Visual Analysis of Movement Behavior using Web Data for Context Enrichment

Size: px
Start display at page:

Download "Visual Analysis of Movement Behavior using Web Data for Context Enrichment"

Transcription

1 Visual Analysis of Movement Behavior using Web Data for Context Enrichment Robert Krueger Dennis Thom Thomas Ertl Institute for Visualization and Interactive Systems (VIS), Universita t Stuttgart, Germany Figure 1: Our system for interactive visual analysis of semantically enriched movement data consists of two components: top) Geographic View visualizing frequent destination clusters, routes, POIs (points of interest) and degree of uncertainty by varying color intensity of the icons; bottom) Temporal View - showing frequent temporal daily patterns. Most frequent patterns are shown on top. Color varies from green (frequent) to white (infrequent). Views are connected by brushing and linking. A BSTRACT With increasing use of GPS devices more and more location-based information is accessible. Thus not only more movements of people are tracked but also POI (point of interest) information becomes available in increasing geo-spatial density. To enable analysis of movement behavior, we present an approach to enrich trajectory data with semantic POI information and show how additional insights can be gained. Using a density-based clustering technique we extract frequent destinations of user movements from a large e-mobility database. We query available context information from Foursquare, a popular location-based social network, to enrich the destinations with semantic background. As GPS measurements can be noisy, often more then one possible destination is found and movement patterns vary over time. Therefore we present highly interactive visualizations that enable an analyst to cope with the inherent geospatial and behavioral uncertainties. Index Terms: Information Storage and Retrieval [H.3.3]: In- {forename.lastname}@vis.uni-stuttgart.de formation Search and Retrieval Information filtering, Query formulation, Selection process; Database Applications [H.2.8]: Spatial databases and GIS ; Information Interfaces and Presentation [H.5.2]: User Interfaces GUI; 1 I NTRODUCTION Nowadays, as data collection becomes easier and cheaper, more and more data is generated. For example, in the traffic domain there is an increasing use of GPS tracking devices to harvest locationbased information. This information is not only used to support drivers but also for analysis purposes, for example in fleet management, engine analysis or to model traffic simulations. Mostly technical details are in scope, but there is more information in the data. To better understand urban dynamics, but also to enable consumer acceptance analysis, there is an increasing interest to look into movement reasons, i.e. why people are moving, where they are going, for what purposes they use a product, and what they are doing at their destinations. However, such information cannot be accessed easily. For example, if a user stops at a specific point in the afternoon, there can be several reasons for this behavior and the recorded GPS positions are often not enough to come up with an explanation. They need to be enriched with further information to support the data analyst in the sensemaking process. As location-

2 enabled Web 2.0 data volumes are growing every day and more and more people are using location-based social media services to give information about POIs, this can be a suitable source for the necessary context data. Foursquare, for example, offers a webservice providing detailed geolocated information of several POIs, called venues. Such as restaurants, bars, business locations, universities, sport parks, and public transport stations. Mainly used for What is here user centered navigation-applications, the services can also be used to enrich geospatial data with context information. In case of movement behavior analysis, one can enrich the movement data with venue information and investigate the destination of movements, i.e. which destinations the users visited and thereby gain more insights of movement reasons. However there are several challenges that hinder a straightforward data enrichment: First, GPS measurements are often not exact, and more inaccuracies can occur while broadcasting and logging this information. Thus data can become noisy. Second, location information is dense in some areas and it is not clear which venues were visited. While mobile phones stick with the user, vehicles have to be parked somewhere and are often not moved all the way to the actual destination. Moreover, there is a high number of venues not recorded by POI services, and thus many map locations cannot be resolved. In this paper we address the challenge by applying a stepwise data processing approach (see Figure 2). First, a clustering algorithm is used to detect ferquently visited destinations and remove noise (a). Second, the frequently visited destinations are enriched with POI information form a webservice (b). Finally an analyst can examine movement behaviors in an interactive visual loop (c) and investigate results (d). Our techniques were developed and evaluated based on an electric mobility dataset generated in a long-term field-study by a large german power company. Electric scooters, particularly suited for urban use, were given to 500 interested users and their usage of the vehicles was recorded in the course of one year using on-board GPS devices. While semantic enrichment of such trajectories with other data recently gained more and more interest and important research has been done in this area, the process and decision of venue assignment and uncertainty handling in space and time has, to the best of our knowledge, not been addressed so far. Therefore, the primary contributions of this paper lie in the semantic enrichment of movements based on Web 2.0 data and the design of highly interactive visualizations that allow an insightful handling of geospatial uncertainties resulting from location inaccuracy as well as temporal uncertainty resulting from changing behavior patterns at the same time. The rest of this paper is structured as follows. In the next section we discuss related work in the field of movement data analysis and semantic enrichment as well as uncertainty visualization. Preprocessing and context enrichment of the data, i.e. steps of clustering and semantic enrichment are discussed in Section 3. In Section 4 we address uncertainty handling and visualization. The interactive analysis is described in Section 5. In Section 6 we demonstrate the applicability of our methods based on two case studies and give a conclusion and outlook in Section 7. 2 RELATED WORK This paper presents an approach to analyze movements using semantic context information, and visualizes uncertainties in geographical and temporal space. Thus, we first discuss related work in the field of visual movement/trajectory analysis in general (section 2.1), before focusing on semantic enrichment. While some works present techniques, applications and results to directly analyze data from location-based web services (2.2), only few use such services to enrich movement data from different sources (2.3). Finally we discuss related work in the more general field of spatial and temporal uncertainty visualization (section 2.4). Figure 2: Analysis Pipeline: a) Extract frequent destinations, b) Data enrichment using a POI Service, c) Interactive Visual Analysis of POI results and movement patterns, d) Interpretation 2.1 Movement Analysis and Semantic Enrichment Movement analysis is a well established research area in the field of Visual Analytics (VA). In recent years many techniques and applications were proposed. Andrienko et al. [6], [4], [3] and Schreck et al. [30] present approaches for automated trajectory aggregation and visualization. Hurter et al. [23] and Krueger et al. [26] investigate exploratory means using focus+context techniques, such as lenses, to visually group and filter trajectories. A visual interactive system for traffic congestion analysis was presented by Wang et al. [35]. While most of these works address macroscopic challenges, some approaches focus on micro pattern analysis. With TripVista a system to analyze trajectory data from crossroads, i.e. turns, speed and temporal behavior was presented by Guo et al. [21]. While the research is mostly about the trajectory data itself (e.g. form, speed, direction), some methods have been proposed to enrich trajectory data to investigate movement reasons. The work of Gonzales et al. [18] suggests techniques to understand basic laws of human motions from individual human mobility patterns. They use a large dataset of mobile phone users, calculate individual travel patterns and aggregate them into a single spatial probability distribution. Andrienko et al. [8] also examine and visualize locations from mobile phones based on call frequencies and guess whether the locations are work or home locations. Alvares et al. [2] propose a generic data preprocessing model for semantic trajectory enrichment by transforming trajectories in stops and movement first, and then add geographic information. They also provide a brief survey on research in the field of semantic trajectory data models. Brakatsoulas et al. [10] formulate a data model to store and query trajectory semantics, like intersections, distances, times, speed, and covered areas, for traffic data. A semantic enrichment for animal movement trajectories was presented by Spaccapietra et al. [32]. They propose modeling approaches that add behavioral semantics (e.g. feeding), but also weather information such as wind, temperature, sky conditions, and area conditions like mountains, water extents, or deserts to the data. 2.2 Social Media Movement Analysis With the increasing use of social media more and more geo-located information is collected and often becomes available via webservices. While there is a lot of research in location based web data, this discussion especially focuses on POI and trip analysis. Mashups [33] is a system to visualize lifelogs from Foursquare, Facebook, etc. on an interactive map. Similarly, Preotiuc-Pietro et al. [29] analyze the behavior of Foursquare users. In contrast to our work Foursquare venues are not used as context to enrich other movement data, but to directly analyze the behavior of Foursquare

3 users. They also use aggregation techniques to investigate periodic behavior of the users. Kling and Pozdnoukhov [24] investigate movement patterns from Foursquare and geo-located Twitter messages to extract semantics behind activities. Movements are visualized in several ways using streamgraphs, geographic heatmaps and word clouds. Fujsaka et al. [17] analyze movements of users based on geolocated Tweets and apply cluster analysis on frequent POIs. In contrast to our work, they focus on the Web 2.0 users and do not use the data as context to enrich trajectory data. 2.3 Semantic Enrichment Using Web Data There is already some related work where trajectories are semantically enriched with POI information. Parent et al. [28] present concepts for (1) the construction of trajectories from plain movement data, (2) semantic enrichment to enable desired interpretations of movements, and (3) the application of data mining techniques to extract knowledge, i.e. behavior patterns of movements. Andrienko et al. [7] focus on privacy preserving movement analysis. They enrich trajectory data from mobile phones with information from a POI database and present results in a semantic space by visualizing flows between activity types. Semantic annotations based on a POI service was discussed by Guc et al. [20]. They propose a conceptual annotation model for semantically enriched trajectories and present components and architecture of a graphical user interface. However, trajectory enrichment is only done based on user knowledge and none of the discussed services are applied. In another work, Andrienko et al. [5] suggest to interpret meanings of personal POIs based on cyclic temporal patterns of visit times. Krueger et al. [25] show work in progress, where movement trajectories are enriched with context from location based Twitter messages. They are visualized using geospatial tag clouds. In context of Nokia s Mobile Data Challenge Hoeferlin et al. [22] present a visual system for explorative analysis of user behavior from a mobile phone dataset. Although, they mention to query POI services for context information it is not the main focus of their work and it is not addressed how they process and integrate those results in an interactive analysis loop and in visualization. 2.4 Uncertainty Visualization in Space and Time Uncertainty visualization is a well establish research field. A good overview of existing techniques has been given by Griethe et al. [19]. Furthermore, MacEachren et al. [27] review progress in geographic uncertainty research towards visual tools and show frameworks suggesting different visual methods for uncertainty representation dependent on data properties such as type and quality. Correa et al. [14] present a framework to support uncertainty analysis in the VA process and presents some information visualizations for uncertain data. A visualization of uncertainty in lattices to support decision making is presented by Collins et al. [12]. They use a hybrid layout with varying color, size and transparency. To the best of our knowledge, we are the first who enrich large movement data with Web 2.0 content in an automated way and allow to exploratively reason the data using interactive visualizations. Furthermore, none of the mentioned research in section 2.3 uses visualization techniques to show the uncertainties, resulting from context enrichment. 3 PREPROCESSING: SEMANTIC ENRICHMENT Preprocessing itself is an important step in information visualization, but it often receives little attention. In recorded movement data, however, inaccuracies in space and time, ambiguities in movement and inconsistent measurements are part of the core problem, especially when data volumes are huge, and make preprocessing necessary. We use a dataset generated by 527 GPS-enabled vehicles as a basis for the development of our methods and evaluation. These vehicles were tracked every 30 seconds for two years, leading to an overall amount of about trips and more than measurements. By visualizing the trips destinations we found that most destinations build dense clusters over time surrounded with some noise, as shown in Figure 3. Noise can occur due to several reasons: First, there are destinations only visited a few times and thus no clear cluster appears. Second, there are measuring inaccuracies due to GPS signal interference, deficient receiver performance, data logging, or broadcasting issues. We choose to accept the information loss of some less dense destinations and focus on frequent visited ones. Later approaches could of course also focus on poorly attended areas to indicate outliers. 3.1 Determine Destination Clusters Figure 3: Popular destinations build dense clusters (red) surrounded with noise (white). Map details are blured due to privacy purposes. In Figure 3 it can be seen that the density in destination clusters is mostly similar and there are clear borders between clusters and whitespaces. For automated destination cluster detection we apply DBScan [16], a density-based clustering technique with a geospatial distance measure. Advantages using DBScan compared to partition based approaches like k-means are that (1) we do not need a-priori knowledge about the number of clusters, (2) DB-Scan implicitly filters out noise and (3) is able to detect clusters of varying shape (longish, etc.). As a result we detect clusters containing of the total trip endpoints. Thus, about two of three endpoints have been often visited by a single user or have been visted at least once by serveral users. We assume that this will frequently be places of employment, shopping centers, homes, and service stations. 3.2 Context Enrichment and Destination Assignment To generate up-to-date context information about the frequently visited areas (clusters) we employ Foursquare as an additional data source. Foursquare is a location-based social networking service with dense information about POIs in city areas. Venue information from Foursquare is often quite recent as the data is regularly updated by users, which can always add, delete, or update POIs (in Foursquare called venues). Furthermore, it delivers additional information such as the number of users that visited a venue and the number of individual Checkins from users, which can help to determine the recent prominence of a location. While there are other providers offering similar services like Google, Microsoft and Facebook, the Foursquare API has less query limitations and has the second highest POI density. The lower query restrictions were also an important factor in our decision, as it allows to compute the data enrichment online (at runtime) enabling more interactive requests. Table 1 compares three major POI APIs in overall number of POIs worldwide, allowed data requests per minute and hour, caching policies, and quality of the data structures.

4 Table 1: Approximated POI API limits and facts (taken from Constine [13], the Facebook API documentation [1] and public forum discussions). Criteria Google Facebook Foursquare Number of POIs missing Requests/min Requests/day Caching forbidden lacking info 30 days Structure fair fair good When querying POI services there are several parameters to influence search results. Most important is the geographic query radius for a given latitute/longitude pair. For the e-mobility dataset we determined a 50 meters radius around the cluster centers as a good range, as users often park their vehicles not directly at a venue but in short walking distance. For some clusters we did not get a venue. These clusters are either private destinations (e.g. home, residence of a friend), charging stations, or anything unknown. Figure 4 shows frequent destination clusters with different assigned destination types (venue, private, charging). Charging Station: As our test dataset is about electric vehicles, we add another data source containing exact locations for power charging stations in that area. Each cluster with a distance 50 meters to such a station is handled as charging cluster. Home Location: Every scooter is assigned a home location based on the destination that is most frequently visited by the user. However, some scooters were used for special testing purposes and are more frequently parked at various special locations (e.g. repair and assembly shop). Therefore we apply cleaning measures to remove vehicles that are missing a clearly identifiable home location. Figure 5: Found venues per cluster, using a 50 meters query radius. For 607 of the clusters venues where found. teractive system. As we want to characterize movements we have to deal with the spatial and temporal dimensions at the same time. We therefore developed two components and integrated them in our movement analysis system: a geographic map view and a temporal view. The geographic map view shows routes and destinations for identified POIs, while the temporal view shows frequent temporal behavior patterns. In both components (Figure 1) the focus is on dealing with uncertainties. 4.1 Geographic Component The map component shows frequent destinations (clusters) and corresponding context information and lets the analyst explore and investigate routes and destinations in an interactive fashion (see Figure 1). The context information, i.e. venues from Foursquare are categorized into three hierarchical levels. Each level aggregates its child categories to more comprehensive ones. The highest level contains 9 overview categories, as shown in Figure 6. Every venue Figure 4: Found Clusters: venues nearby (red), home (green), chargings (yellow). Map details are removed due to privacy purposes. Overall we were able to enrich 75% of the data (clusters). From 1215 clusters we identified 266 as home locations, 27 as electric scooter charging stations, and 607 clusters could be connected to venues via Foursquare. For some of these latter clusters we got up to 42 possible venues in urban areas and often just one identifiable venue for clusters in more remote areas. Figure 5 illustrates the distribution of venues per cluster. 4 INTERACTIVE VISUAL MOVEMENT REASONING After preprocessing (i.e. clustering and context enrichment, described in section 3) the analyst can start reasoning using an in- Figure 6: Foursquare category tree: each level summarizes its child nodes in parent categories. Leaf nodes are high detailed venue types, such as Post Office. belongs to one of these categories. For example Ann s Blue Jeans Shop belongs to the subsubcategory Boutique, subcategory Clothing Store and category Shops and Services. Simply visualizing the most detailed venue information (420 different venue types) however hampers fluent analysis and renders fast situation assessment impossible. Thus, we first show the main categories (Figure 6, left), which can then be interactively explored by changing the current category level to see more or less details. Doing so we follow the interactive visualization mantra Overview first, then zoom and filter, details on demand [31]. However, only in few cases the category

5 of a cluster can be clearly determined. Often more than one venue is found (see Figure 5), and often these venues belong to different categories, thus it is uncertain which venue was actually visited. This uncertainty has to be addressed and visualized. For example if 30 of 40 found venues for a cluster are restaurants, there is a good chance that the area is a food district and we can thus be quite confident that the scooter user went there to have a meal. According to the Foursquare category structure we define three degrees of detail for each category: main category (level 1), subcategory (level 2) and subsubcategory (level 3), while level 0 holds all found venues. For each level l we can now determine the most likely category by comparing the number of venues of direct siblings (i.e. categories at the same level having the same parent category). Depending on the current detail level the certainty c, 0 c 1 (c = 1 is certain) of the determined category cat at level l is calculated as a ratio of the number of venues found in this category and the overall number of venues found for the destination cluster. c catl = count(cat l )/count(cat 0 ) (1) Thus, with higher level of detail categories usually get more and more uncertain, since only a low amount of found venues belong to exactly that category, e.g Australian Restaurant, while with lower detail it becomes more certain that most venues belong to a main category Food. Using the proposed calculations we can now visualize venue categories for each cluster at the current level by showing the cluster points and the corresponding category icons. Uncertainty is visualized using a gradient color mapping for the icons and ranges from blue (certainty 1) to white (certainty 0). To investigate categories and venues the analyst can place the mouse over an icon and drill down to lower category levels to get more specific information (Figure 7). Figure 8: Map snippet showing clusters and found venues at category level 1. Top: Ungrouped icons may give better hints but clutter; Bottom: Aggregated icons avoid clutter. Map details are removed in the image due to privacy. destination clusters visited with these scooters are then visualized. In addition the scooter tracks are shown on the map with a certain degree of opacity, making frequent routes appear with higher intensity. To filter for individual scooters, the analyst can select their Figure 7: Found venues for a cluster in the inner city area, where venues are dense. From left to right: most likely main category, subcategory, subsubcategory with venue name. With higher detail certainty is decreasing Reducing Visual Clutter While in outer city areas and suburbs distances between clusters are usually quite high for our e-mobility dataset, in the downtown area clusters can be much closer to each other, causing heavy visual clutter when an icon is to be displayed for each of them. There have been several grouping techniques proposed to allow aggregation of icons, which are now quite common in interactive map visualizations. In our case, we group nearby icons together to form representative icons showing the number of aggregated POIs. To determine the certainty of the group icons we accumulate the certainties of the included venues and compute their average. As this is done based on pixel distance it can easily be recomputed for all geographic zoom levels of the interactive map. Figure 8 shows the map with detail view (top) and aggregated view (bottom) Filtering To investigate movements the analyst can either apply an area based filtering or filter the tracks on a per scooter basis. Filtering by area can be done using a lens magnifier tool and move it over the map [26]. The lens filter includes all movements of scooters having their home location in the lens area (see Figure 9). Only the Figure 9: Area based analysis. Map snippet shows the scooter home location and frequent destinations like shops and offices. The area was selected using a lens magnifier. Map is hidden, due to privacy. home location by clicking on the corresponding Residence icon. In urban areas often more than one venue is found for a given destination cluster and our automated approach discussed before is used to select the most likely venue. However, sometimes the analyst may disagree with the systems choice and it should thus be possible to integrate her knowledge. For example, the analyst may recognize that a cluster has been formed on the large parking area of a warehouse store, which also happens to house a small hot-dog stand. Because of the spatial proximity on the parking area, the system may have falsely suggested that most scooter users within the cluster went there for the hot-dog stand. Therefore the analyst should be allowed to change the icon and category on the current category level. Figure 10 shows Shops and Services is preselected, since the certainty is the highest for all categories in that level. However with expert knowledge about a certain area she is aware that this destination was most likely a restaurant, and switch to that category. The same procedure can be done on each category level. For example

6 she stays with the Shop and Services category but on lower level selects the hairdresser category instead of the preselected bank. Figure 10: The analyst can integrate expert knowledge and change preselected categories on each level of detail. 4.2 Temporal Component To allow detail analysis of frequent usage behavior of the scooters we also provide a temporal analysis component (lower half of Figure 1). If one or more scooters are selected on the map, for example to investigate the typical usage behavior within a specific suburb, frequently visited POIs are highlighted within a linear temporal view. The view represents cyclically repeating venue occurrences by aggregating them based on a definable timeframe, e.g. daily, weekly or restricted to weekends. For example, if the analyst is interested in daily behavior, the visualization subdivides the day in 24 sections. For each hour of the day, the most frequently visited POI is displayed on top, while less frequent ones are stacked below in descending order of frequency. This way the analyst can always identify the most probable current location for a given hour of the day up top. This aggregation is restricted to POIs that have been visited for a duration not exceeding 12 hours in order to remove periods where it was not used for a long time. In addition to the vertical order we also map the frequency to a color scheme ranging from green (most frequent) to white (least frequent). Depending on the current level of detail (see section 4.1) the venues are aggregated according to respective categories. Thus the temporal visualization unfolds on higher detail levels (i.e. shows more distinct destinations) and folds up when lower detail levels like the primary categories are selected. The map view and temporal view are interactively linked. By investigating the destinations in the temporal view details of the aggregated venues (type, number, frequency) are shown and all corresponding destinations are consistently highlighted in the map, revealing their geographic position. Changes in the map view (e.g. selecting a more appropriate POI for a given destination) are also directly reflected in the temporal view. The chosen visualization for temporal analysis has many similarities to common Streamgraph and Sankey diagram variations (see [34], [15], [9], [11]) where the categories are shown as continuous stacked sorted ribbons. Although it might be easier to follow the temporal changes for a given category in these visualizations, we experienced that they also lead to heavy clutter due to frequently crossing ribbons in the case of several changes in the category order. Because it is more important for our domain to understand the probability order of POIs for a given time of the day, instead of the temporal changes that a POI experiences, we chose to cut the ribbon for each segment and indicate the category by a corresponding symbol at this point. 5 CASE STUDY To demonstrate the applicability of our methods we apply our approach to our evaluation dataset. In our first case study we want to investigate consumer acceptance in distinct city suburbs. The second study centers on individual product usage. 5.1 Motivation To produce the dataset, the 500 study participants signed a contract that the complete usage of the bikes would be tracked and analyzed. While also several vehicle specific statistics were recorded and evaluated, many questions remained unanswered, such as whether e- mobility was able to find it s way into every day life. Thus in the first case study we want to investigate for what purposes the products (scooters) were applied for and when, e.g. for daily means of transportation or for leisure. Before analyzing such higher aggregated data, it might also be important to focus on the usage of individual products, where venues can be validated on a more detailed and precise level. This is shown in the second study. Although users agreed to the analysis of their movements, this study still holds a high risk of violating their privacy. Therefore we anonymized the following results as good as possible. 5.2 Residence dependent consumer acceptance To compare consumer acceptance in different city areas we use the lens magnifier and place it over a specific suburb. The lens is adjustable in size so that it can be enlarged or shrinked until it covers the area to be analyzed. In this case we select an area which is about 20 kilometers outside of the city. We want to investigate the range of uses that people have for their scooters. Besides some technical statistics for the current selection (e.g. average distance, number of trips and battery drainage) we also see the tracks of the users and the corresponding venues. In Figure 11 it can be seen that the main destinations are characterized in the categories Professionals, Shops and Services and University. This reveals that these scooters were Figure 11: Map snippet. All users living in the filtered suburb are selected. Routes show movements. Icons show frequently visited POIs. probably employed for daily transportation means, like work, shopping and education. The thickness and intensity of the highlighted routes also gives us a general idea how frequent certain POIs were visited. The certainty coloring of the icons reveals that it is quite sure the selected users employ their scooters to head for the cities university, while office places are more infrequently visited in their movements. For more details, symbols can be examined by mousehover to show a tooltip with more information and the level of detail can be changed, as described in section To investigate the discovered locations we can now check other venues that were found for a destination by clicking on the icon. This way we discover that at the offices there is also a restaurant, but it is less frequently visited by Foursquare users and thus also less likely to be a frequent destination for the scooter users. To investigate the periodic frequencies and temporal behavior we continue our analysis using the temporal view. It shows that for the selected scooters the home location covers most of the time, indicating that the scooters are not

7 Figure 12: Temporal component showing frequent temporal patterns per hour on a daily basis. It can be seen that at night times, the selected scooters are usually at home, while they are used to get to work and education at daytimes. In the early evening shopping becomes more frequent. used every day (see Figure 12). Earlier investigations showed that the vehicles were rarely used during the winter months. One can also see that the university and business areas were primarily visited during the office hours. In the afternoon other venues are also visited frequently, such as the category Outdoors. The users obviously use their scooter to go shopping and for dinner. By mousehovering over the venue bars in the temporal view a tooltip provides more detail about the venues, e.g. how often they were visited at this time of the day during the year. We can now switch to a different temporal template, showing only days during weekends. This reveals that shopping is a frequent activity for which the scooters were used during leisure times. 5.3 Individual scooter usage In our second use case we focus on individual product usage. We start by selecting a scooter home location, which will highlight corresponding routes, visited POIs and typical behavior, as can be seen in Figure 13. The map shows that the scooter is likely applied to visit the shopping sites Penny Markt and ALDI, two supermarkets and also Bauhaus, a hardware store. While users can charge their scooters at home, this scooter was also charged at a public charging station. Further analysis suggests that the scooter was employed to drive to the hospital, for dining (McDonalds) and shopping at Stylecode. However there are more sites found in the vicinity of these POIs, all showing a lower certainty. There is also a business location found in the north west of the map, but with expert knowledge we are able to identify this location as repair and assembly shop for the scooters. Examining the place Grillwagen (hot-dog stand) reveals that there is also a different venue available here, another hardware store (Figure 13). We thus zoom in the map and indeed see a large OBI (large german hardware store chain) sign on the map as well. Using our background knowledge we can infer that the scooter user went to this store rather than to the Grillwagen (Figure 13). Final investigations in the temporal behavior reveals that the user is not a frequent driver, that the scooter was seldom used for daily transportation means, but that it was used now and then, primarily for shopping. 6 CONCLUSION In this work we presented a stepwise approach to generate insights from recorded vehicle movements. We first processed a massive movement dataset and extracted frequent and dense destination areas. We then enriched the data with context information using Foursquare, a popular POI webservice. We employed an interactive visual analysis loop to gain insights from the enriched data and cope with different types of uncertainty in the data. Our interactive visual analysis system consists of two linked components for (1) geographic information and (2) periodic temporal information of frequent movement behavior. We not only focused on uncertainty visualization in the geographical and temporal domain but also presented ideas how to interactively deal with the uncertainties and Figure 13: Map snippet. An individual scooter is selected. Routes show movements. Icons visualize frequent visited POIs. Map details are removed for privacy protection. integrate them in fluent analysis. Future work will further improve the automated venue decisions based on temporal probabilities [29]. Considering time, one could also make better guesses which venues were visited and whether a user visited one or multiples based on the number and popularity of available POIs. Moreover, we want to take more context data, such as weather information and traffic models into account. Also enrichment results should be validated. Therefore a dataset with a-prori known destinations will be generated and used for evaluation. We also plan to apply our approach to movement data from other domains, such as floating car data or animal movements. ACKNOWLEDGEMENTS This work has been conducted in the context of the project VASA (13N11254) funded by the German Federal Ministry of Education and Research (BMBF). Furthermore, it was supported by the graduate program Digital Media of the Universities of Stuttgart and Tübingen, and the Stuttgart Media University (HdM). The electric scooter data was kindly provided by EnBW Energie Baden- Württemberg AG. We would like to thank them for their collaboration. REFERENCES [1] Graph api rate limiting [2] L. O. Alvares, V. Bogorny, B. Kuijpers, J. A. F. de Macedo, B. Moelans, and A. Vaisman. A model for enriching trajectories with seman-

8 tic geographical information. In Proceedings of the 15th annual ACM international symposium on Advances in geographic information systems, page 22. ACM, [3] G. Andrienko and N. Andrienko. Spatio-temporal aggregation for visual analysis of movements. In Visual Analytics Science and Technology, VAST 08. IEEE Symposium on, pages IEEE, [4] G. Andrienko and N. Andrienko. Sammon s projection for clustering complex geographical objects. et al.: GIScience, [5] G. Andrienko and N. Andrienko. Visual Analytics of Movement. Springer-Verlag Berlin An, [6] G. Andrienko, N. Andrienko, S. Rinzivillo, M. Nanni, D. Pedreschi, and F. Giannotti. Interactive visual clustering of large collections of trajectories. In Visual Analytics Science and Technology, VAST IEEE Symposium on, pages IEEE, [7] N. Andrienko, G. Andrienko, and G. Fuchs. Towards privacypreserving semantic mobility analysis. [8] N. Andrienko, G. Andrienko, H. Stange, T. Liebig, and D. Hecker. Visual analytics for understanding spatial situations from episodic movement data. KI-Künstliche Intelligenz, 26(3): , [9] Z. Beane. Movie box office charts, oct [10] S. Brakatsoulas, D. Pfoser, and N. Tryfona. Modeling, storing and mining moving object databases. In Database Engineering and Applications Symposium, IDEAS 04. Proceedings. International, pages IEEE, [11] L. Byron and M. Wattenberg. Stacked graphs geometry & aesthetics. Visualization and Computer Graphics, IEEE Transactions on, 14(6): , [12] C. Collins, S. Carpendale, and G. Penn. Visualization of uncertainty in lattices to support decision-making. In Proceedings of the 9th Joint Eurographics/IEEE VGTC conference on Visualization, pages Eurographics Association, [13] J. Constine. What are the pros and cons of each places api? [14] C. D. Correa, Y.-H. Chan, and K.-L. Ma. A framework for uncertaintyaware visual analytics. In Visual Analytics Science and Technology, VAST IEEE Symposium on, pages IEEE, [15] W. Cui, S. Liu, L. Tan, C. Shi, Y. Song, Z. Gao, H. Qu, and X. Tong. Textflow: Towards better understanding of evolving topics in text. Visualization and Computer Graphics, IEEE Transactions on, 17(12): , [16] M. Ester, H.-P. Kriegel, J. Sander, and X. Xu. A density-based algorithm for discovering clusters in large spatial databases with noise. In KDD, volume 96, pages , [17] T. Fujisaka, R. Lee, and K. Sumiya. Discovery of user behavior patterns from geo-tagged micro-blogs. In Proceedings of the 4th International Conference on Uniquitous Information Management and Communication, page 36. ACM, [18] M. C. Gonzalez, C. A. Hidalgo, and A.-L. Barabasi. Understanding individual human mobility patterns. Nature, 453(7196): , [19] H. Griethe and H. Schumann. The visualization of uncertain data: Methods and problems. In SimVis, pages , [20] B. Guc, M. May, Y. Saygin, and C. Körner. Semantic annotation of gps trajectories. In 11th AGILE International Conference on Geographic Information Science, [21] H. Guo, Z. Wang, B. Yu, H. Zhao, and X. Yuan. Tripvista: Triple perspective visual trajectory analytics and its application on microscopic traffic data at a road intersection. In Pacific Visualization Symposium (PacificVis), 2011 IEEE, pages IEEE, [22] B. Hoeferlin, M. Hoeferlin, and J. Raeuchle. Visual analytics of mobile data. In Nokia Mobile Data Challenge 2012 Workshop (2012), [23] C. Hurter, B. Tissoires, and S. Conversy. FromDaDy: Spreading aircraft trajectories across views to support iterative queries. IEEE Transactions on Visualization and Computer Graphics (Proceedings InfoVis), 15(6): , [24] F. Kling and A. Pozdnoukhov. When a city tells a story: urban topic analysis. In Proceedings of the 20th International Conference on Advances in Geographic Information Systems, pages ACM, [25] R. Krüger, S. Lohmann, D. Thom, H. Bosch, and T. Ertl. Using social media content in the visual analysis of movement data. In 2nd Workshop on Interactive Visual Text Analytics, [26] R. Krüger, D. Thom, M. Wörner, H. Bosch, and T. Ertl. Trajectorylenses a set-based filtering and exploration technique for long-term trajectory data. In Computer Graphics Forum, volume 32, pages Wiley Online Library, [27] A. M. MacEachren, A. Robinson, S. Hopper, S. Gardner, R. Murray, M. Gahegan, and E. Hetzler. Visualizing geospatial information uncertainty: What we know and what we need to know. Cartography and Geographic Information Science, 32(3): , [28] C. Parent, S. Spaccapietra, C. Renso, G. Andrienko, N. Andrienko, V. Bogorny, M. L. Damiani, A. Gkoulalas-Divanis, J. Macedo, N. Pelekis, et al. Semantic trajectories modeling and analysis. ACM Computing Surveys (CSUR), 45(4):42, [29] D. Preotiuc-Pietro and T. Cohn. Mining user behaviours: A study of check-in patterns in location based social networks. [30] T. Schreck, J. Bernard, T. Von Landesberger, and J. Kohlhammer. Visual cluster analysis of trajectory data with interactive kohonen maps. Information Visualization, 8(1):14 29, [31] B. Shneiderman. The eyes have it: A task by data type taxonomy for information visualizations. In Visual Languages, Proceedings., IEEE Symposium on, pages IEEE, [32] S. Spaccapietra, C. Parent, M. L. Damiani, J. A. de Macedo, F. Porto, and C. Vangenot. A conceptual view on trajectories. Data & knowledge engineering, 65(1): , [33] K. Takahashi, A. Shimojo, S. Matsumoto, and M. Nakamura. Mashmap: Application framework for map-based visualization of lifelog with location. In Information and Telecommunication Technologies (APSITT), th Asia-Pacific Symposium on, pages 1 6. IEEE, [34] E. R. Tufte and P. Graves-Morris. The visual display of quantitative information, volume 2. Graphics press Cheshire, CT, [35] Z. Wang, M. Lu, X. Yuan, J. Zhang, and H. van de Wetering. Visual traffic jam analysis based on trajectory data. IEEE, 2013.

Constructing Semantic Interpretation of Routine and Anomalous Mobility Behaviors from Big Data

Constructing Semantic Interpretation of Routine and Anomalous Mobility Behaviors from Big Data Constructing Semantic Interpretation of Routine and Anomalous Mobility Behaviors from Big Data Georg Fuchs 1,3, Hendrik Stange 1, Dirk Hecker 1, Natalia Andrienko 1,2,3, Gennady Andrienko 1,2,3 1 Fraunhofer

More information

Big Data in Pictures: Data Visualization

Big Data in Pictures: Data Visualization Big Data in Pictures: Data Visualization Huamin Qu Hong Kong University of Science and Technology What is data visualization? Data visualization is the creation and study of the visual representation of

More information

The STC for Event Analysis: Scalability Issues

The STC for Event Analysis: Scalability Issues The STC for Event Analysis: Scalability Issues Georg Fuchs Gennady Andrienko http://geoanalytics.net Events Something [significant] happened somewhere, sometime Analysis goal and domain dependent, e.g.

More information

MOBILITY DATA MODELING AND REPRESENTATION

MOBILITY DATA MODELING AND REPRESENTATION PART I MOBILITY DATA MODELING AND REPRESENTATION 1 Trajectories and Their Representations Stefano Spaccapietra, Christine Parent, and Laura Spinsanti 1.1 Introduction For a long time, applications have

More information

Extend Table Lens for High-Dimensional Data Visualization and Classification Mining

Extend Table Lens for High-Dimensional Data Visualization and Classification Mining Extend Table Lens for High-Dimensional Data Visualization and Classification Mining CPSC 533c, Information Visualization Course Project, Term 2 2003 Fengdong Du fdu@cs.ubc.ca University of British Columbia

More information

CONTENTS. List of Contributors Preface Acknowledgments. mobility data modeling and representation

CONTENTS. List of Contributors Preface Acknowledgments. mobility data modeling and representation CONTENTS List of Contributors Preface Acknowledgments page xi xv xix part i mobility data modeling and representation 1 Trajectories and Their Representations 3 S. Spaccapietra, C. Parent, L. Spinsanti

More information

How To Create An Analysis Tool For A Micro Grid

How To Create An Analysis Tool For A Micro Grid International Workshop on Visual Analytics (2012) K. Matkovic and G. Santucci (Editors) AMPLIO VQA A Web Based Visual Query Analysis System for Micro Grid Energy Mix Planning A. Stoffel 1 and L. Zhang

More information

GEO-VISUALIZATION SUPPORT FOR MULTIDIMENSIONAL CLUSTERING

GEO-VISUALIZATION SUPPORT FOR MULTIDIMENSIONAL CLUSTERING Geoinformatics 2004 Proc. 12th Int. Conf. on Geoinformatics Geospatial Information Research: Bridging the Pacific and Atlantic University of Gävle, Sweden, 7-9 June 2004 GEO-VISUALIZATION SUPPORT FOR MULTIDIMENSIONAL

More information

Visualizing Repertory Grid Data for Formative Assessment

Visualizing Repertory Grid Data for Formative Assessment Visualizing Repertory Grid Data for Formative Assessment Kostas Pantazos 1, Ravi Vatrapu 1, 2 and Abid Hussain 1 1 Computational Social Science Laboratory (CSSL) Department of IT Management, Copenhagen

More information

Information Visualization WS 2013/14 11 Visual Analytics

Information Visualization WS 2013/14 11 Visual Analytics 1 11.1 Definitions and Motivation Lot of research and papers in this emerging field: Visual Analytics: Scope and Challenges of Keim et al. Illuminating the path of Thomas and Cook 2 11.1 Definitions and

More information

Gain insight, agility and advantage by analyzing change across time and space.

Gain insight, agility and advantage by analyzing change across time and space. White paper Location Intelligence Gain insight, agility and advantage by analyzing change across time and space. Spatio-temporal information analysis is a Big Data challenge. The visualization and decision

More information

Visualization methods for patent data

Visualization methods for patent data Visualization methods for patent data Treparel 2013 Dr. Anton Heijs (CTO & Founder) Delft, The Netherlands Introduction Treparel can provide advanced visualizations for patent data. This document describes

More information

Tracking System for GPS Devices and Mining of Spatial Data

Tracking System for GPS Devices and Mining of Spatial Data Tracking System for GPS Devices and Mining of Spatial Data AIDA ALISPAHIC, DZENANA DONKO Department for Computer Science and Informatics Faculty of Electrical Engineering, University of Sarajevo Zmaja

More information

Spatio-Temporal Networks:

Spatio-Temporal Networks: Spatio-Temporal Networks: Analyzing Change Across Time and Place WHITE PAPER By: Jeremy Peters, Principal Consultant, Digital Commerce Professional Services, Pitney Bowes ABSTRACT ORGANIZATIONS ARE GENERATING

More information

Context-aware taxi demand hotspots prediction

Context-aware taxi demand hotspots prediction Int. J. Business Intelligence and Data Mining, Vol. 5, No. 1, 2010 3 Context-aware taxi demand hotspots prediction Han-wen Chang, Yu-chin Tai and Jane Yung-jen Hsu* Department of Computer Science and Information

More information

International Journal of Computer Science Trends and Technology (IJCST) Volume 2 Issue 3, May-Jun 2014

International Journal of Computer Science Trends and Technology (IJCST) Volume 2 Issue 3, May-Jun 2014 RESEARCH ARTICLE OPEN ACCESS A Survey of Data Mining: Concepts with Applications and its Future Scope Dr. Zubair Khan 1, Ashish Kumar 2, Sunny Kumar 3 M.Tech Research Scholar 2. Department of Computer

More information

Introduction to Data Mining

Introduction to Data Mining Introduction to Data Mining 1 Why Data Mining? Explosive Growth of Data Data collection and data availability Automated data collection tools, Internet, smartphones, Major sources of abundant data Business:

More information

PhoCA: An extensible service-oriented tool for Photo Clustering Analysis

PhoCA: An extensible service-oriented tool for Photo Clustering Analysis paper:5 PhoCA: An extensible service-oriented tool for Photo Clustering Analysis Yuri A. Lacerda 1,2, Johny M. da Silva 2, Leandro B. Marinho 1, Cláudio de S. Baptista 1 1 Laboratório de Sistemas de Informação

More information

Real-time Processing and Visualization of Massive Air-Traffic Data in Digital Landscapes

Real-time Processing and Visualization of Massive Air-Traffic Data in Digital Landscapes Real-time Processing and Visualization of Massive Air-Traffic Data in Digital Landscapes Digital Landscape Architecture 2015, Dessau Stefan Buschmann, Matthias Trapp, and Jürgen Döllner Hasso-Plattner-Institut,

More information

Big Data: Rethinking Text Visualization

Big Data: Rethinking Text Visualization Big Data: Rethinking Text Visualization Dr. Anton Heijs anton.heijs@treparel.com Treparel April 8, 2013 Abstract In this white paper we discuss text visualization approaches and how these are important

More information

Linköpings Universitet - ITN TNM033 2011-11-30 DBSCAN. A Density-Based Spatial Clustering of Application with Noise

Linköpings Universitet - ITN TNM033 2011-11-30 DBSCAN. A Density-Based Spatial Clustering of Application with Noise DBSCAN A Density-Based Spatial Clustering of Application with Noise Henrik Bäcklund (henba892), Anders Hedblom (andh893), Niklas Neijman (nikne866) 1 1. Introduction Today data is received automatically

More information

Topic Maps Visualization

Topic Maps Visualization Topic Maps Visualization Bénédicte Le Grand, Laboratoire d'informatique de Paris 6 Introduction Topic maps provide a bridge between the domains of knowledge representation and information management. Topics

More information

131-1. Adding New Level in KDD to Make the Web Usage Mining More Efficient. Abstract. 1. Introduction [1]. 1/10

131-1. Adding New Level in KDD to Make the Web Usage Mining More Efficient. Abstract. 1. Introduction [1]. 1/10 1/10 131-1 Adding New Level in KDD to Make the Web Usage Mining More Efficient Mohammad Ala a AL_Hamami PHD Student, Lecturer m_ah_1@yahoocom Soukaena Hassan Hashem PHD Student, Lecturer soukaena_hassan@yahoocom

More information

Introduction. A. Bellaachia Page: 1

Introduction. A. Bellaachia Page: 1 Introduction 1. Objectives... 3 2. What is Data Mining?... 4 3. Knowledge Discovery Process... 5 4. KD Process Example... 7 5. Typical Data Mining Architecture... 8 6. Database vs. Data Mining... 9 7.

More information

Location-Based Social Networks: Users

Location-Based Social Networks: Users Chapter 8 Location-Based Social Networks: Users Yu Zheng Abstract In this chapter, we introduce and define the meaning of location-based social network (LBSN) and discuss the research philosophy behind

More information

APPLICATION OF DATA MINING TECHNIQUES FOR BUILDING SIMULATION PERFORMANCE PREDICTION ANALYSIS. email paul@esru.strath.ac.uk

APPLICATION OF DATA MINING TECHNIQUES FOR BUILDING SIMULATION PERFORMANCE PREDICTION ANALYSIS. email paul@esru.strath.ac.uk Eighth International IBPSA Conference Eindhoven, Netherlands August -4, 2003 APPLICATION OF DATA MINING TECHNIQUES FOR BUILDING SIMULATION PERFORMANCE PREDICTION Christoph Morbitzer, Paul Strachan 2 and

More information

Information Visualisation and Visual Analytics for Governance and Policy Modelling

Information Visualisation and Visual Analytics for Governance and Policy Modelling Information Visualisation and Visual Analytics for Governance and Policy Modelling Jörn Kohlhammer 1, Tobias Ruppert 1, James Davey 1, Florian Mansmann 2, Daniel Keim 2 1 Fraunhofer IGD, Fraunhoferstr.

More information

SPATIAL DATA CLASSIFICATION AND DATA MINING

SPATIAL DATA CLASSIFICATION AND DATA MINING , pp.-40-44. Available online at http://www. bioinfo. in/contents. php?id=42 SPATIAL DATA CLASSIFICATION AND DATA MINING RATHI J.B. * AND PATIL A.D. Department of Computer Science & Engineering, Jawaharlal

More information

Clustering. Adrian Groza. Department of Computer Science Technical University of Cluj-Napoca

Clustering. Adrian Groza. Department of Computer Science Technical University of Cluj-Napoca Clustering Adrian Groza Department of Computer Science Technical University of Cluj-Napoca Outline 1 Cluster Analysis What is Datamining? Cluster Analysis 2 K-means 3 Hierarchical Clustering What is Datamining?

More information

Big Data Analytics in Mobile Environments

Big Data Analytics in Mobile Environments 1 Big Data Analytics in Mobile Environments 熊 辉 教 授 罗 格 斯 - 新 泽 西 州 立 大 学 2012-10-2 Rutgers, the State University of New Jersey Why big data: historical view? Productivity versus Complexity (interrelatedness,

More information

An Analysis on Density Based Clustering of Multi Dimensional Spatial Data

An Analysis on Density Based Clustering of Multi Dimensional Spatial Data An Analysis on Density Based Clustering of Multi Dimensional Spatial Data K. Mumtaz 1 Assistant Professor, Department of MCA Vivekanandha Institute of Information and Management Studies, Tiruchengode,

More information

VISUALIZING HIERARCHICAL DATA. Graham Wills SPSS Inc., http://willsfamily.org/gwills

VISUALIZING HIERARCHICAL DATA. Graham Wills SPSS Inc., http://willsfamily.org/gwills VISUALIZING HIERARCHICAL DATA Graham Wills SPSS Inc., http://willsfamily.org/gwills SYNONYMS Hierarchical Graph Layout, Visualizing Trees, Tree Drawing, Information Visualization on Hierarchies; Hierarchical

More information

Gerard Mc Nulty Systems Optimisation Ltd gmcnulty@iol.ie/0876697867 BA.,B.A.I.,C.Eng.,F.I.E.I

Gerard Mc Nulty Systems Optimisation Ltd gmcnulty@iol.ie/0876697867 BA.,B.A.I.,C.Eng.,F.I.E.I Gerard Mc Nulty Systems Optimisation Ltd gmcnulty@iol.ie/0876697867 BA.,B.A.I.,C.Eng.,F.I.E.I Data is Important because it: Helps in Corporate Aims Basis of Business Decisions Engineering Decisions Energy

More information

Discovering Trajectory Outliers between Regions of Interest

Discovering Trajectory Outliers between Regions of Interest Discovering Trajectory Outliers between Regions of Interest Vitor Cunha Fontes 1, Lucas Andre de Alencar 1, Chiara Renso 2, Vania Bogorny 1 1 Dep. de Informática e Estatística Universidade Federal de Santa

More information

Similarity Search and Mining in Uncertain Spatial and Spatio Temporal Databases. Andreas Züfle

Similarity Search and Mining in Uncertain Spatial and Spatio Temporal Databases. Andreas Züfle Similarity Search and Mining in Uncertain Spatial and Spatio Temporal Databases Andreas Züfle Geo Spatial Data Huge flood of geo spatial data Modern technology New user mentality Great research potential

More information

Selection of Optimal Discount of Retail Assortments with Data Mining Approach

Selection of Optimal Discount of Retail Assortments with Data Mining Approach Available online at www.interscience.in Selection of Optimal Discount of Retail Assortments with Data Mining Approach Padmalatha Eddla, Ravinder Reddy, Mamatha Computer Science Department,CBIT, Gandipet,Hyderabad,A.P,India.

More information

Scalable Cluster Analysis of Spatial Events

Scalable Cluster Analysis of Spatial Events International Workshop on Visual Analytics (2012) K. Matkovic and G. Santucci (Editors) Scalable Cluster Analysis of Spatial Events I. Peca 1, G. Fuchs 1, K. Vrotsou 1,2, N. Andrienko 1 & G. Andrienko

More information

Information Management course

Information Management course Università degli Studi di Milano Master Degree in Computer Science Information Management course Teacher: Alberto Ceselli Lecture 01 : 06/10/2015 Practical informations: Teacher: Alberto Ceselli (alberto.ceselli@unimi.it)

More information

Customer Analytics. Turn Big Data into Big Value

Customer Analytics. Turn Big Data into Big Value Turn Big Data into Big Value All Your Data Integrated in Just One Place BIRT Analytics lets you capture the value of Big Data that speeds right by most enterprises. It analyzes massive volumes of data

More information

CHAPTER-24 Mining Spatial Databases

CHAPTER-24 Mining Spatial Databases CHAPTER-24 Mining Spatial Databases 24.1 Introduction 24.2 Spatial Data Cube Construction and Spatial OLAP 24.3 Spatial Association Analysis 24.4 Spatial Clustering Methods 24.5 Spatial Classification

More information

Clustering & Visualization

Clustering & Visualization Chapter 5 Clustering & Visualization Clustering in high-dimensional databases is an important problem and there are a number of different clustering paradigms which are applicable to high-dimensional data.

More information

Visualization of Streaming Data: Observing Change and Context in Information Visualization Techniques

Visualization of Streaming Data: Observing Change and Context in Information Visualization Techniques Visualization of Streaming Data: Observing Change and Context in Information Visualization Techniques Miloš Krstajić, Daniel A. Keim University of Konstanz Konstanz, Germany {milos.krstajic,daniel.keim}@uni-konstanz.de

More information

TOWARDS SIMPLE, EASY TO UNDERSTAND, AN INTERACTIVE DECISION TREE ALGORITHM

TOWARDS SIMPLE, EASY TO UNDERSTAND, AN INTERACTIVE DECISION TREE ALGORITHM TOWARDS SIMPLE, EASY TO UNDERSTAND, AN INTERACTIVE DECISION TREE ALGORITHM Thanh-Nghi Do College of Information Technology, Cantho University 1 Ly Tu Trong Street, Ninh Kieu District Cantho City, Vietnam

More information

Identifying User Behavior in domainspecific

Identifying User Behavior in domainspecific Identifying User Behavior in domainspecific Repositories Wilko VAN HOEK a,1, Wei SHEN a and Philipp MAYR a a GESIS Leibniz Institute for the Social Sciences, Germany Abstract. This paper presents an analysis

More information

Big Data Collection and Utilization for Operational Support of Smarter Social Infrastructure

Big Data Collection and Utilization for Operational Support of Smarter Social Infrastructure Hitachi Review Vol. 63 (2014), No. 1 18 Big Data Collection and Utilization for Operational Support of Smarter Social Infrastructure Kazuaki Iwamura Hideki Tonooka Yoshihiro Mizuno Yuichi Mashita OVERVIEW:

More information

20 A Visualization Framework For Discovering Prepaid Mobile Subscriber Usage Patterns

20 A Visualization Framework For Discovering Prepaid Mobile Subscriber Usage Patterns 20 A Visualization Framework For Discovering Prepaid Mobile Subscriber Usage Patterns John Aogon and Patrick J. Ogao Telecommunications operators in developing countries are faced with a problem of knowing

More information

Robust Outlier Detection Technique in Data Mining: A Univariate Approach

Robust Outlier Detection Technique in Data Mining: A Univariate Approach Robust Outlier Detection Technique in Data Mining: A Univariate Approach Singh Vijendra and Pathak Shivani Faculty of Engineering and Technology Mody Institute of Technology and Science Lakshmangarh, Sikar,

More information

What is Visualization? Information Visualization An Overview. Information Visualization. Definitions

What is Visualization? Information Visualization An Overview. Information Visualization. Definitions What is Visualization? Information Visualization An Overview Jonathan I. Maletic, Ph.D. Computer Science Kent State University Visualize/Visualization: To form a mental image or vision of [some

More information

COMP 150-04 Visualization. Lecture 11 Interacting with Visualizations

COMP 150-04 Visualization. Lecture 11 Interacting with Visualizations COMP 150-04 Visualization Lecture 11 Interacting with Visualizations Assignment 5: Maps Due Wednesday, March 17th Design a thematic map visualization Option 1: Choropleth Map Implementation in Processing

More information

Data Visualization Handbook

Data Visualization Handbook SAP Lumira Data Visualization Handbook www.saplumira.com 1 Table of Content 3 Introduction 20 Ranking 4 Know Your Purpose 23 Part-to-Whole 5 Know Your Data 25 Distribution 9 Crafting Your Message 29 Correlation

More information

Visually driven analysis of movement data by progressive clustering

Visually driven analysis of movement data by progressive clustering Visually driven analysis of movement data by progressive clustering Salvatore Rinzivillo* Dino Pedreschi Mirco Nanni Fosca Giannotti KDD Lab, University of Pisa {rinziv,pedre}@di.unipi.it Natalia Andrienko

More information

Understanding Web personalization with Web Usage Mining and its Application: Recommender System

Understanding Web personalization with Web Usage Mining and its Application: Recommender System Understanding Web personalization with Web Usage Mining and its Application: Recommender System Manoj Swami 1, Prof. Manasi Kulkarni 2 1 M.Tech (Computer-NIMS), VJTI, Mumbai. 2 Department of Computer Technology,

More information

Spatiotemporal Clustering of Twitter Feeds for Activity Summarization

Spatiotemporal Clustering of Twitter Feeds for Activity Summarization Spatiotemporal Clustering of Twitter Feeds for Activity Summarization N. Wayant 1, A. Crooks 2, A. Stefanidis 3, A. Croitoru 3, J. Radzikowski 3, J. Stahl 2, J. Shine 2 1 US Army ERDC Topographic Engineering

More information

TEXT-FILLED STACKED AREA GRAPHS Martin Kraus

TEXT-FILLED STACKED AREA GRAPHS Martin Kraus Martin Kraus Text can add a significant amount of detail and value to an information visualization. In particular, it can integrate more of the data that a visualization is based on, and it can also integrate

More information

Knowledge Discovery from patents using KMX Text Analytics

Knowledge Discovery from patents using KMX Text Analytics Knowledge Discovery from patents using KMX Text Analytics Dr. Anton Heijs anton.heijs@treparel.com Treparel Abstract In this white paper we discuss how the KMX technology of Treparel can help searchers

More information

How To Make Sense Of Data With Altilia

How To Make Sense Of Data With Altilia HOW TO MAKE SENSE OF BIG DATA TO BETTER DRIVE BUSINESS PROCESSES, IMPROVE DECISION-MAKING, AND SUCCESSFULLY COMPETE IN TODAY S MARKETS. ALTILIA turns Big Data into Smart Data and enables businesses to

More information

MOBILE SOCIAL NETWORKS FOR LIVE MEETINGS

MOBILE SOCIAL NETWORKS FOR LIVE MEETINGS Computer Science 13 (4) 2012 http://dx.doi.org/10.7494/csci.2012.13.4.87 Michał Wrzeszcz Jacek Kitowski MOBILE SOCIAL NETWORKS FOR LIVE MEETINGS Abstract In this article, we present an idea of combining

More information

Exploring Big Data in Social Networks

Exploring Big Data in Social Networks Exploring Big Data in Social Networks virgilio@dcc.ufmg.br (meira@dcc.ufmg.br) INWEB National Science and Technology Institute for Web Federal University of Minas Gerais - UFMG May 2013 Some thoughts about

More information

3D Interactive Information Visualization: Guidelines from experience and analysis of applications

3D Interactive Information Visualization: Guidelines from experience and analysis of applications 3D Interactive Information Visualization: Guidelines from experience and analysis of applications Richard Brath Visible Decisions Inc., 200 Front St. W. #2203, Toronto, Canada, rbrath@vdi.com 1. EXPERT

More information

Integration of GPS Traces with Road Map

Integration of GPS Traces with Road Map Integration of GPS Traces with Road Map Lijuan Zhang Institute of Cartography and Geoinformatics Leibniz University of Hannover Hannover, Germany +49 511.762-19437 Lijuan.Zhang@ikg.uni-hannover.de Frank

More information

Security visualisation

Security visualisation Security visualisation This thesis provides a guideline of how to generate a visual representation of a given dataset and use visualisation in the evaluation of known security vulnerabilities by Marco

More information

Visualizing e-government Portal and Its Performance in WEBVS

Visualizing e-government Portal and Its Performance in WEBVS Visualizing e-government Portal and Its Performance in WEBVS Ho Si Meng, Simon Fong Department of Computer and Information Science University of Macau, Macau SAR ccfong@umac.mo Abstract An e-government

More information

Visual Exploration of Sparse Traffic Trajectory Data

Visual Exploration of Sparse Traffic Trajectory Data Visual Exploration of Sparse Traffic Trajectory Data Zuchao Wang, Tangzhi Ye, Min Lu, Xiaoru Yuan, Member, IEEE, Huamin Qu, Member, IEEE, Jacky Yuan and Qianliang Wu Fig. 1. Our sparse traffic trajectory

More information

Information Visualization and Visual Analytics

Information Visualization and Visual Analytics Information Visualization and Visual Analytics Pekka Wartiainen University of Jyväskylä pekka.wartiainen@jyu.fi 23.4.2014 Outline Objectives Introduction Visual Analytics Information Visualization Our

More information

Course 803401 DSS. Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization

Course 803401 DSS. Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization Oman College of Management and Technology Course 803401 DSS Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization CS/MIS Department Information Sharing

More information

Comparative Analysis of EM Clustering Algorithm and Density Based Clustering Algorithm Using WEKA tool.

Comparative Analysis of EM Clustering Algorithm and Density Based Clustering Algorithm Using WEKA tool. International Journal of Engineering Research and Development e-issn: 2278-067X, p-issn: 2278-800X, www.ijerd.com Volume 9, Issue 8 (January 2014), PP. 19-24 Comparative Analysis of EM Clustering Algorithm

More information

Behavior Analysis in Crowded Environments. XiaogangWang Department of Electronic Engineering The Chinese University of Hong Kong June 25, 2011

Behavior Analysis in Crowded Environments. XiaogangWang Department of Electronic Engineering The Chinese University of Hong Kong June 25, 2011 Behavior Analysis in Crowded Environments XiaogangWang Department of Electronic Engineering The Chinese University of Hong Kong June 25, 2011 Behavior Analysis in Sparse Scenes Zelnik-Manor & Irani CVPR

More information

SuperViz: An Interactive Visualization of Super-Peer P2P Network

SuperViz: An Interactive Visualization of Super-Peer P2P Network SuperViz: An Interactive Visualization of Super-Peer P2P Network Anthony (Peiqun) Yu pqyu@cs.ubc.ca Abstract: The Efficient Clustered Super-Peer P2P network is a novel P2P architecture, which overcomes

More information

The Scientific Data Mining Process

The Scientific Data Mining Process Chapter 4 The Scientific Data Mining Process When I use a word, Humpty Dumpty said, in rather a scornful tone, it means just what I choose it to mean neither more nor less. Lewis Carroll [87, p. 214] In

More information

Tableau Your Data! Wiley. with Tableau Software. the InterWorks Bl Team. Fast and Easy Visual Analysis. Daniel G. Murray and

Tableau Your Data! Wiley. with Tableau Software. the InterWorks Bl Team. Fast and Easy Visual Analysis. Daniel G. Murray and Tableau Your Data! Fast and Easy Visual Analysis with Tableau Software Daniel G. Murray and the InterWorks Bl Team Wiley Contents Foreword xix Introduction xxi Part I Desktop 1 1 Creating Visual Analytics

More information

Continuous Fastest Path Planning in Road Networks by Mining Real-Time Traffic Event Information

Continuous Fastest Path Planning in Road Networks by Mining Real-Time Traffic Event Information Continuous Fastest Path Planning in Road Networks by Mining Real-Time Traffic Event Information Eric Hsueh-Chan Lu Chi-Wei Huang Vincent S. Tseng Institute of Computer Science and Information Engineering

More information

Information & Data Visualization. Yasufumi TAKAMA Tokyo Metropolitan University, JAPAN ytakama@sd.tmu.ac.jp

Information & Data Visualization. Yasufumi TAKAMA Tokyo Metropolitan University, JAPAN ytakama@sd.tmu.ac.jp Information & Data Visualization Yasufumi TAKAMA Tokyo Metropolitan University, JAPAN ytakama@sd.tmu.ac.jp 1 Introduction Contents Self introduction & Research purpose Social Data Analysis Related Works

More information

A Tool For Active FLEET Management and Analysis of Activities of a Snow PlowING and a Road Salting Fleet

A Tool For Active FLEET Management and Analysis of Activities of a Snow PlowING and a Road Salting Fleet A Tool For Active FLEET Management and Analysis Activities a Snow PlowING and a Road Salting Fleet Rok Strašek, Tina Vukasović Abstract - Current economic crisis combined with increasing fuel costs rises

More information

Text Mining - Scope and Applications

Text Mining - Scope and Applications Journal of Computer Science and Applications. ISSN 2231-1270 Volume 5, Number 2 (2013), pp. 51-55 International Research Publication House http://www.irphouse.com Text Mining - Scope and Applications Miss

More information

Data Visualisation and Its Application in Official Statistics. Olivia Or Census and Statistics Department, Hong Kong, China ooyor@censtatd.gov.

Data Visualisation and Its Application in Official Statistics. Olivia Or Census and Statistics Department, Hong Kong, China ooyor@censtatd.gov. Data Visualisation and Its Application in Official Statistics Olivia Or Census and Statistics Department, Hong Kong, China ooyor@censtatd.gov.hk Abstract Data visualisation has been a growing topic of

More information

RiskLab. Discussion: Peter Sarlin. Visual Network Analysis in the Regulation of Financial Systemic Risk

RiskLab. Discussion: Peter Sarlin. Visual Network Analysis in the Regulation of Financial Systemic Risk Discussion: Visual Network Analysis in the Regulation of Financial Systemic Risk The Application of Visual Analytics to Financial Stability Monitoring Peter Sarlin Goethe University Frankfurt and European

More information

TOP-DOWN DATA ANALYSIS WITH TREEMAPS

TOP-DOWN DATA ANALYSIS WITH TREEMAPS TOP-DOWN DATA ANALYSIS WITH TREEMAPS Martijn Tennekes, Edwin de Jonge Statistics Netherlands (CBS), P.0.Box 4481, 6401 CZ Heerlen, The Netherlands m.tennekes@cbs.nl, e.dejonge@cbs.nl Keywords: Abstract:

More information

The 4 Pillars of Technosoft s Big Data Practice

The 4 Pillars of Technosoft s Big Data Practice beyond possible Big Use End-user applications Big Analytics Visualisation tools Big Analytical tools Big management systems The 4 Pillars of Technosoft s Big Practice Overview Businesses have long managed

More information

EasyTracker: An Android application for capturing mobility behavior

EasyTracker: An Android application for capturing mobility behavior EasyTracker: An Android application for capturing mobility behavior Alexandros Doulamis Dept. of Informatics University of Piraeus Piraeus, Greece alex.doulamis@gmail.com Nikos Pelekis Dept. of Statistics

More information

Data Mining for Fun and Profit

Data Mining for Fun and Profit Data Mining for Fun and Profit Data mining is the extraction of implicit, previously unknown, and potentially useful information from data. - Ian H. Witten, Data Mining: Practical Machine Learning Tools

More information

Public Transportation BigData Clustering

Public Transportation BigData Clustering Public Transportation BigData Clustering Preliminary Communication Tomislav Galba J.J. Strossmayer University of Osijek Faculty of Electrical Engineering Cara Hadriana 10b, 31000 Osijek, Croatia tomislav.galba@etfos.hr

More information

Building Data Cubes and Mining Them. Jelena Jovanovic Email: jeljov@fon.bg.ac.yu

Building Data Cubes and Mining Them. Jelena Jovanovic Email: jeljov@fon.bg.ac.yu Building Data Cubes and Mining Them Jelena Jovanovic Email: jeljov@fon.bg.ac.yu KDD Process KDD is an overall process of discovering useful knowledge from data. Data mining is a particular step in the

More information

Dr. Shih-Lung Shaw s Research on Space-Time GIS, Human Dynamics and Big Data

Dr. Shih-Lung Shaw s Research on Space-Time GIS, Human Dynamics and Big Data Dr. Shih-Lung Shaw s Research on Space-Time GIS, Human Dynamics and Big Data for Geography Department s Faculty Research Highlight October 12, 2014 Shih-Lung Shaw, Ph.D. Alvin and Sally Beaman Professor

More information

<no narration for this slide>

<no narration for this slide> 1 2 The standard narration text is : After completing this lesson, you will be able to: < > SAP Visual Intelligence is our latest innovation

More information

The Role of Visualization in Effective Data Cleaning

The Role of Visualization in Effective Data Cleaning The Role of Visualization in Effective Data Cleaning Yu Qian Dept. of Computer Science The University of Texas at Dallas Richardson, TX 75083-0688, USA qianyu@student.utdallas.edu Kang Zhang Dept. of Computer

More information

Data Mining Cluster Analysis: Basic Concepts and Algorithms. Lecture Notes for Chapter 8. Introduction to Data Mining

Data Mining Cluster Analysis: Basic Concepts and Algorithms. Lecture Notes for Chapter 8. Introduction to Data Mining Data Mining Cluster Analysis: Basic Concepts and Algorithms Lecture Notes for Chapter 8 by Tan, Steinbach, Kumar 1 What is Cluster Analysis? Finding groups of objects such that the objects in a group will

More information

INFORMATION VISUALIZATION TECHNIQUES USAGE MODEL

INFORMATION VISUALIZATION TECHNIQUES USAGE MODEL INFORMATION VISUALIZATION TECHNIQUES USAGE MODEL Akanmu Semiu A. 1 and Zulikha Jamaludin 2 1 Universiti Utara Malaysia, Malaysia, ayobami.sm@gmail.com 2 Universiti Utara Malaysia, Malaysia, zulie@uum.edu.my

More information

Mapping Linear Networks Based on Cellular Phone Tracking

Mapping Linear Networks Based on Cellular Phone Tracking Ronen RYBOWSKI, Aaron BELLER and Yerach DOYTSHER, Israel Key words: Cellular Phones, Cellular Network, Linear Networks, Mapping. ABSTRACT The paper investigates the ability of accurately mapping linear

More information

Unresolved issues with the course, grades, or instructor, should be taken to the point of contact.

Unresolved issues with the course, grades, or instructor, should be taken to the point of contact. Graphics and Data Visualization CS1501 Fall 2013 Syllabus Course Description With the advent of powerful data-mining technologies, engineers in all disciplines are increasingly expected to be conscious

More information

Chapter 5 Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization

Chapter 5 Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization Turban, Aronson, and Liang Decision Support Systems and Intelligent Systems, Seventh Edition Chapter 5 Business Intelligence: Data Warehousing, Data Acquisition, Data Mining, Business Analytics, and Visualization

More information

An example. Visualization? An example. Scientific Visualization. This talk. Information Visualization & Visual Analytics. 30 items, 30 x 3 values

An example. Visualization? An example. Scientific Visualization. This talk. Information Visualization & Visual Analytics. 30 items, 30 x 3 values Information Visualization & Visual Analytics Jack van Wijk Technische Universiteit Eindhoven An example y 30 items, 30 x 3 values I-science for Astronomy, October 13-17, 2008 Lorentz center, Leiden x An

More information

Search Result Optimization using Annotators

Search Result Optimization using Annotators Search Result Optimization using Annotators Vishal A. Kamble 1, Amit B. Chougule 2 1 Department of Computer Science and Engineering, D Y Patil College of engineering, Kolhapur, Maharashtra, India 2 Professor,

More information

Big Data Mining Services and Knowledge Discovery Applications on Clouds

Big Data Mining Services and Knowledge Discovery Applications on Clouds Big Data Mining Services and Knowledge Discovery Applications on Clouds Domenico Talia DIMES, Università della Calabria & DtoK Lab Italy talia@dimes.unical.it Data Availability or Data Deluge? Some decades

More information

COMPACT GUIDE. Camera-Integrated Motion Analysis

COMPACT GUIDE. Camera-Integrated Motion Analysis EN 05/13 COMPACT GUIDE Camera-Integrated Motion Analysis Detect the movement of people and objects Filter according to directions of movement Fast, simple configuration Reliable results, even in the event

More information

International Journal of Advancements in Research & Technology, Volume 3, Issue 5, May-2014 18 ISSN 2278-7763. BIG DATA: A New Technology

International Journal of Advancements in Research & Technology, Volume 3, Issue 5, May-2014 18 ISSN 2278-7763. BIG DATA: A New Technology International Journal of Advancements in Research & Technology, Volume 3, Issue 5, May-2014 18 BIG DATA: A New Technology Farah DeebaHasan Student, M.Tech.(IT) Anshul Kumar Sharma Student, M.Tech.(IT)

More information

Database Marketing, Business Intelligence and Knowledge Discovery

Database Marketing, Business Intelligence and Knowledge Discovery Database Marketing, Business Intelligence and Knowledge Discovery Note: Using material from Tan / Steinbach / Kumar (2005) Introduction to Data Mining,, Addison Wesley; and Cios / Pedrycz / Swiniarski

More information

T he complete guide to SaaS metrics

T he complete guide to SaaS metrics T he complete guide to SaaS metrics What are the must have metrics each SaaS company should measure? And how to calculate them? World s Simplest Analytics Tool INDEX Introduction 4-5 Acquisition Dashboard

More information

Traffic Prediction and Analysis using a Big Data and Visualisation Approach

Traffic Prediction and Analysis using a Big Data and Visualisation Approach Traffic Prediction and Analysis using a Big Data and Visualisation Approach Declan McHugh 1 1 Department of Computer Science, Institute of Technology Blanchardstown March 10, 2015 Summary This abstract

More information

Density Map Visualization for Overlapping Bicycle Trajectories

Density Map Visualization for Overlapping Bicycle Trajectories , pp.327-332 http://dx.doi.org/10.14257/ijca.2014.7.3.31 Density Map Visualization for Overlapping Bicycle Trajectories Dongwook Lee 1, Jinsul Kim 2 and Minsoo Hahn 1 1 Digital Media Lab., Korea Advanced

More information

siemens.com/mobility Traffic data analysis in Sitraffic Scala/Concert The expert system for visualization, quality management and statistics

siemens.com/mobility Traffic data analysis in Sitraffic Scala/Concert The expert system for visualization, quality management and statistics siemens.com/mobility Traffic data analysis in Sitraffic Scala/Concert The expert system for visualization, quality management and statistics 2 Traffic data analysis produces transparent intersections The

More information