David Coufal 2 Institute of Computer Science Academy of Sciences of the Czech Republic Pod Vodarenskou vezi 2, Prague 8, Czech Republic
|
|
- Estella Holmes
- 8 years ago
- Views:
Transcription
1 V.012 SHORT TERM PREDICTION OF HIGHWAY TRAVEL TIME USING DATA MINING AND NEURO-FUZZY METHODS 1 David Coufal 2 Institute of Computer Science Academy of Sciences of the Czech Republic Pod Vodarenskou vezi 2, Prague 8, Czech Republic Esko Turunen 34 Tampere University of Technology P.O. Box 692, Tampere, Finland Abstract We show that prediction of travel time on a 28-km long highway section based on on-line travel time measurements with video is practicable by data mining and neuro-fuzzy methods. We introduce two new prediction models. The first one is a result of GUHA style data mining analysis and Total Fuzzy Similarity method, and the second one is a hierarchical model based on neuro-fuzzy modelling. Comparing results with the existing Traficon model, both new models improve the travel time prediction. The results obtained by the new methods are comparable to MLP neural network model, too. Key words: fuzzy logic, neural networks, data mining. 1. Introduction The aim of this study is to show that short-term travel time prediction presented in [3] can be carried out by data mining and neuro-fuzzy methods, too, and that results are comparable. Research [3] was carried out on main road 4 between points A (Lahti) and D (Heinola) in Southern Finland. According to [3], the average daily summertime traffic on this 28-kilometer section is about vehicles per day, in particular, the traffic volumes are high during summer weekends. The study section AD is divided into three sub-sections AB, BC and CD with camera stations approximately equally distributed over link AD length and equipped with an automatic travel time monitoring system. The system is based on an artificial vision and neural network application, which automatically reads license plates. Moreover, there is an inductive loop detector on station C gathering information on traffic volumes and point speeds. A variable message sign (VMS) at point A gives upper and lower bounds of a estimations about the travel to the point D. In an unpublished preliminary study of the problem done by Laura Lanne, the estimation categories are below 25 min, min, min, min and above 50 min. In [3], travel time from point A to point D is regarded as congested if it is above 25 min. In [3], travel time prediction p is regarded as acceptable if the real travel time lies in the interval [0.9*p, 1.1*p], i.e., it is used ±10% marginal error based correctness of results. 1 This research is part of research project COST Action 274 [TARSKI] 2 Supported by grant OC of Ministry of Education, Youth and Sports of the Czech Republic 3 Supported by grant of Finnish Academy 4 Correspondence via esko.turunen@tut.fi
2 The erroneous predictions can be divided into two categories: the travel time prediction can be too pessimistic (too long travel times, i.e. road users arriving to their destinations earlier than expected) or too optimistic (too short travel times, i.e. road users arriving to their destinations later than expected). The too optimistic travel time prediction is worse for the road user than the too pessimistic travel time prediction. In this study we used two data sets we received in autumn 2001 and in summer 2002 from Helsinki University of Technology, Transportation Engineering. The data was given in form of the following table. Input Output traffic flow at C avg. speed at C avg. tt AB avg. tt BC avg. tt CD avg. tt real travel time AD [min] [vehicles/5min] [km/hour] [min] [min] [min] AD [min] the value to be predicted avg. tt = average travel time There are 4541 rows (cases) in the first set and 9333 rows (cases) it the second set. Notice that average travel time AD is not the sum AB+BC+CD but it is the average travel time of the vehicles that passed point D during the last 5 minutes spent in the whole section. In other words, the sum AB+BC+CD contains travel time information of at least three different vehicles while AD can be based on travel time information of one single car. The structures of the data sets, in a sense that all predictions are correct, are presented in following Table1 and Table 2. Predicted travel time Travel time <= 25 (25, 30] (30, 40] (40, 50] > 50 row sum <= (25, 30] (30, 40] [40-50) > column sum Table 1. The structure of the first data set. From Table 1, we see that in first data set there are 4426 (97%) cases of non-congested conditions and 115 (3%) cases of congested conditions. In congested conditions we have distribution into particular categories given a s 16.5% - (25, 30]; 37.4% - (30,40]; 34% - (40,50]; 12.1% - >50.
3 Predicted travel time Travel time <= 25 (25, 40] (30, 40] (40, 50] > 50 row sum <= (25, 30] (30, 40] (40, 50) > Column sum Table 2. The structure of the second data set. According to Table 2, in the second data set there are 8693 (93%) non-congested cases and 640 (7%) congested cases. With respect to congested cases, the distribution into particular categories is 26.9% - (25,30]; 60% - (30,40], 13.1% - (40,50], 0.02% - >50. In the present application, called Traficon model, the estimated value of the travel time, which is around 20 minutes in normal non-congested circumstances, is based on the last measurements on each sub-link AB, BC and CD. The estimation is normally updated every 5 minutes. However, a problem is that the estimation is always more or less outdated. Based on a similar kind of data and presented in [3], the Traficon model estimates the congested travel times right within 10% error marginal in 32,9%.. Therefore, one of the objectives in [3] was to develop a model that would improve the travel time prediction, in particular, in the congested cases. This was done by MLP-neural networks and, indeed, the prediction results improved. In the new model the travel times are predicted right in 98,4% of all cases, while 0,5% of all cases are predicted too high and 1,1% of all cases are predicted too low. Among the congested cases the improvement is significant: 66,5% of the congested cases are predicted right, while 5,3% are predicted too high and 28,2% too low. In this paper we show that comparable results can be attained by other methods, too. 2. A Travel time prediction by GUHA analysis and Total Fuzzy Similarity method The aim in the first new approach is to develop a prediction method that would increase correctness of forecasts and, simultaneously, be as simple and reasonable as possible. The underlying consumption is that input values similar enough imply relatively similar output values. Indeed, the idea in Total Fuzzy Similarity method introduced in [6] is to present an experts knowledge by a finite set of fuzzy IF - THEN inference rules and, in each particular case, to search for the most similar IF - part in the role base and, finally, to fire the corresponding THEN - part. In the travel time prediction problem there was, however, no expert to tell which kinds of traffic situations lead congested travel times and which ones to normal travel times. Thus, we used GUHA method [2] to create the rule base. GUHA - General Unary Hypotheses Automation - is a logically and statistically founded data mining method to find all interesting facts following from a data matrix containing n columns and m rows. In the core of GUHA analysis there are frequency tables of a form
4 succedent non(succedent) antecedent a B non(antecedent) c D In practice, the GUHA analysis was carried out by means of a computer program LISp-Miner [4]. A results of such an analysis is presented in Appendix 1. To have as up-to-date information as possible, we added two new columns to the data, the sums AB+BC and AB+BC+CD. The main difference in values AB+BC and AB+BC+CD with respect to values AC and AD, respectively, is that e.g. travel times of vehicles calculated into the value AD have already left the whole section, while most vehicles calculated into the values AB+BC+CD have not jet reached destination D and, if they are in a congested situation, then this congestion has an effect on travel time of the vehicles just starting from point A. We used the first data set of 4541 cases as a model teaching material. First we searched for simple conditions that should divide the data into two subsets X = {predicted travel time and real travel time are both above 30 min} and Y = {predicted travel time and real travel time are both below 30 min} such that union of X and Y would be as large as possible. It turned out that such a condition really exists. Indeed, if the data is divided by a condition (A) AD is at least 23 min and AB+BC is at least 17,5 min and AB is at least 5,58 min then only 9 cases out of 4541 are not in X nor in Y. Moreover, in the set X, predicted travel time is a more or less linear function of the value AB. However, as we wanted to keep things simple, we did not include the function into the model but let the output be a prediction class. Second, we searched for simple conditions that would divide the set Y into suitable subsets Y1 = {predicted travel time and real travel time are both above 25 min} and Y2 = {predicted travel time and real travel time are both below 25 min}. We found such a condition. Indeed, by a condition (B) (AB+BC+CD is at least 21,25 min or AD is at least 35 min) and CD is at least 6,3 min the set Y can be divided such that 30 cases out of 4409 are not in Y1 nor in Y2. Third, we searched by GUHA method for simple conditions that would divide the set Y2 into suitable subsets Y21 = {predicted travel time and real travel time are both above 20 min} and Y22 = {predicted travel time and real travel time are both below 20 min}. It turned out that such rules exist, however, they are no more simple nor reasonable. This is, after all, not restrictive as in the VMS the cases 'Travel time < 20 min' and 'Travel time min' are not distinguished from each other. Another result of GUHA analysis is that the input values 'Traffic flow at point C' and 'Average Speed at point C' do not seem have any significant importance with respect to the travel time prediction.
5 Based on these kinds of GUHA analyses, the rule base of a Total Fuzzy Similarity - inference system is the following TÄHÄN ASTI KOPOITU IFSA 2003 ARTIKKELIINl IF AD >= 23 AND AB+BC >= 17.5 AND 23 <= AB THEN PREDICTION is ' > 50 min' IF AD >= 23 AND AB+BC >= 17.5 AND 12 =< AB < 23 THEN PREDICTION is '(40,50] min' IF AD >= 23 AND AB+BC >= 17.5 AND 5.58 <= AB < 12 THEN PREDICTION is '(30,40] min' IF AB+BC+CD >= AND CD>=6.3 THEN PREDICTION is '(25,30] min' IF AD >= 35 AND CD>=6.3 THEN PREDICTION is '(25,30] min' ELSE PREDICTION is '(20,25] min' The corresponding fuzzy sets reduce to crisp ones. If the output would not be unique i.e. there are several IF - parts possessing the maximal total similarity degree, then - corresponding to the 'pessimistic prediction principle' - the prediction should be the longest one. In Table 3, we see the rate of correct predictions among the teaching data set obtained by the above IF-THEN inference system. Indeed, 98,9% among all data fall into the proper prediction class and 80,9% among the congested cases. Predictions travel time < 25] (25, 30] (30, 40] (40, 50] >50 row sum <= (25, 30] (30, 40] (40, 50] > Column sum Table 3. Rate of correct predictions among teaching data obtained by Total Fuzzy Similarity Since the rules of the above IF-THEN inference system are tuned with respect to the first data set, a real and relevant test is to see correctness of predictions among the second data set of 9333 cases. Real Prediction in test data Travel time < 25] [25-30) [30-40) [40-50) > 50 Row sum < 25] [25-30) [30-40) [40-50) > column sum Table 4. Rate of correct predictions among test data obtained by Total Fuzzy Similarity According to the results in Table 4, in all test data, prediction is right in 96,5 % of all 9333 cases, too low in 1,6 % of these cases and too high in 1,9 % of cases. In congested situations, and there are 640 such cases, the figures are 60,5 %, 23,1 % and 16,4 %, respectively. Note that due to fact
6 that output of the Total Similarity model is a class and not an exact value, thus, we cannot use here ±10% marginal error, but we form the table without this marginal test so our results are stronger than if we would use ±10% marginal error. For example, a real travel time 50,3 min is predicted to fall into the class (40,50]. This prediction is regarded as an error. It is worth emphasising that the present Total Fuzzy Similarity model is extremely simple, indeed, it contains only six rules. Adding more rules and counting the output e.g. by a linear function would improve correctness of predictions. This we have, however, not done as our main purpose is to show that hidden in the data, there is a reasonable structure that can be found by GUHA data mining method and then implemented by an IF-THEN rule base. 3. Hierarchical neuro-fuzzy model of traffic data The above prediction model based on GUHA analysis and Total Fuzzy Similarity method gives better results than the one introduced in [3]. However, the Total Fuzzy Similarity model does not predict exact travel times but only prediction classes. This disadvantage could be overcome by adding linear functions to count the exact output, however, the model would no more be simple. From this reason and to evaluate other approach as well, we created a hierarchical neuro-fuzzy model of the data. is AD<=20 YES neuro-fuzzy model for non-congested conditions NO neuro-fuzzy model for congested conditions Figure 1. Hierarchical neuro-fuzzy model. The hierarchical structure of the model is presented in Figure1. Actually, it is very simple. On the first level, data are divided into two classes according to travel time AD. Consequently, data in each class are treated by special neuro-fuzzy model. More precisely, if travel time AD is less or equal to 20 min the case is treated by a neuro-fuzzy model regarding non-congested conditions and, if travel time AD is greater than 20 min, the case is treated by neuro-fuzzy model regarding congested conditions. In the following section we aim to describe the architecture of employed neuro-fuzzy model Neuro-fuzzy model o ( x ) = n j = 1 The model is actually a Wang fuzzy inference system (FIS) designed in neural network fashion to be capable to learn its parameters [7]. Wang FIS computes according to formula c j A j ( x )
7 (1) Note that, to avoid division by zero, the denominator is omitted from a more general equation in [7]. In formula (1), the Aj(x) are multidimensional Gaussians, given by A j (x) n = exp i= 1 (x i a 2b ji 2 ji ) 2 (2) where x=(x 1,, x n ) is a n-dimensional input, a j,=( a j1,, a jn ), b j,=( b j1,, b jn ), b ji 0 are parameters. The Gaussians are combined by linear combination, formula (1), employing coefficients c j. From fuzzy inference system's point of view A j (x) are the antecedents of the system's particular rules and c j are the centroids of their succedents fuzzy sets. Having the architecture of fuzzy model given, its parameters are centers of Gaussians a j, their widths b j and centroids c j. To be able to learn the parameters of the model, we interpret it in a form of neural network of architecture given in Figure 2. h 1 x 1 u 1 w 1 x i u i h 2 w 2 o o x w m1 x n h m1 w m u n h m Figure 2. Neural network representation of Wang FIS. Neurons of input layers are the transmitting ones, i.e., u i (x)= x i. The hidden neurons compute according to (2), i.e., h j (x)= A j (x) and output neuron computes linear combination of hj via weights. Weights corresponds to parameters of linear combination, i.e., w j =c j. Apparently, computation of whole network is given just by formula (1). That is, we have neural network s representation of Wang FIS, i.e., Wang neuro-fuzzy system. To let the network to learn its parameters, we have to apply on it an appropriate structure and parameters learning algorithm. In this study, we used for structure learning incremental learning algorithm [1]. For parameter learning we used Levenberg-Marquard method [5] with respect to error function given by E = i ( t o( )) 2 i x i (3)
8 In (3), it is considered that pairs (x i, t i,) are elements of training set. In the next section we present results of four experiments regarding both traffic data sets. Because traffic flow and average speed variables were observed to do not have an effect on the real travel time, input to the neuro-fuzzy model was only four-dimensional, i.e., x=(x 1, x 2, x 3, x 4 ). - travel time AB, travel time BC, travel time CD and travel time AD Results obtained by neuro-fuzzy model In the first experiment we split the first data set (4541 cases) on two halves. The first half we used for learning of model and the second for its testing, the results of the experiment with respect to ±10% marginal error are presented in Table 5. Note that due to fact that output of neuro-fuzzy system are exact values we can use the ±10% marginal error based correctness to compare results with these given in [3]. Predictions Travel time (0,25] (25,30] (30,40] (40,50] > 50 row sum < (25, 30] (30, 40] (40, 50] > col sum Table 5. Results of the first experiment with neuro-fuzzy model. In the second experiment we take the same strategy is in the first one, but we used the second data set (9333 cases). That is, we split the data on two halves, the first one was used for learning and the second for testing. Results with respect to ±10% marginal are in Table 6. Predictions Travel time < 25 (25, 30] (30, 40] (40, 50] > 50 row sum < (25, 30] (30, 40] (40, 50] > col sum Table 6. Results of the second experiment with neuro-fuzzy model Regarding the correctness figures for both models we have in the first case figures as 99,2 % of all predictions on whole testing data set correct, 0.5% too low and 0.3% to high. Aiming only at predictions in congested conditions the figures are 70.2% correct, 21% too low and 8.8% too high. In the case of second model (for the second data set) with respect to whole testing set we have 98.1% predictions correct, 1.1% too low and 0.9% too high. In congested conditions figures are as 77.5% correct, 15.6 too low and 6.9% too high. We see that both models have almost the same behaviour with respect to their correctness. The correctness is comparable well with results
9 in [3] and of Total Fuzzy Similarity method presented above. Note that neuro-fuzzy models are proper predictive ones, i.e., different data were used for building (learning) the models from data used for their testing. In the third experiment we use for learning the whole first data set (4541 cases) and created model was tested on data of the second set (9333 cases). The results with respect to ±10% marginal are presented in Table 7. Predictions Travel time < 25 (25, 30] (30, 40] (40, 50] >50 row sum < (25, 30] (30, 40] (40, 50] > col sum Table 7. The results of the third experiment with neuro-fuzzy model. Finally, reverse experiment to the third one was performed. That is, we learned the model on the second data set (9333 cases) and tested on the first one (4541 cases). The results with respect to ±10% marginal are in Table 8. Predictions Travel time < 25 (25, 30] (30, 40] (40, 50] >50 row sum < (25, 30] (30, 40] (40, 50] > col sum Table 8. The results of the fourth experiment with neuro-fuzzy model. When analysing Table 8 the figures are really disgusting, especially for congested conditions. For all 9333 testing cases we have 95.4% correctness of predictions, 3.8% are too low and 0.8% too high, which is still quite good but for congested conditions we have only 34.7% predictions correct, 55.8% are too low and 9.5% are too high. The reasons for such a bad performance lies in different distributions of congested cases in both data sets as presented in section % - (25,30]; 37.4% - (30,40]; 34% - (40,50]; 12.1% - >50 for the first data set and 26.9% - (25,30], 60% - (30,40], 13.1% - (40,50], 0.02% - >50 for the second set. We see that when learning on the first data set then the cases from (30,40] and (40,50] classes have almost equal distribution but in the second set the class (30,40] is much more preferred. The other reason is that in the first data set there are only 115 cases of congested conditions but in the second set we have 640 of them so there is more information about congested cases in the second set. This is also reflected in figures for fourth experiment where for all predictions we have 98.6% correct, 0.8% too low and 0.5% too high. and in congested conditions 60.0% correct predictions, 33% too low and 7% too high which is comparable with results of the first two experiments and of [3], especially noting the predictive character of our models.
10 4. Concluding remarks We have presented two new prediction models for a travel time prediction problem. The models are based on two data sets one containing 4541 cases and the second 9333 cases. The first model utilises GUHA data mining analysis and Total Fuzzy Similarity method, the second model is based on hierarchical neuro-fuzzy inference system. The introduced new models improve the results obtained in [3]. Both new prediction models are simple, too. The first experiments we have introduced strengthen an impression that similar but more complicated models would rise the prediction above 90% exactness level, a target imposed in [3]. Acknowledgements The authors wish to thank Satu Innamaa, Laura Lanne, Matti Pursula and the staff at Helsinki University of Technology, Transportation Engineering, for fruitful scientific comments and ideas in preparing the manuscript. In particular, thanks are due to Laura Lanne for her kind assistance. References [1] Coufal, D. Incremental Structure Learning of Three-Layered Gaussian RBF Networks. International Conference on Computational Science - ICCS 2002, Amsterdam, The Netherlands, April 21-24, 2002, Proceedings, Part III, Springer-Verlag [2] Hajek, P., Havranek, T. Mechanizing Hypothesis Formation - Mathematical Foundations for a General Theory. Berlin - Heidelberg -New York, Springer -Verlag, 1978, 396 p. Internet version (free) [3] Innamaa, S. Short-term prediction of highway travel time using MLP-neural networks. 8.th World Congress on Intelligent Transportation Systems. Sydney, Australia, 30 Sept. - 4 Oct [4] Rauch, J.: Mining for Scientific Hypotheses. In Meij, J.(Editor): Dealing with the data flood. Mining Data, Text and Multimedia. STT/Beweton, The Hague pp (An internet version of software available at [5] Press W.H., Teukolsky S.A., Vetterling W.T, Flannery B.P., Numerical Recipes in C, The Art of Scientific Computing, Second Edition, Cambridge University Press, 1992; internet version available at [6] Turunen, E. Mathematics Behind Fuzzy Sets. Springer-Verlag, Heidelberg [7] Wang L.X., Mendel J.M. Fuzzy basis function, universal approximation, and orthogonal least-squares learning. IEEE Trans. on Neural Networks, vol.3, no.5, pp APPENDIX 1. A GUHA ANALYSIS RESULT Antecedent Travel time AB from 11 to 12 min and Travel time CD from 6 to 7 min Succedent Prediction from 30 to 40 min Frequency table
11 Succeden NOT Succedent Row Sum Antecedent NOT Antecedent Column Sum
Chapter 2 The Research on Fault Diagnosis of Building Electrical System Based on RBF Neural Network
Chapter 2 The Research on Fault Diagnosis of Building Electrical System Based on RBF Neural Network Qian Wu, Yahui Wang, Long Zhang and Li Shen Abstract Building electrical system fault diagnosis is the
More informationFUZZY CLUSTERING ANALYSIS OF DATA MINING: APPLICATION TO AN ACCIDENT MINING SYSTEM
International Journal of Innovative Computing, Information and Control ICIC International c 0 ISSN 34-48 Volume 8, Number 8, August 0 pp. 4 FUZZY CLUSTERING ANALYSIS OF DATA MINING: APPLICATION TO AN ACCIDENT
More informationData quality in Accounting Information Systems
Data quality in Accounting Information Systems Comparing Several Data Mining Techniques Erjon Zoto Department of Statistics and Applied Informatics Faculty of Economy, University of Tirana Tirana, Albania
More informationLU Factorization Method to Solve Linear Programming Problem
Website: wwwijetaecom (ISSN 2250-2459 ISO 9001:2008 Certified Journal Volume 4 Issue 4 April 2014) LU Factorization Method to Solve Linear Programming Problem S M Chinchole 1 A P Bhadane 2 12 Assistant
More informationOVERVIEW OF THE GUHA METHOD AS A DATA MINING TECHNIQUE
OVERVIEW OF THE GUHA METHOD AS A DATA MINING TECHNIQUE Viktor Todorovski Faculty of Electrical Engineering and Information Technology, UKIM Skopje, R. of Macedonia Ivan Chorbev Faculty of Electrical Engineering
More informationDecember 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B. KITCHENS
December 4, 2013 MATH 171 BASIC LINEAR ALGEBRA B KITCHENS The equation 1 Lines in two-dimensional space (1) 2x y = 3 describes a line in two-dimensional space The coefficients of x and y in the equation
More informationProgramming Risk Assessment Models for Online Security Evaluation Systems
Programming Risk Assessment Models for Online Security Evaluation Systems Ajith Abraham 1, Crina Grosan 12, Vaclav Snasel 13 1 Machine Intelligence Research Labs, MIR Labs, http://www.mirlabs.org 2 Babes-Bolyai
More informationNTC Project: S01-PH10 (formerly I01-P10) 1 Forecasting Women s Apparel Sales Using Mathematical Modeling
1 Forecasting Women s Apparel Sales Using Mathematical Modeling Celia Frank* 1, Balaji Vemulapalli 1, Les M. Sztandera 2, Amar Raheja 3 1 School of Textiles and Materials Technology 2 Computer Information
More informationSome Polynomial Theorems. John Kennedy Mathematics Department Santa Monica College 1900 Pico Blvd. Santa Monica, CA 90405 rkennedy@ix.netcom.
Some Polynomial Theorems by John Kennedy Mathematics Department Santa Monica College 1900 Pico Blvd. Santa Monica, CA 90405 rkennedy@ix.netcom.com This paper contains a collection of 31 theorems, lemmas,
More informationCHARACTERISTICS IN FLIGHT DATA ESTIMATION WITH LOGISTIC REGRESSION AND SUPPORT VECTOR MACHINES
CHARACTERISTICS IN FLIGHT DATA ESTIMATION WITH LOGISTIC REGRESSION AND SUPPORT VECTOR MACHINES Claus Gwiggner, Ecole Polytechnique, LIX, Palaiseau, France Gert Lanckriet, University of Berkeley, EECS,
More informationAnalysis of an Artificial Hormone System (Extended abstract)
c 2013. This is the author s version of the work. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purpose or for creating
More informationOptimizing the prediction models of the air quality state in cities
Air Pollution XV 89 Optimizing the prediction models of the air quality state in cities J. Skrzypski, E. Jach-Szakiel & W. Kamiński Faculty of Process and Environmental Engineering, Technical University
More informationClassification of Fuzzy Data in Database Management System
Classification of Fuzzy Data in Database Management System Deval Popat, Hema Sharda, and David Taniar 2 School of Electrical and Computer Engineering, RMIT University, Melbourne, Australia Phone: +6 3
More informationSINGULAR SPECTRUM ANALYSIS HYBRID FORECASTING METHODS WITH APPLICATION TO AIR TRANSPORT DEMAND
SINGULAR SPECTRUM ANALYSIS HYBRID FORECASTING METHODS WITH APPLICATION TO AIR TRANSPORT DEMAND K. Adjenughwure, Delft University of Technology, Transport Institute, Ph.D. candidate V. Balopoulos, Democritus
More informationDATA MINING TECHNOLOGY. Keywords: data mining, data warehouse, knowledge discovery, OLAP, OLAM.
DATA MINING TECHNOLOGY Georgiana Marin 1 Abstract In terms of data processing, classical statistical models are restrictive; it requires hypotheses, the knowledge and experience of specialists, equations,
More informationARTIFICIAL NEURAL NETWORKS FOR ADAPTIVE MANAGEMENT TRAFFIC LIGHT OBJECTS AT THE INTERSECTION
The 10 th International Conference RELIABILITY and STATISTICS in TRANSPORTATION and COMMUNICATION - 2010 Proceedings of the 10th International Conference Reliability and Statistics in Transportation and
More informationTwo classes of ternary codes and their weight distributions
Two classes of ternary codes and their weight distributions Cunsheng Ding, Torleiv Kløve, and Francesco Sica Abstract In this paper we describe two classes of ternary codes, determine their minimum weight
More informationVisualization of Breast Cancer Data by SOM Component Planes
International Journal of Science and Technology Volume 3 No. 2, February, 2014 Visualization of Breast Cancer Data by SOM Component Planes P.Venkatesan. 1, M.Mullai 2 1 Department of Statistics,NIRT(Indian
More informationFurther Analysis Of A Framework To Analyze Network Performance Based On Information Quality
Further Analysis Of A Framework To Analyze Network Performance Based On Information Quality A Kazmierczak Computer Information Systems Northwest Arkansas Community College One College Dr. Bentonville,
More informationEFFICIENT DATA PRE-PROCESSING FOR DATA MINING
EFFICIENT DATA PRE-PROCESSING FOR DATA MINING USING NEURAL NETWORKS JothiKumar.R 1, Sivabalan.R.V 2 1 Research scholar, Noorul Islam University, Nagercoil, India Assistant Professor, Adhiparasakthi College
More informationTHE DIMENSION OF A VECTOR SPACE
THE DIMENSION OF A VECTOR SPACE KEITH CONRAD This handout is a supplementary discussion leading up to the definition of dimension and some of its basic properties. Let V be a vector space over a field
More informationA Comparative Study of the Pickup Method and its Variations Using a Simulated Hotel Reservation Data
A Comparative Study of the Pickup Method and its Variations Using a Simulated Hotel Reservation Data Athanasius Zakhary, Neamat El Gayar Faculty of Computers and Information Cairo University, Giza, Egypt
More informationAccurate and robust image superresolution by neural processing of local image representations
Accurate and robust image superresolution by neural processing of local image representations Carlos Miravet 1,2 and Francisco B. Rodríguez 1 1 Grupo de Neurocomputación Biológica (GNB), Escuela Politécnica
More informationFollow links Class Use and other Permissions. For more information, send email to: permissions@pupress.princeton.edu
COPYRIGHT NOTICE: David A. Kendrick, P. Ruben Mercado, and Hans M. Amman: Computational Economics is published by Princeton University Press and copyrighted, 2006, by Princeton University Press. All rights
More informationOPRE 6201 : 2. Simplex Method
OPRE 6201 : 2. Simplex Method 1 The Graphical Method: An Example Consider the following linear program: Max 4x 1 +3x 2 Subject to: 2x 1 +3x 2 6 (1) 3x 1 +2x 2 3 (2) 2x 2 5 (3) 2x 1 +x 2 4 (4) x 1, x 2
More informationThe Combination Forecasting Model of Auto Sales Based on Seasonal Index and RBF Neural Network
, pp.67-76 http://dx.doi.org/10.14257/ijdta.2016.9.1.06 The Combination Forecasting Model of Auto Sales Based on Seasonal Index and RBF Neural Network Lihua Yang and Baolin Li* School of Economics and
More informationThe Gravity Model: Derivation and Calibration
The Gravity Model: Derivation and Calibration Philip A. Viton October 28, 2014 Philip A. Viton CRP/CE 5700 () Gravity Model October 28, 2014 1 / 66 Introduction We turn now to the Gravity Model of trip
More informationHigh Frequency Trading using Fuzzy Momentum Analysis
Proceedings of the World Congress on Engineering 2 Vol I WCE 2, June 3 - July 2, 2, London, U.K. High Frequency Trading using Fuzzy Momentum Analysis A. Kablan Member, IAENG, and W. L. Ng. Abstract High
More informationARTIFICIAL INTELLIGENCE METHODS IN STOCK INDEX PREDICTION WITH THE USE OF NEWSPAPER ARTICLES
FOUNDATION OF CONTROL AND MANAGEMENT SCIENCES No Year Manuscripts Mateusz, KOBOS * Jacek, MAŃDZIUK ** ARTIFICIAL INTELLIGENCE METHODS IN STOCK INDEX PREDICTION WITH THE USE OF NEWSPAPER ARTICLES Analysis
More informationKnowledge Acquisition Approach Based on Rough Set in Online Aided Decision System for Food Processing Quality and Safety
, pp. 381-388 http://dx.doi.org/10.14257/ijunesst.2014.7.6.33 Knowledge Acquisition Approach Based on Rough Set in Online Aided ecision System for Food Processing Quality and Safety Liu Peng, Liu Wen,
More informationOptimization of Fuzzy Inventory Models under Fuzzy Demand and Fuzzy Lead Time
Tamsui Oxford Journal of Management Sciences, Vol. 0, No. (-6) Optimization of Fuzzy Inventory Models under Fuzzy Demand and Fuzzy Lead Time Chih-Hsun Hsieh (Received September 9, 00; Revised October,
More informationKnowledge Discovery in Stock Market Data
Knowledge Discovery in Stock Market Data Alfred Ultsch and Hermann Locarek-Junge Abstract This work presents the results of a Data Mining and Knowledge Discovery approach on data from the stock markets
More informationSYSM 6304: Risk and Decision Analysis Lecture 5: Methods of Risk Analysis
SYSM 6304: Risk and Decision Analysis Lecture 5: Methods of Risk Analysis M. Vidyasagar Cecil & Ida Green Chair The University of Texas at Dallas Email: M.Vidyasagar@utdallas.edu October 17, 2015 Outline
More informationD A T A M I N I N G C L A S S I F I C A T I O N
D A T A M I N I N G C L A S S I F I C A T I O N FABRICIO VOZNIKA LEO NARDO VIA NA INTRODUCTION Nowadays there is huge amount of data being collected and stored in databases everywhere across the globe.
More informationPredictive Dynamix Inc
Predictive Modeling Technology Predictive modeling is concerned with analyzing patterns and trends in historical and operational data in order to transform data into actionable decisions. This is accomplished
More informationIntrusion Detection via Machine Learning for SCADA System Protection
Intrusion Detection via Machine Learning for SCADA System Protection S.L.P. Yasakethu Department of Computing, University of Surrey, Guildford, GU2 7XH, UK. s.l.yasakethu@surrey.ac.uk J. Jiang Department
More informationTowards Rule-based System for the Assembly of 3D Bricks
Universal Journal of Communications and Network 3(4): 77-81, 2015 DOI: 10.13189/ujcn.2015.030401 http://www.hrpub.org Towards Rule-based System for the Assembly of 3D Bricks Sanguk Noh School of Computer
More informationOn the Interaction and Competition among Internet Service Providers
On the Interaction and Competition among Internet Service Providers Sam C.M. Lee John C.S. Lui + Abstract The current Internet architecture comprises of different privately owned Internet service providers
More informationExtracting Fuzzy Rules from Data for Function Approximation and Pattern Classification
Extracting Fuzzy Rules from Data for Function Approximation and Pattern Classification Chapter 9 in Fuzzy Information Engineering: A Guided Tour of Applications, ed. D. Dubois, H. Prade, and R. Yager,
More informationData Mining Project Report. Document Clustering. Meryem Uzun-Per
Data Mining Project Report Document Clustering Meryem Uzun-Per 504112506 Table of Content Table of Content... 2 1. Project Definition... 3 2. Literature Survey... 3 3. Methods... 4 3.1. K-means algorithm...
More informationEM Clustering Approach for Multi-Dimensional Analysis of Big Data Set
EM Clustering Approach for Multi-Dimensional Analysis of Big Data Set Amhmed A. Bhih School of Electrical and Electronic Engineering Princy Johnson School of Electrical and Electronic Engineering Martin
More informationProceedings of the 9th WSEAS International Conference on APPLIED COMPUTER SCIENCE
Automated Futures Trading Environment Effect on the Decision Making PETR TUCNIK Department of Information Technologies University of Hradec Kralove Rokitanskeho 62, 500 02 Hradec Kralove CZECH REPUBLIC
More informationComponent Ordering in Independent Component Analysis Based on Data Power
Component Ordering in Independent Component Analysis Based on Data Power Anne Hendrikse Raymond Veldhuis University of Twente University of Twente Fac. EEMCS, Signals and Systems Group Fac. EEMCS, Signals
More informationImproving Decision Making and Managing Knowledge
Improving Decision Making and Managing Knowledge Decision Making and Information Systems Information Requirements of Key Decision-Making Groups in a Firm Senior managers, middle managers, operational managers,
More informationPrediction Model for Crude Oil Price Using Artificial Neural Networks
Applied Mathematical Sciences, Vol. 8, 2014, no. 80, 3953-3965 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2014.43193 Prediction Model for Crude Oil Price Using Artificial Neural Networks
More informationContinued Fractions and the Euclidean Algorithm
Continued Fractions and the Euclidean Algorithm Lecture notes prepared for MATH 326, Spring 997 Department of Mathematics and Statistics University at Albany William F Hammond Table of Contents Introduction
More informationGESTIKULATOR Generator of a tetrahedral mesh on a sphere
GESTIKULATOR Generator of a tetrahedral mesh on a sphere Jakub Velímský Department of Geophysics Faculty of Mathematics and Physics Charles University in Prague V Holešovičkách 2 180 00 Prague Czech Republic
More information2x + y = 3. Since the second equation is precisely the same as the first equation, it is enough to find x and y satisfying the system
1. Systems of linear equations We are interested in the solutions to systems of linear equations. A linear equation is of the form 3x 5y + 2z + w = 3. The key thing is that we don t multiply the variables
More informationAn Anomaly-Based Method for DDoS Attacks Detection using RBF Neural Networks
2011 International Conference on Network and Electronics Engineering IPCSIT vol.11 (2011) (2011) IACSIT Press, Singapore An Anomaly-Based Method for DDoS Attacks Detection using RBF Neural Networks Reyhaneh
More informationEvaluation of the Time Needed for e-learning Course Developing *
BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 9, No 3 Sofia 2009 Evaluation of the Time Needed for e-learning Course Developing * Petar Halachev, Ivan Mustakerov Institute
More informationNotes on Factoring. MA 206 Kurt Bryan
The General Approach Notes on Factoring MA 26 Kurt Bryan Suppose I hand you n, a 2 digit integer and tell you that n is composite, with smallest prime factor around 5 digits. Finding a nontrivial factor
More informationMATH 590: Meshfree Methods
MATH 590: Meshfree Methods Chapter 7: Conditionally Positive Definite Functions Greg Fasshauer Department of Applied Mathematics Illinois Institute of Technology Fall 2010 fasshauer@iit.edu MATH 590 Chapter
More informationAPPLYING GMDH ALGORITHM TO EXTRACT RULES FROM EXAMPLES
Systems Analysis Modelling Simulation Vol. 43, No. 10, October 2003, pp. 1311-1319 APPLYING GMDH ALGORITHM TO EXTRACT RULES FROM EXAMPLES KOJI FUJIMOTO* and SAMPEI NAKABAYASHI Financial Engineering Group,
More informationMaximization versus environmental compliance
Maximization versus environmental compliance Increase use of alternative fuels with no risk for quality and environment Reprint from World Cement March 2005 Dr. Eduardo Gallestey, ABB, Switzerland, discusses
More informationLecture 3: Finding integer solutions to systems of linear equations
Lecture 3: Finding integer solutions to systems of linear equations Algorithmic Number Theory (Fall 2014) Rutgers University Swastik Kopparty Scribe: Abhishek Bhrushundi 1 Overview The goal of this lecture
More informationIntroduction to Machine Learning and Data Mining. Prof. Dr. Igor Trajkovski trajkovski@nyus.edu.mk
Introduction to Machine Learning and Data Mining Prof. Dr. Igor Trakovski trakovski@nyus.edu.mk Neural Networks 2 Neural Networks Analogy to biological neural systems, the most robust learning systems
More informationA Data Mining Analysis to evaluate the additional workloads caused by welding distortions
A Data Mining Analysis to evaluate the additional workloads caused by welding distortions Losseau N., Caprace J.D., Aracil Fernandez F., Rigo P. ABSTRACT: This paper presents a way to minimize cost in
More informationOBJECTIVE ASSESSMENT OF FORECASTING ASSIGNMENTS USING SOME FUNCTION OF PREDICTION ERRORS
OBJECTIVE ASSESSMENT OF FORECASTING ASSIGNMENTS USING SOME FUNCTION OF PREDICTION ERRORS CLARKE, Stephen R. Swinburne University of Technology Australia One way of examining forecasting methods via assignments
More informationEvaluation of Processor Health within Hierarchical Condition Monitoring System
Evaluation of Processor Health within Hierarchical Condition Monitoring System Lenka Pavelková and Ladislav Jirsa Department of Adaptive Systems, Institute of Information Theory and Automation, Czech Academy
More informationDiscussion on the paper Hypotheses testing by convex optimization by A. Goldenschluger, A. Juditsky and A. Nemirovski.
Discussion on the paper Hypotheses testing by convex optimization by A. Goldenschluger, A. Juditsky and A. Nemirovski. Fabienne Comte, Celine Duval, Valentine Genon-Catalot To cite this version: Fabienne
More informationModelling Complex Patterns by Information Systems
Fundamenta Informaticae 67 (2005) 203 217 203 IOS Press Modelling Complex Patterns by Information Systems Jarosław Stepaniuk Department of Computer Science, Białystok University of Technology Wiejska 45a,
More informationA note on companion matrices
Linear Algebra and its Applications 372 (2003) 325 33 www.elsevier.com/locate/laa A note on companion matrices Miroslav Fiedler Academy of Sciences of the Czech Republic Institute of Computer Science Pod
More informationArtificial Neural Networks are bio-inspired mechanisms for intelligent decision support. Artificial Neural Networks. Research Article 2014
An Experiment to Signify Fuzzy Logic as an Effective User Interface Tool for Artificial Neural Network Nisha Macwan *, Priti Srinivas Sajja G.H. Patel Department of Computer Science India Abstract Artificial
More informationIntroduction to Machine Learning Using Python. Vikram Kamath
Introduction to Machine Learning Using Python Vikram Kamath Contents: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Introduction/Definition Where and Why ML is used Types of Learning Supervised Learning Linear Regression
More informationImproving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation
Improving the Performance of TCP Using Window Adjustment Procedure and Bandwidth Estimation R.Navaneethakrishnan Assistant Professor (SG) Bharathiyar College of Engineering and Technology, Karaikal, India.
More information2004 Networks UK Publishers. Reprinted with permission.
Riikka Susitaival and Samuli Aalto. Adaptive load balancing with OSPF. In Proceedings of the Second International Working Conference on Performance Modelling and Evaluation of Heterogeneous Networks (HET
More informationChapter 11. Managing Knowledge
Chapter 11 Managing Knowledge VIDEO CASES Video Case 1: How IBM s Watson Became a Jeopardy Champion. Video Case 2: Tour: Alfresco: Open Source Document Management System Video Case 3: L'Oréal: Knowledge
More informationA New Approach for Evaluation of Data Mining Techniques
181 A New Approach for Evaluation of Data Mining s Moawia Elfaki Yahia 1, Murtada El-mukashfi El-taher 2 1 College of Computer Science and IT King Faisal University Saudi Arabia, Alhasa 31982 2 Faculty
More informationE3: PROBABILITY AND STATISTICS lecture notes
E3: PROBABILITY AND STATISTICS lecture notes 2 Contents 1 PROBABILITY THEORY 7 1.1 Experiments and random events............................ 7 1.2 Certain event. Impossible event............................
More informationFigure 1. The cloud scales: Amazon EC2 growth [2].
- Chung-Cheng Li and Kuochen Wang Department of Computer Science National Chiao Tung University Hsinchu, Taiwan 300 shinji10343@hotmail.com, kwang@cs.nctu.edu.tw Abstract One of the most important issues
More informationMATHEMATICS FOR ENGINEERING BASIC ALGEBRA
MATHEMATICS FOR ENGINEERING BASIC ALGEBRA TUTORIAL 3 EQUATIONS This is the one of a series of basic tutorials in mathematics aimed at beginners or anyone wanting to refresh themselves on fundamentals.
More informationPredicting the Risk of Heart Attacks using Neural Network and Decision Tree
Predicting the Risk of Heart Attacks using Neural Network and Decision Tree S.Florence 1, N.G.Bhuvaneswari Amma 2, G.Annapoorani 3, K.Malathi 4 PG Scholar, Indian Institute of Information Technology, Srirangam,
More informationRecurrent Neural Networks
Recurrent Neural Networks Neural Computation : Lecture 12 John A. Bullinaria, 2015 1. Recurrent Neural Network Architectures 2. State Space Models and Dynamical Systems 3. Backpropagation Through Time
More informationDATA SECURITY BASED ON NEURAL NETWORKS
TASKQUARTERLY9No4,409 414 DATA SECURITY BASED ON NEURAL NETWORKS KHALED M. G. NOAMAN AND HAMID ABDULLAH JALAB Faculty of Science, Computer Science Department, Sana a University, P.O. Box 13499, Sana a,
More informationFRAUD DETECTION IN ELECTRIC POWER DISTRIBUTION NETWORKS USING AN ANN-BASED KNOWLEDGE-DISCOVERY PROCESS
FRAUD DETECTION IN ELECTRIC POWER DISTRIBUTION NETWORKS USING AN ANN-BASED KNOWLEDGE-DISCOVERY PROCESS Breno C. Costa, Bruno. L. A. Alberto, André M. Portela, W. Maduro, Esdras O. Eler PDITec, Belo Horizonte,
More informationCSC384 Intro to Artificial Intelligence
CSC384 Intro to Artificial Intelligence What is Artificial Intelligence? What is Intelligence? Are these Intelligent? CSC384, University of Toronto 3 What is Intelligence? Webster says: The capacity to
More informationAdvanced Ensemble Strategies for Polynomial Models
Advanced Ensemble Strategies for Polynomial Models Pavel Kordík 1, Jan Černý 2 1 Dept. of Computer Science, Faculty of Information Technology, Czech Technical University in Prague, 2 Dept. of Computer
More informationMATRIX ALGEBRA AND SYSTEMS OF EQUATIONS
MATRIX ALGEBRA AND SYSTEMS OF EQUATIONS Systems of Equations and Matrices Representation of a linear system The general system of m equations in n unknowns can be written a x + a 2 x 2 + + a n x n b a
More informationTHREE DIMENSIONAL GEOMETRY
Chapter 8 THREE DIMENSIONAL GEOMETRY 8.1 Introduction In this chapter we present a vector algebra approach to three dimensional geometry. The aim is to present standard properties of lines and planes,
More informationBayesian Statistics: Indian Buffet Process
Bayesian Statistics: Indian Buffet Process Ilker Yildirim Department of Brain and Cognitive Sciences University of Rochester Rochester, NY 14627 August 2012 Reference: Most of the material in this note
More informationLEO I{ATZ MICHIGAN STATE COLLEGE
PSYCHOMETRIKA~VOL. ]8, NO. 1 UAaCH, 1953 A NEW STATUS INDEX DERIVED FROM SOCIOMETRIC ANALYSIS* LEO I{ATZ MICHIGAN STATE COLLEGE For the purpose of evaluating status in a manner free from the deficiencies
More informationReal Time Traffic Balancing in Cellular Network by Multi- Criteria Handoff Algorithm Using Fuzzy Logic
Real Time Traffic Balancing in Cellular Network by Multi- Criteria Handoff Algorithm Using Fuzzy Logic Solomon.T.Girma 1, Dominic B. O. Konditi 2, Edward N. Ndungu 3 1 Department of Electrical Engineering,
More informationSimilarity and Diagonalization. Similar Matrices
MATH022 Linear Algebra Brief lecture notes 48 Similarity and Diagonalization Similar Matrices Let A and B be n n matrices. We say that A is similar to B if there is an invertible n n matrix P such that
More informationLearning is a very general term denoting the way in which agents:
What is learning? Learning is a very general term denoting the way in which agents: Acquire and organize knowledge (by building, modifying and organizing internal representations of some external reality);
More informationSymbolic Determinants: Calculating the Degree
Symbolic Determinants: Calculating the Degree Technical Report by Brent M. Dingle Texas A&M University Original: May 4 Updated: July 5 Abstract: There are many methods for calculating the determinant of
More informationDESIGN AND STRUCTURE OF FUZZY LOGIC USING ADAPTIVE ONLINE LEARNING SYSTEMS
Abstract: Fuzzy logic has rapidly become one of the most successful of today s technologies for developing sophisticated control systems. The reason for which is very simple. Fuzzy logic addresses such
More informationData Analysis Tools. Tools for Summarizing Data
Data Analysis Tools This section of the notes is meant to introduce you to many of the tools that are provided by Excel under the Tools/Data Analysis menu item. If your computer does not have that tool
More informationOptimized Fuzzy Control by Particle Swarm Optimization Technique for Control of CSTR
International Journal of Computer, Electrical, Automation, Control and Information Engineering Vol:5, No:, 20 Optimized Fuzzy Control by Particle Swarm Optimization Technique for Control of CSTR Saeed
More informationA FUZZY LOGIC APPROACH FOR SALES FORECASTING
A FUZZY LOGIC APPROACH FOR SALES FORECASTING ABSTRACT Sales forecasting proved to be very important in marketing where managers need to learn from historical data. Many methods have become available for
More informationFlexible mobility management strategy in cellular networks
Flexible mobility management strategy in cellular networks JAN GAJDORUS Department of informatics and telecommunications (161114) Czech technical university in Prague, Faculty of transportation sciences
More informationNine Common Types of Data Mining Techniques Used in Predictive Analytics
1 Nine Common Types of Data Mining Techniques Used in Predictive Analytics By Laura Patterson, President, VisionEdge Marketing Predictive analytics enable you to develop mathematical models to help better
More informationON INTEGRATING UNSUPERVISED AND SUPERVISED CLASSIFICATION FOR CREDIT RISK EVALUATION
ISSN 9 X INFORMATION TECHNOLOGY AND CONTROL, 00, Vol., No.A ON INTEGRATING UNSUPERVISED AND SUPERVISED CLASSIFICATION FOR CREDIT RISK EVALUATION Danuta Zakrzewska Institute of Computer Science, Technical
More informationApproaches to Qualitative Evaluation of the Software Quality Attributes: Overview
4th International Conference on Software Methodologies, Tools and Techniques Approaches to Qualitative Evaluation of the Software Quality Attributes: Overview Presented by: Denis Kozlov Department of Computer
More informationQOS Based Web Service Ranking Using Fuzzy C-means Clusters
Research Journal of Applied Sciences, Engineering and Technology 10(9): 1045-1050, 2015 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scientific Organization, 2015 Submitted: March 19, 2015 Accepted: April
More informationFeasibility Study Inputs based on Requirements Engineering
Feasibility Study Inputs based on Requirements Engineering Robert Pergl Department of Information Engineering, Faculty of Economics and Management, Czech University of Life Sciences, Prague, Czech Republic
More informationReal-Time Traffic Patrol Allocation for Abu Dhabi Emirate (UAE)
Real-Time Traffic Patrol Allocation for Abu Dhabi Emirate (UAE) Hussain Al-Harthei Traffic and Patrols Directorate, Abu Dhabi Police, Email: alharthei@saaed.ae Oualid (Walid) Ben Ali University of Sharjah,
More informationForecasting of Economic Quantities using Fuzzy Autoregressive Model and Fuzzy Neural Network
Forecasting of Economic Quantities using Fuzzy Autoregressive Model and Fuzzy Neural Network Dušan Marček 1 Abstract Most models for the time series of stock prices have centered on autoregressive (AR)
More informationParallel Data Selection Based on Neurodynamic Optimization in the Era of Big Data
Parallel Data Selection Based on Neurodynamic Optimization in the Era of Big Data Jun Wang Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Shatin, New Territories,
More informationEngineering Problem Solving and Excel. EGN 1006 Introduction to Engineering
Engineering Problem Solving and Excel EGN 1006 Introduction to Engineering Mathematical Solution Procedures Commonly Used in Engineering Analysis Data Analysis Techniques (Statistics) Curve Fitting techniques
More information