Role of Social Networking in Marketing using Data Mining Mrs. Saroj Junghare Astt. Professor, Department of Computer Science and Application St. Aloysius College, Jabalpur, Madhya Pradesh, India Abstract: Social Networking is a phenomenon that has transformed the interaction and communication of individuals throughout the world. Data mining is the exploration and analysis of large quantities of data in order to discover valid, novel, potentially useful and ultimately understandable patterns in data. Huge amount of data generated by Social Networking sites can be used by analysts to analyze useful data. This data is utilized in many fields such as making business strategies for sales, marketing, government agencies etc. Analysts use various Data Mining techniques to generate different patterns based on their selection criteria. In this paper we are discussing some parameters of social networking sites, data mining techniques and its uses in business strategy based on statics of social media. Keywords: Social Networking Sites, Data Mining, Data Mining Techniques, Social Media Marketing, Marketing. Social Networking Sites Social Networking Sites connect people with similar interests and background. E.g.: Facebook and LinkedIn. There are 10 parameters of social media marketing. Parameters are as follows [2] : Goal describes what are the goals for social media campaign? Specific explains whom do we want to engage in? What states the needs of the audience? How defines how do you wish to engage the audience? Where suggests what type or platform of social media are we going to use? Timed tells what is the best time to post according to audience usage? Consumer-Mindset suggests what kind of content does the consumer relates to? What are the views of consumer about the branded the usage of platform. Where the campaign is carried out? Mail: editor@globalresearch.co.in 27
Assessment shows how are we going to measure results of the campaign? E.g.: use of Google analytics to observe the traffic on your website. Observe your competitor states what are the strategy of your competitor both in online as well as offline domain. Customer Characteristics include location, gender, age, likes and interests, relationship status, workplace and education. Data Mining Data mining is the process of analyzing data from different perspectives and summarizing it into useful information, information that can be used to increase revenue, cuts costs or both. Data mining is the process of finding correlations or patterns among dozens of fields in large relational databases [7]. Data mining tasks include predictive tasks and descriptive tasks. Predictive tasks make predictions based on user data collected whereas the descriptive tasks determine relationships among data, formulate patterns to depict relationship [4]. Table 1: Predictive Tasks: Classification, Regression and Anomaly Detection [1]. Techniques Classification Classification function is created by analyzing the relationship between attributes and data objects whose class labels are well known [5]. Fig. (a) : Classification 1. Logistic Regression 2. Naive Bayes 3. Support Vector Machine 4. Decision Tree Regression Regression is a function that maps a data item to a real valued prediction variable [5]. Fig. (b) : Regression Anomaly Detection Deviation detection detects deviation from normal behavior. Multiple Regression Support Vector Machine Fig. (c) : Anomaly Detection One-Class Support Vector Machine Mail: editor@globalresearch.co.in 28
Table 2: Descriptive Tasks: Attribute Importance, Clustering, Association and Feature Selection and Extraction [1]. Technique Attribute Importance Attribute importance is a supervised function that identifies and ranks the attributes that are most important in predicting a target attribute. Fig. (d) : Attribute Importance Minimum Description Length Clustering Clustering is the process of making group of abstract objects into similar subclasses [5] Fig. (e) : Clustering Enhanced K-Means Orthogonal Partitioning Clustering Expectation Maximization Association Association learning is if/then statements that identify associations/relationships between unrelated data [5] Fig. (f) : Association Apriori Feature Selection and Extraction Feature extraction is the second class of methods for dimension reduction. It creates new attributes (features) using linear combinations of the (original existing) attributes. Fig. (g) : Feature Selection and Extraction Data Mining and Social Networking Sites Non-negative Matrix Factorization Principal Components Analysis (PCA) Singular Vector Decomposition Mail: editor@globalresearch.co.in 29
While large-scale information technology has been evolving separate transaction and analytical systems, data mining provides the link between the two. Data mining software analyzes relationships and patterns in stored user data based on open-ended user queries. Social media has vast amount of user-generated data which can be utilized for data mining. Data mining of social media can amplify use of social media and perk up commercial intelligence to transport enhanced services. For example, data mining techniques can identify user sentiments for anticipatory preparation to develop suggestion systems for business of specific products and even to build new friendships or connect certain interest groups.(facebook uses likes, groups as well as posts of users to recommend users specific ads as well as new pages and groups.) [2]. Marketing experts are searching for means to utilize them for their sales and advertising teams. Quite a lot of types of pattern detection that are normally used in social media data mining include: [6] Sequential Patterns, Association learning, Clustering, Classification, Regression, Deviation detection, and Summarization as described above [2]. Data mining can be used to make business strategy for [2] : 1 Cart Analysis: It is based on future prediction of customer behavior by past performance, including purchases and preferences. 2 Future sales: Predicting future sales plays vital role for the retail company. 3 Database Marketing: Database marketing is one of the most successful business applications of data mining. Mining of historical customer data helps determine patterns and trends to build customer profile for effective marketing. 4 Merchandise Planning: This is helpful for offline or online companies. For the offline, a company looking to grow by adding stores can evaluate the amount of merchandise they will need by looking at the exact layout of a current store. For an online business, merchandise planning can help you determine stocking options and inventory warehousing. 5 Card Marketing: If your business involves issuing credit cards, you can collect the information from usage, identify customer segments and then based on information on these segments boost acquisition, target products to develop and design prices. 6 Customer Loyalty: Retailers maintain loyalty management system to provide data for customer driven marketing. Most retailers today have adopted such a system, which is extensively used for reward point accumulation & redemption. 7 Market Segmentation: One of the best uses of data mining is to segment your customers. Identify the common characteristics of customers who buy the same products from your company. Given below is the list of Statics of Social Media Users based on gender, age, urbanity, education attainment and who use social networking tools [3]. Table 3: Statics based on Urbanity and Age Urbanity Sites Urban Suburban Rural Twitter 20 14 12 Pinterest 13 16 18 Instagram 17 11 11 Tumblr 7 5 6 Facebook 72 65 63 Total 70 67 61 Age Sites 18-29 30-49 50-64 65+ Twitter 13 5 3 1 Pinterest 19 19 12 4 Instagram 28 16 3 2 Tumblr 27 17 11 2 Mail: editor@globalresearch.co.in 30
Facebook 86 73 57 35 Total 83 77 52 32 Table 4: Statics based on Education Attainment and Gender Education Attainment Sites College+ Some College Up to high school grad Twitter 15 17 15 Pinterest 20 16 11 Instagram 12 15 12 Tumblr 7 6 5 Facebook 69 73 60 Total 65 69 66 Gender Sites Female Male Twitter 15 17 Pinterest 25 5 Instagram 16 10 Tumblr 6 6 Facebook 72 62 Total 71 62 Table 5: Statics based on Users who use social networking tools Users who use social networking tools Sites Users (%) Twitter 16 Pinterest 15 Instagram 13 Tumblr 6 Facebook 67 Total 67 Fig. (h) : Users who use social networking tools Marketing experts use these statistics for marketing/ advertising their products / applications / games on social media networks. For example Gender can be used to market the products suitable for that particular gender such as female group are attracted towards jewelry,fashionable clothes, cosmetic products etc. whereas male group are attracted towards gadgets, Mail: editor@globalresearch.co.in 31
games, applications etc. Similarly other criteria can be used to classify the data and can be used in different marketing strategies. Conclusion It can be concluded that social media mining is a new initiative to build new business strategies. The Social media houses vast amount of user-generated data which can be used for data mining, therefore guarantee a huge potential in terms of knowledge. Market ventures are using various mining techniques to gain insight into business information for the intake of their sales/marketing and advertising teams. The mined information from social platforms can significantly impact business strategy of any business enterprise. References [1] http://www.oracle.com/technetwork/database/options/advanced-analytics/odm/odmtechniques-algorithms-097163.html; last accessed on 17.4.2015 [2] Pippal S., Batra L., Krishna A., Gupta H., Arora K., Data mining in social networking sites: A social media mining approach to generate effective business strategies, IJIACS, Vol. 3, Issue 2, April 2014 [3] http://www.mediabistro.com/alltwitter/social-media-user-demographics_b38095; last accessed on 15.04.2015 [4] Data Mining Tasks, http://shodhganga.inflibnet.ac.in:8080/jspui/bitstream/10603/6897/13/13_chapter%208.pdf; last accessed on: 18.04.2015 [5] The Primary Tasks of Data Mining, http://www2.cs.uregina.ca/~dbd/cs831/notes/kdd/2_tasks.html; last accessed on 15.04.2015 [6] Data Mining: Tasks, Techniques and Applications, http://academic.csuohio.edu/fuy/pub/pot97.pdf; last accessed on 18.04.2015 [7] Data Mining, http://iveybusinessjournal.com/topics/strategy/four-strategies-to-captureand-create-value-from-big-data#.vm8bt7zi9ok; last accessed on 15.04.2015 Mail: editor@globalresearch.co.in 32