8 ijaers feb 2016 18 effective ranking in social media

Page 1

International Journal of Advanced Engineering Research and Science (IJAERS)

Vol-3, Issue-2 , Feb- 2016] ISSN: 2349-6495

Effective Ranking in Social Media Mekala.R1, Sindhu.S2, Suvathi.T3, Pavithra.V4 1

Assistant Professor, Department of CSE, K.S.Rangasamy College of Technology, Tiruchengode, Tamil Nadu, India 2, 3, 4 Students, Department of CSE, K.S.Rangasamy College of Technology, Tiruchengode, Tamil Nadu, India

Abstract—Data mining, the extraction of predictive information from large databases allowing businesses to make proactive decisions. Automated prediction of trends and behaviors improves the targeted marketing. Social media analytics help organizations to make the consumer at the center of the social marketing process. This mines the customer sentiment and opinion by effective tracking of information to support marketing. The analysis helps companies to connect directly with customers. The relationship between the groups of products and the items are identified by effective analysis method to improve the target business market. Apriori algorithm is used to mine frequent item set from the market basket databases and the number of lift are also recorded. Filtering through these rules makes effective transaction of data. Visualization rule is used to manage the large amount of data. The techniques used are aimed at uncovering the associations and interpretation of the data sets. The pattern of frequently purchased products analyzed to improve marketing strategies. Keywords— Apriori, Visualization, Social Media, Ranking. I. INTRODUCTION The number of customers shifting to online shopping is increased rapidly, the demand of online business deal is also getting very much increased. Nowadays everybody wants high-speed and direct home delivery without taking any efforts. Online shopping is one of the ways to buy the items easily and develop the business growth. It is based on reviews which represent excellence of the particular goods. Every time the customers are buying the product in online by viewing the reviews of the particular item. By viewing all these reviews before buying the product, gives an idea about the particular product which helps the customer to find best product among millions of the products and also know how other customers have recommended the product based on the quality. The market basket analysis provides the supplier to understand the purchase behavior of the consumer by discovering the products that were often purchased together, which helps to capture new buyers by www.ijaers.com

cross selling concept. This analysis used to divide consumers into groups. A company could look at what products are most frequently sold together. The analysis has significantly increased by the introduction of the electronic point of sale. The ranking of product increases the business development and popularity of the product. The different users have different needs and an objective for a product and it is essential that these differences are beneficial for the organization to provide advertisement for the particular product. II. EXISTING SYSTEM In the presented system many rules are generated and it becomes difficult to filter those association rules which lead to inefficient retrieval of data. The hereditary algorithm used leads to transaction problem when multiple transactions take place during purchase. No clear description and procedure for how to choose data. The numbers of rules are very large, trivial for analyzing and it is difficult to isolate interesting patterns. In the existing system there is an occurrence of Exponential computation complexity. The accuracy of gathering customer opinion and thoughts is difficult in the previous process. It does not give a clear view about the product for providing product advertisement. III. PROPOSED SYSTEM The purpose of market basket analysis using the apriori algorithm is to generate a set of rules that link products together. The apriori generate strong association rules from the frequent item sets. The output of the analysis reflects how frequently items are used in transactions and used to drive business decision-making. The algorithm manages the large result set of data. Candidate generation performed by algorithm generates large numbers of subsets until successful extensions are found. This algorithm uses the data structure used for storing the candidate sets, and counting their frequencies. The first step of Apriori is to count up the number of occurrences, called the support by scanning the database. Here it count the number of transactions in which each item occurs. Page | 34


International Journal of Advanced Engineering Research and Science (IJAERS)

Fig. 1: A scatter plot of the confidence and the support The next step is to generate a frequent item in pairs. In further step find frequent triples in the database and remove all the items that are bought less than minimum support. IV. MODULES DESCRIPTION 4.1. CUSTOMER REGISTRATION Registration process collects the basic information of person’s age, sexual category, work, income, hobbies that can be useful for highly developed advertising and marketing strategy. The idea of their income helps with pricing approach. The services are provided in order to satisfy customers. The behavior of the customer is also recorded to make key business decisions, after finding your target customer, the next step is to develop products to please them. Segmentation techniques can help marketers to optimize their targeting processes. The identification of similar data sets, understanding ding the similarities and differences within data are performed. The patterns obtained can then be seen as a kind of summary of the input data, and may be used in further analysis. The information from the customer is used by the businesses for direct marketing eting and customer relationship management. The marketing provides services to satisfy the customer by the prediction of customer behavior. In this the forecasting of buyers habits and life styles are analyzed. Through this customer analytics the business markets can make decisions with confidence. 4.2. PRODUCT CATEGORY Product Categorization is the process in which products are recognized, differentiated, and stated. It is systematic, disciplined approach to manage the product category to improve the business market. The groups of products are categorized usually for some specific purpose. A category illuminates a relationship between different items of knowledge. The categorization uses classical and conceptual approaches. The products are considered in large market segment which are generally consumed regularly www.ijaers.com

Vol-3, Vol Issue-2 , Feb- 2016] ISSN: 2349-6495

and purchased frequently. In this module first step is to formulate the abstract descriptions and then classifying c the entities according to that description. The task involves recognizing inherent structure in a data set and grouping products together by similarity presented in them. Product category consist of emergency products in which the customer seekss due to sudden events where there is no proper purchase planning. The unsought products are unplanned by the consumer but occurs as a result of marketers actions during special discounts offered to certain online shoppers. Customers profiling p determines and groups what kind of people buys what kind of products. It increases the growth rate of particular items and also increases the business strategy. 4.3. PRODUCT MANAGEMENT The role of product management is to maximize sales, market share, and profit fit percent. This deals with the planning, forecasting and production of the products. It provides product information for the companies. The product management also includes profit and loss. Product management involves both product addition and product elimination. The product manager is responsible for analyzing the market conditions. The benefits of the organization can be improved by product management on its own without dependence. During seasonal time the products are included and after that the products are removed. The product management consists of defining features of a product. To maximize the benefits, a Product management plays a vital role. The product for advertisement varies depends upon the stock level. This module includes product development and a product marketing. The product marketing includes product life cycle, product customers, partners, launching new products to market and monitoring the competition. The product development includes gathering the customers, defining product requirements and developing the products on schedule.

Page | 35


International Journal of Advanced Engineering Research and Science (IJAERS)

Vol-3, Vol Issue-2 , Feb- 2016] ISSN: 2349-6495

frequency of the items is recorded r and the count indicates how many lifts. V. CONCLUSION The breadth-first search and a Hash tree structure efficiently count the frequency set. The market basket analyses manage the placement of goods in their store layout. Segmentation and the association rules are generated to satisfy the customers. The algorithm is assumed to generate the candidate sets from the large item sets and count their frequencies of the databases. Apriori uses a bottom up approach which transforms specific data to less specific. Here the numbers of occurrences called support are calculated and when the minimum support meets or exceeds. The lift of a rule is the ratio of the support of the items and co-occurrence occurrence if the two are independent independent. By analyzing the data improves marketing by encourage them to spend more on their shopping basket. Fig. 2: Market basket analysis ysis for retails 4.4. TREND MONITORING AND ANALYTIC REPORT Trend monitoring module uses market basket analysis which suggests additional products to a customer for purchase. For example if we buy tea power and coffee power then the product sugar is automa automatically display by guessing the customers need. From the data set then information is recorded. The association rules are used to place the items next to each other so that the user buys more items.

[1]

[2]

[3]

[4]

[5]

[6]

Fig. 3: Steps involved in apriori algorithm [7] Now the apriori is the classic and basic algorithm to do it. This trend monitoring is used to detect and diagnose the customer opinion. From the recorded reports the apriori algorithm is applied to extract the required data. The www.ijaers.com

REFERENCE Aharon Bar-Hillel, Hillel, Daphna Weinshall and Tomer Hertz (2015), “Learning Distance Function for Image Retrival”, IEEE, vol.8, no.5, pp. 751-764. 751 Dr.N.Anbazhagan, N.Anbazhagan, A.Kannan, Dr.V.Mohan (2014), “An Effective Method of Image Retrival Using Image Mining Techniques”, The International Journal of Multimedia and its Applications (IJMA) Vol.2, No.4, pp.56-64. Andrea Passarella, R.I.M. Dunbara, Marco Contib and Valerio Amaboldia (2015), “Image Based Mechanical Analysis of Stent Deformation”, IEEE Social Networks, vol. 2, no, 2, pp. 39 39-47. Andrei Yakushev1 and Sergey Mityagin ITMO (2014), “System Design of a Super Super-Peer Network or Content-Based Based Image Retrieval”, IEEE vol. no. 6, pp.520-535. E.Annasaro and A.Hemal (2014), “A Survey in Need of Image Mining Techniques”, International Journal of Advanced Research in Computer and Communication Engineering, Vol. 2, no. 2, pp. 35-43. 35 Bo Pang and Lillian Lee (2014), “Emotional “Emot Profiling of Locations Based on Social Media”, Journal of Universal Computer Science, vol. 18, no. 8, pp. 973973 992. Brian Fisher, Nadya A. Calderon and Richard AriasArias Hernandez (2014), “A Design Study of Social Media Analytics in Emergency Management”, IEEE, Vol. 5, no. 2, pp. 224-233.

Page | 36


International Journal of Advanced Engineering Research and Science (IJAERS) [8] Ribeiro, M.X.; Bugatti, Jr.; Marques, P.M.A.; Rosa, P.H.; Traina, C., N.A.; Traina, A.J.M. Supporting content-based image retrieval and computer-aided diagnosis systems with association rule-based techniques. Data Knowl. Eng. 2011, 68, 1370–1382. [9] David Tsai, Michele Covell, James M. Rehg and Yushi Jing (2014), “Learning Query Specific Distance Functions for Large-Scale Web Image Search”, IEEE transactions on multimedia, vol. 15, no. 8, pp. 10211032. [10] Dewen Zhuang and Shoujue Wang (2013), “Contentbased image retrival based on integrating region segmentation and relevance feedback”, IEEE, vol. 3, no. 4, pp. 121-127. [11] Graeme Shanks and Bekmamedova (2014), “Social Media Analytics and Business Value”, IEEE, vol. 4, no. 3, pp. 178-186. [12] Mennis, J. Guo, D. Spatial data mining and geographic knowledge discovery—An introduction. Comput. Environ. Urban Syst. (2013), 33, 403–408. [13] Horia-Nicolai Teodorescu (2015), “Using analytics and social media for monitoring and mitigation of social disasters”, Horia-Nicolai Teodorescu / Procedia Engineering IEEE, vol. 10, no.3, pp.325-334. [14] Liviu Gabriel Cabaua, Mihaela Munteana, Vlad Rinciog (2013), “Using Image Mining to Discover Association Rules Between Image Objects”, Social and Behavioural Sciences IEEE, Vol. 6, no. 2, pp. 562-567. [15] Navya Nandakumar, Poulose Jacob K, Ramakrishnan K and Vimina E.R (2015), “An Efficient Multi Query System for Content-Based Image Retrival Using Query Replacement”, IEEE, vol. 5, no. 3, pp. 424-432. [16] Nesar Ahmad, Rashid Ali 1 and S.M.Zakariyal (2014), “Combining Visual Features of an Image at Different Precision Value of Unsupervised Content Based Image Retrival”, IEEE, vol. 4, no. 3, pp. 310-318. [17] Pattarasinee Bhattarakosol, Tippakorn Rungkasiri and Wilas Chamlertwat (2014), “Recommending Venues Using Continious Predictive Social Media Analytics”, Vol. 2, No 1-2, 1-135. [18] Klaric, M.; Scott, G.; Shyu, C.-R. Multi-index multiobject content-based retrieval. IEEE Trans. Geosci. Remote Sens. 2012, 50, 4036–4049. [19] Datcu, M. Seidel, K. Image Information Mining: Exploration of Image Content in Large Archives. In Proceedings of 2012 IEEE Aerospace Conference, Big Sky, MT, USA, 18–25 Volume 3, pp. 253–264.

www.ijaers.com

Vol-3, Issue-2 , Feb- 2016] ISSN: 2349-6495

[20] Liu, Y.; Zhang, D.; Lu, G.; Ma, W.-Y. A survey of content-based image retrieval with high-level semantics. Pattern Recogn. 2010, 40, 262–282.

Page | 37


Turn static files into dynamic content formats.

Create a flipbook
Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.