An Intelligent E-commerce Recommender System Based on Web Mining

Abstract

The prosperity of e-commerce has changed the whole outlook of traditional trading behavior. More and more people are willing to conduct Internet shopping. However, the massive product information provided by the Internet Merchants causes the problem of information overload and this will reduces the customer's satisfaction and interests. To overcome this problem, a recommender system based on web mining is proposed in this paper. The system utilizes web mining techniques to trace the customer's shopping behavior and learn his/her up-to-date preferences adaptively. The experiments have been conducted to evaluate its recommender quality and the results show that the system can give sensible recommendations, and is able to help customers save enormous time for Internet shopping.

http://www.jestr.org/downloads/Volume10Issue4/fulltext171042017.pdf

“…[65], [21], [66], [67], [68], [69] [7], [175], [106], [8], [176], [177], [178], [179], [180], [181], [182], [57], [116] [187] , [190], [191], [75], [192], [31], [76], [193], [194] 5…”

“…A detailed discussion about the development of a personalized product recommendation system based on customer's click streams is performed in [67]. The authors have proposed a recommender system based on web mining to overcome the problem of data overload so that satisfactory recommendation can be made for users.…”

Recommendation System: State of the Art Approach

“…Traditionally, users get a word of mouth recommendation for a product, and they ask for the recommended product at the shop. A similar analogy works for library and information services because people tend to buy books from bookstalls which are suggested to them by their faculty, friends and/or colleagues [5]. These days many online book stores are available which have a huge collection of material available with them in the form of ebooks, research papers and journals.…”

“…The designer that designs an application has a collection of algorithms, and must make a pertinent decision for selecting an algorithm to achieve his goals; the selection process is solely based on experiments, in which the performance of numerous users recommendations are collated [5]. For the given structural constraints the designer can then plump for the best performing algorithm.…”

A Survey on E-Marketing by Behavioral Analysis in Online Social Network

“…Certain other measures are also available like  Network Centralization which gives an insight of individual user's location in the network [13] and it will also provide the relationship among the centralities of all nodes. A less centralized network may be less prone to single point failure  Shortest path to reach a particular node is also one major factor which decides the faster reachability of a node.…”