Online Book Recommendation System Using Collaborative Filtering
Objective
The Book Recommendation System aims to provide the best suggestion to the user by analyzing the buyer’s interest. The quality and the content are taken into consideration by employing content filtering, association rule mining and collaborative filtering.
Project Overview
The booming technology of the modern world has given rise to the enormous book websites. This makers the buyers to choose the best books to read as books play a vital role in many people’s life. The various kinds of books come into existence on day to day basis. So in order to eliminate this critical situation the recommendation system has been introduced in which the suggestion on the various books can be provided based on the analysis of the buyer’s interest. The Book Recommendation System is an intelligent algorithm which reduces the overhead of the people. This provides benefit to both the seller and the consumer creating the win-win situation. The E-commerce site to network security, all demands the need for the recommended system to increase their revenue rate. The content filtering, association rule mining and collaborative filtering are the various decision making techniques employed in the recommendation system as it helps buyers by the strong recommendations as there are various books, buyer’s sometimes cannot find the item they search for. The Book Recommendation System is widely implemented using search engines comprising of data sets.
Proposed System
The online book recommendation system involves various techniques for providing effective suggestion for the buyers. The association mining, collaborative filtering and content filtering are the three widely employed methods for strong impact using search engines. The content based filtering system is one in which the recommendation to the buyers are provided based on the items they have searched for. The items are generally in the form of text, comprising e-mail and web pages. This method analyse the similarities between the items to bring out the best recommendation.
The collaborative filtering involves the analysis of the opinions in which the recommendation is provided based on the ratings provided by the users. The quality of the item cannot be analysed in the content based filtering. But the collaborative filtering can expose the quality of the item. The collaborative filtering is employed in two ways namely, the user based collaborative filtering and item based collaborative filtering.
The next process to be performed is association rule mining in which association and correlation relationship is mined for the best outcome. The exemplar for the association rule mining would be market basket in which the set of data are analysed to obtain the buying pattern of the user. The two measures namely, support and confidence is used for analysis.
Book Recommendation System Modules
The modules include
- User module
- Admin module
The dataset contains the information of the user and the book which are provided as input. The output obtained is recommendation. The user module involves activities such as searching of books, entering text into web pages and so on. The admin module analyse the pattern by the above methods to provide the optimal suggestion to the user. For this analysis the user should create an account by using social media like Facebook such that the analysis like the recently searched books, books read can be taken into account for suggestion.
Software Requirements
- Windows OS
- My SQL 5.6
Hardware Requirements
- Hard Disk – 1 TB or Above
- RAM required – 8 GB or Above
- Processor – Core i3 or Above
Technology Used
- Recommender System
- Data Mining
- Sentiment Analysis
