A book review is a web based application that holds individual comments, critics and suggestion on books. Any viewer of the application can see all information related to any book searched by the viewer and also see the opinions of other users of the application.A user who is registered, when logs in the system, can do comment in any book and add any book to his/her favorite list. A user can also follow other registered users in this application. Depending on the comment and analyzing the comment, a rating for that book is generated by the system. For generating the rating for books, the technology used is called sentimental analysis which is the most important feature of this project. Two recommendation algorithms are applied in this system for suggesting books to users: non-personalized recommendation and personalized recommendation.
Personalized recommendation is based on user preference. The algorithm used for this type of recommendation is rating-based item-to-item recommender system based on Slope One algorithm. User to user recommendation is also used in the application. Co-efficient cosine similarity is used for user to user recommendation. A streaming system is included in the application to view the activities of a user by others who follow that user. A real time notification system is also implemented in this system. In the current time, the user activities followed by any user are shown in the notification. An advanced search is also implemented in this system for finding books based on different terms: book name, category of book, tags, ISBN or any keyword that remains in the description of any book. This application also shows the top rated books in all time and also the best rated books in the current month. A suggestion for books related to any search is also shown. The demo data inserted in this application are taken form Amazon book data set.