International Journal of Engineering Science and Computing, April 2017 10942 http://ijesc.org/ ISSN XXXX XXXX © 2017 IJESC Hybrid Book Recommendation System Yash Wani 1 , Darsh Bakshi 2 , Vinesh Desai 3 , Sheetal Pereira 4 UG Student 1, 2, 3 , Assistant Professor 4 Department of Computer Engineering K.J Somaiya College of Engineering, Vidyavihar, Mumbai, India Abstract: Internet has been overwhelmed with mass amount of information. In order to provide necessary and accurate information to the user filtering mechanism were used. Recommendation systems are filtering systems which provide personalization in terms of the information coming to a user based on the similarities, interests, relevance of the information etc. Recommender systems are used widely for recommending movies, articles, restaurants, places to visit, items to buy etc. Its main goal is the management of the information overload. We propose a Book Recommendation system which integrates content based and collaborative filtering approaches. In Content based approach system uses keywords in order to find similar books, while in Collaborative filtering data is distributed evenly across all peers. In this system we attempt to present to the user various books the user is interested in filtering various redundant books using above two methodologies. Keywords: Collaborative filtering, content based filtering, Apriori algorithm, Karl Pearson algorithm I. INTRODUCTION With the rapid popularization of the web services, Internet has begun overflowing with information. Among the many web resources, users do not know how to quickly find the resources they really need. Therefore, a user specific recommendation system becomes an urgent demand product. Related research has become an important topic in this field of research and has gained wide attention of many research pioneers. Recommendation system [1] gets the reader’s interest hobby by online analyzing, recommendations. Collaborative filtering [1] [2] is one of the most important technologies working towards recommendations. eg grouplens, its can automatically provide their users with information which comes from people with similar interests. Users make evaluation to books and the system also provides users with a list of books that comes from other users. It is different from recommended system that needs to show the user’s evaluation information which comes from people with similar interests. Content-based filtering [2] also referred to as cognitive filtering, recommends items based on a comparison between the content of the items and a user profile. The content of each item is represented as a set of descriptors or terms, typically the words that occur in a document. The user profile is represented with the same terms and built up by analyzing the content of items which have been seen by the user. II.LITERATURE SURVEY According to survey, techniques used for recommendation are classified on the basis of knowledge sources. For instance, Collaborative Filtering technique works on user-item preference data, Content-based technique [2] is based on item features. Collaborative Filtering [1] [2] technique filters out the recommendation with the help of user behavior in the form of ratings. As mentioned in the paper we are going to implement collaborative filtering with the help of finding Pearson correlation ratio and applying nearest neighbor algorithm. This technique generates rating for an unrated item for the user, based on the commonalities among users and their ratings. Recommendation quality is directly proportional to the size of rating dataset. Content-based technique uses item features and user preference to provide recommendations. In this technique, attributes like genres, authors etc. are used to describe items, while user rating indicates the items liked by the user. It recommends items that are similar to the items preferred by the user. As mentioned in the papers the attribute matrix formed can be simplified by using matrix factorization by using singular value decomposition. Hybrid systems: Combines two or more recommendation techniques to predict recommendation. Using Hybrid technique, it is possible to overcome the drawbacks set by one recommendation technique and sum up advantages of different recommendation techniques. For example, Collaborative Filtering technique have problem when limited user-item ratings are available whereas Demographic and Content-based technique do not use rating data and therefore can overcome cold start problem. [1] There are various ways to combine recommendation techniques to achieve effective hybrid recommendation. III.PROPOSED SYSTEM We propose to build a website for book recommendation using Content Based and Collaborative Filtering technique. We will be making an attribute matrix for content based filtering and apriori algorithm [3] for Collaborative filtering. We propose to build a website for book recommendation using Content Based and Collaborative Filtering technique. [2] User Input: While making an account the user will be asked to give his/her details like name, age etc. Also the user can rate books that they like. Output: According to the data and the system the user will get book recommendations which will be personal. Research Article Volume 7 Issue No.4