International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS) Volume V, Issue VII, July 2016 | ISSN 2278-2540 www.ijltemas.in Page 25 Recommendation System Based on Content Filtering for Specific Commodity Niranjan C Kundur 1 , Praveen M Dhulavvagol 2 , Prasad M R 3 1 Dept. of CSE, JSSATE, VTU, Belagavi. 2 Dept. of ISE, BVBCET, 3 Dept. of CSE, JSSATE, VTU, Belagavi. Abstract— Internet Content-based recommendation systems may be used in a variety of domains ranging from recommending web sites, news items, restaurants, television programs, and commodities for sale. Content-based recommendation systems share a common means for describing the items that may be recommended. In this paper, we propose Recommendation System that uses Keywords as input query from user for extracting specific items that match user query from the list. User keywords may consists of keywords words from name of the item, brand and popularity. Here we are calculating the similarity between user given item names and collected item name in the database by using vector space model which in turn uses TF-IDF, Cosine Similarity and finally re-rank top recommended items. We measured satisfaction and accuracy for each system-recommended item to test and evaluated performances of the suggested system. Finally Recommendation System for item based represents high level of satisfaction and accuracy. Keywords: item based, recommendation, vector space model, hash-map I. INTRODUCTION ecommender systems have become extremely common in recent years, and are applied in a variety of applications. The most popular ones are probably movies, music, news, books, research articles, search queries, social tags, and products in general. However, there are also recommender systems for experts, jokes, restaurants, financial services, life insurance, persons (online dating), and Twitter followers. Recommender systems are software tools and techniques providing suggestions for items to be of use to a user. The term item here is generic. It may represent many concepts. For instance recommender systems may recommend news on a news portal, or products in an online shop, or even services. The recommendations are usually tailored to a given type of user or a given type of user group. Since recommendations are personalized, they may vary from one user to another or from one user group to another. Due to the development of technology of Internet, Web Programming and Web environment in recent years, the huge amount of data extremely increases in the Web[6], then following new exceed information overload problems occur. So, newly high technology search engines are developed and made to solve these problems and to provide user-wanted information quickly and accurately. Content-based recommendation systems analyse item descriptions to identify items that are of particular interest to the user. Recommender system is an active research area in the data mining and machine learning areas. [1] II. RELATED WORKS Collaborative Filtering Technique: Collaborative filtering systems work by collecting user feedback in the form of ratings for items in a given domain and exploit similarities and differences among profiles of several users in determining how to recommend an item. The limitations of using collaborative filtering are i. Most users do not rate most items and hence the user-item rating matrix is typically very sparse. Therefore the probability of finding a set of users with significantly similar ratings is usually low[2]. This is often the case when systems have a very high item-to-user ratio. This problem is also very significant when the system is in the initial stage of use[4]. ii. First-rater Problem: An item cannot be recommended unless a user has rated it before. This problem applies to new items and also obscure items and is particularly detrimental to users with eclectic tastes. Click-Through Rate: It is used in recommendation of papers, it’s mainly online. The click-through rate is the number of times a click is made on the advertisement divided by the total impressions (the number of times an advertisement was served). The limitations of click-through rate are: i. It does not help you with conversions. A high CTR might actually have a low conversion rate (and often does). Some Internet users just have a higher propensity to click, which does not actually mean that they want to buy anything. Usually, these people can be found in higher proportion at less popular sites (which is probably how they got there in the first place)[5]. R