C. Sathya Charanya
Recommendation system acts as a tool in providing most appropriate service to the user. Currently, information through online services increases. This leads to the overhead of data in online and there is a possibility of getting less accurate results. In previous approaches, recommendation of service is based on the feedbacks and ranking from the previous user. It doesn’t consider the suggestion of the user at a time, who in need of searching for the particular service. The proposed system deals with the implementation of personalized recommendation to provide services for hotel reservation system. Preferences are collected from the active user about particular service for each application. Similar user’s opinions are taken from the reviews using keyword extraction method and Supervised learning algorithms are used to identify sentiment orientation. It determines positive or negative opinion along with negation word near to each opinion word and then identifies the number of positive and negative opinions of reviews. Keywords with positive opinion are considered and similarity is calculated between user preferences with reviews of the previous user by jaccord and cosine measures. From this most similar keywords are provided to the user as recommended service. To provide more accurate prediction of the services needed by the active user the proposed system is implemented using MapReduce framework.
Preferences, Recommender system, Hadoop, MapReduce