Toward personalize user-based recommendation system for big data application

Abstract

The use of web are increases the option of user choices

also increases. Recommendation system is guideline for users in

personalized way to choose/choice their option from the large

space datasets. Large volume dataset is refers to “Big Data”, Big

data is nothing but the data which is beyond the capacity of

storing, managing and processing within a short time period. In

this paper purposes the personalized user-based

recommendation system in which the movies recommendation

list is generated as per user interest. In previous service

recommendation system collaborative filtering algorithm is

adopted but they are faces problem with scalability and

inefficiency at the time of data retrieval. The existing service

recommendation systems are fails to meet user requirements

because without considering users preferences/interest’s it

display same ratings and rankings to different users. Also in

traditional recommendation system yielding the big data

discovery and analysis problem. In purposed system ratings or

features are used to filtering the information by applying multi

criteria selection policy. Basically to manage and solve scalability

and efficiency problem Hadoop is broadly adopted distributed

computing platform with MapReduce parallel processing

environment. Finally, Experiment is conducted on real-world

data set and results demonstrate the accuracy, efficiency and

scalability to improve recommendations