Paper Details

Asper today’s generation the demand of internet and social networking service, the micro-videos arebecoming more popular. The users spend their lot of time to search their favourite micro-videos from the amount ofvideos on the web, and also the producers of the micro-videos. Therefore, we have proposed a micro-videorecommendation system. The recommendation algorithm are the important part of this system. The slope one schemealgorithm is based on rating of the users, and it is simple, efficient, easy to implement. However, the slope one schemesuffers from both new item problems and data sparsity, which affect the performance of recommender systems. Toovercome the drawbacks of slope one scheme, we proposed an improved slope one algorithm which containcollaborative filtering algorithm. The slope one scheme is improved by introducing content similarity computation toovercome the new item problem. In item-based collaborative filtering algorithms, the target users rating to the targetitem can be predicted based on the ratings that the target user has rated and the content similarities of items. Andclustering algorithm is used to overcome the problem of data sparsity. By combining the set of items into several differentclusters based on the item rating data, the target users rating to the target users rating to the target item can be predictedbased on which cluster the target item belongs to. The rating of target user to the target item is the linear combination ofthe above algorithms used.

Prof. M. P. Nerkar, Shraddha Bhurke, Nikita Kadam, Prajkta kavitake, Komal Handore, “A Micro-video recommendation system using improved slope one algorithm based on Big Data”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 12, pp. -, December 2017.