Offering Recommendations on Netflix dataset by Associations among Users as Trust Metric

  • Rajeswari Nakka, Dr. G.V.S.N.R.V.Prasad and R.Kiran Kumar

Abstract

As the start of internet revolution, data is generated every day in large amounts that led to many opportunities in improvement and/or enabling wider areas of research and decision based systems providing unprecedented value towards real-world applications, business strategies, science, engineering, and developments etc. Gathering insights and making profits out of this large data is being the main objective of many service recommending companies like Netflix, Amazon, etc. One of the vital problem encountered in recommender systems is data sparsity. To address this problem, the proposed work incorporates users confidence in trust computation in addition to the similarity among users while offering the recommendations. Thus the addition of user confidence in trust calculation helps to identify the trust among users even though the users don’t share any similarity. This paper focuses on applying machine learning techniques to offer recommendations on Netflix dataset by implementing associations among users as a trust metric. As the machine learning process requires huge computations and consumes more time for processing large datasets, the services of cloud is used for computations and to perform the analysis. Here the experiments are performed in Google Colab platform which is a free cloud service that provides python based Jupyter notebook environment for computing and analysing the large datasets. As the Google Colab is a free cloud service it facilitates the required resources for analyzing the data, especially this kind of services are helpful in analyzing large quantities of data.

Published
2020-05-12
How to Cite
Rajeswari Nakka, Dr. G.V.S.N.R.V.Prasad and R.Kiran Kumar. (2020). Offering Recommendations on Netflix dataset by Associations among Users as Trust Metric. International Journal of Advanced Science and Technology, 29(7), 989 - 1000. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/14957
Section
Articles