A Hybrid Recommendation System Using Incremental Learning Algorithm
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
To overcome research problems in recommendation system, all basis models obtain separately capability. However with different source data, they become ineffective respectively. Moreover in the big data era, the rapid data process are also highlighted as necessary requirement beside the accuracy. The existing researches mainly focus on single model and only a few research considers to the hybrid system to combine various robustness of them. In this research, we design a hybrid system by combining the feature of content base and the robustness of matrix factorization in collaboration filtering using the stochastic gradient descent algorithm. With the results in experiments, we show the better results of model in both accuracy and the time learning of model. This study also direct a novel trend in recommendation system research.
Abstract
To overcome research problems in recommendation system, all basis models obtain separately capability. However with different source data, they become ineffective respectively. Moreover in the big data era, the rapid data process are also highlighted as necessary requirement beside the accuracy. The existing researches mainly focus on single model and only a few research considers to the hybrid system to combine various robustness of them. In this research, we design a hybrid system by combining the feature of content base and the robustness of matrix factorization in collaboration filtering using the stochastic gradient descent algorithm. With the results in experiments, we show the better results of model in both accuracy and the time learning of model. This study also direct a novel trend in recommendation system research.
참고문헌 (7)
- Çano, E., "Hybrid recommender syste ms: A systematic literature review", Intelligent Data Analysis, 21(6), pp.1487~1524 (2017)
- Battiti, R., "Accelerated backpropagation learning:Two optimization methods", Complex systems, 3(4), pp.331~342 (1989)
- Needell, D., "Advances in neural informa tion processing systems", pp.1017~1025 (2014)
- Gemulla, R., "Largescalematrixfactorizationwithdistributedstochasticgr adientdescent", Proceedings of the 17th ACM SI GKDD international conference on Knowledge disc overy and data mining, pp.69~77 (2011)
- Ruder, S., "An overview of gradient descent optimiza tion algorithms", arXiv preprint arXiv:1609.04747, ~ (2016)
- Duchêne, S., "Cross-validation to select Bayesian hierarchic al models in phylogenetics", BMC evolutionary biology, 16(1), pp.115~ (2016)
- FU, Wei, "Estimating the number of clusters using cross-validation", Journal of Comp utational and Graphical Statistics, 29(1), pp.162~173 (2020)