Academic Journal of Computing & Information Science, 2021, 4(1); doi: 10.25236/AJCIS.2021.040115.
Li Zhitao, Zhou Xiaokang
Department of Software Engineering, Nanjing University of Finance and Economics, Nanjing, Jiangsu, China
[purposes] The recommendation effect of the user-based collaborative filtering algorithm is improved by extracting the effective dense sub-sequence between the user and the project and generating the data information of the project that the user participates in with high probability, The problem of data sparsity is alleviated to some extent, and the validity of binary data method in expressing user interest is verified. [methods]There are two steps to extract the effective dense subsequences from the user rate sequence, Firstly, the effective dense subsequence is extracted based on the time range; secondly, the subsequence is extracted twice by judging the items in the subsequence with high probability that do not conform to the user's interest. Generating data for projects with high probability of user participation requires the use of project attributes of user participation projects. The validity of the binary data method is mainly verified by the comparison experiment with the scoring data and the final result of the recommendation algorithm. [results] The experimental results show that RMSE (root mean square error) is reduced by more than 0.04 on the ml-latest-small data set compared with the classical traditional collaborative filtering algorithm. When F value (the harmonic mean of accuracy and recall rate) is taken as the evaluation standard, the algorithm proposed in this paper is more accurate than other classical algorithms, and the recommendation accuracy has been significantly improved. [conclusions] The method of extracting the effective dense subsequence between the user and the project is better than the traditional collaborative filtering method in terms of recommendation effect, and using binary data in the algorithm to replace the original rate data can bring better recommendation effect.
Dense Subsequence, Binary Data, Generate The Data, Mitigation of Sparsity, Recommendation System Algorithm
Li Zhitao, Zhou Xiaokang. A Recommendation Algorithm Based on Efficient Dense Subsequence and Data Generation. Academic Journal of Computing & Information Science (2021), Vol. 4, Issue 1: 94-104. https://doi.org/10.25236/AJCIS.2021.040115.
 Marken A. (2006) The Long Tail Why the Future of Business is Selling Less of More[J]. Hyperion, 24(3):274-276.
 Zhang S, Yao L, Sun A X, (2018) Tay Y.Deep learning based recommender system[J]. CM Computing Surveys, 1(1):1-35.
 Hong L, Ren Q H, Liang S X. (2016) Comparative Study on Information Service Quality of Recommendation System of Domestic E-commerce Websites -- A case study of Taobao,JINGdong and Amazon[J]. Library and Information Work, 60(23):97-110.
 Ma X T. (2016) Research on library Personalized Service Recommendation System based on Situational Big data[J]. Modern Intelligence, 36(4):90-94.
 Amatriain X, Pujol J M, Tintarev N. (2009) Rate it again:increasing recommendation accuracy by user re-rating[C]. ACM Conference on Recommender Systems. ACM, pp. 173-180.
 Young O K. (2008) Improving top-N recommendation techniques using rating variance[C]. In Proceedings of the Conference on Recommender Systems (RecSys’08), ACM Press, New York. NY. pp. 307–310.
 Jugovac M, Jannach D. (2017) Interacting with Recommenders—Overview and Research Directions[J]. ACM Transactions on Interactive Intelligent Systems, 7(3):1-46.
 Xing C X, Gao F R, Zhan S N, Zhou L Z. (2007) A collaborative filtering recommendation algorithm that ADAPTS to user interest changes[J]. Computer Research and Development, pp. 296-301.
 Liu K, Shi X, Kumar A. (2016) Temporal learning and sequence modeling for a job recommender system[C]. Proceedings of the Recommender Systems Challenge, ACM, pp. 1-4.
 Ding S H, Ji D H, Wang L L. (2015) Collaborative filtering recommendation algorithm based on user attributes and ratings[J]. Computer Engineering and Design, 36(2):487-491+497.
 Mao Y Y, Liu J H, Hu R, Tang M D, Shi M. (2017) A collaborative filtering recommendation algorithm for Web services using Sigmoid function[J]. Computer Science and Exploration, 11(2):314-322.
 Qian R, Wu Y, Kong G Q. (2018) Research on collaborative filtering algorithm based on fusion sparsity weighting[J]. Computer Technology and Development, 28(7):21-24+42.
 Yu H, Li Z Y. (2010) Collaborative filtering recommendation algorithm based on forgetting curve[J]. Journal of Nanjing University (Natural Science Edition), 46(5):520-527.
 Jia W Y, Li S Q, Li X Y, Liu B. (2018)A collaborative filtering recommendation algorithm based on discrete variables and user interest proximity[J]. Computer Engineering, 44(1):226-232+237.
 Benjamin M M, Richard S Z. (2009) Collaborative prediction and ranking with non-random missing data[C]. In Proceedings of the 3rd ACM Conference on Recommender Systems (RecSys’09). ACM Press, New York. NY. pp. 5-12.
 Koren Y. (2010) Factor in the Neighbors: Scalable and Accurate Collaborative Filtering[J]. Acm Transactions on Knowledge Discovery from Data, 4(1):1-24.
 Guo H Y, Liu G S, Su B, Meng K. (2016)Collaborative filtering recommendation algorithm combining community structure and interest clustering[J]. Computer research and development, 53(8):1664-1672.
 Liu Q H. (2016) A hybrid recommendation study based on community of interest and trusted neighbors[J]. Information science, 34(2):65-69.
 Yu B, Yang H L, Len M. (2018) Recommendation algorithm based on user interest model[J]. Computer System Application, 27(9):182-187.
 He M, Xiao R, Liu W S, Sun W. (2017) Collaborative filtering recommendation algorithm combining category information and user interest[J]. Computer Science, 44(8):230-235+269.
 Zhao W T, Wang C C, Cheng Y F, Meng L J, Zhao H H. (2016) Collaborative filtering algorithm based on user multiple attributes and interests[J]. Computer Application Research, 33(12):3630-3633+3653.
 Dong L Y, Wang Y Q, He J N, Sun M H, Li Y L. (2017) Collaborative filtering recommendation algorithm based on time attenuation[J]. Journal of Jilin University (Engineering Science), 47(4):1268-1272.
 Ren L. (2015)A collaborative recommendation algorithm combined with scoring time characteristics[J]. Computer applications and software, 32(5):112-115.
 Wu Y T, Zhang X M, Wang X M, Li H. (2016)Collaborative filtering algorithm based on user fuzzy similarity[J]. Journal of communication, 37(1):198-206.
 Liu J, Wu W Q, Li X, Liu Y L, (2017) Wang J F.Collaborative filtering algorithm based on user interests and project attributes[J]. Computer Applications and Software, 34(5):33-37.