International Journal of New Developments in Engineering and Society, 2020, 4(1); doi: 10.25236/IJNDES.040109.
Huimin Yang1, Jin Liu1,*
1 School of Computer & Software, Nanjing University of Information Science &Technology, Jiangsu, Nanjing, 210044, China
*Corresponding Author
Aiming at the unreasonable resource allocation problem of traditional mass diffusion and heat conduction algorithms, this paper proposes a dynamic resource allocation algorithm based on popularity. Taking into account the different degree of influence of the popularity of items in different periods, this paper proposes the concept of item weighting to improve the traditional algorithm, and design the non-equilibrium mass diffusion and heat conduction algorithm. The experimental results prove that compared with the original mass diffusion algorithm and heat conduction algorithm and some improved algorithms based on mass diffusion and heat conduction, the algorithm proposed in this paper can significantly improve the performance of the Recommender system.
Recommender systems, Heat conduction, popularity, Accuracy
Huimin Yang, Jin Liu. Dynamic resource allocation recommendation algorithm based on popularity. International Journal of New Developments in Engineering and Society (2020) Vol.4, Issue 1: 61-69. https://doi.org/10.25236/IJNDES.040109.
[1] Maes, Pattie(1995). Agents the Reduce Work and Information Overload. Communications of the ACM, p.811-821.
[2] Song W J , Guo Q , Liu J G(2014). Improved hybrid information filtering based on limited time window. Physica A Statal Mechanics & Its Applications, vol.416, p.192-197.
[3] Ding Y, Li X(2015). Time weight collaborative filtering. Proceedings of the 14th ACM international conference on Information and knowledge management.p.485-492.
[4] Zhou Y, Zeng A, Wang W H(2015). Temporal effects in trend prediction: identifying the most popular nodes in the future. PloS one, vol.10, no.3, e0120735.
[5] Li W J, Dong Q, Shi Y B, et al(2017). Effect of recent popularity on heat-conduction based recommendation models. Physica A: Statistical Mechanics and its Applications, vol.474, p.334-343.
[6] Song W J, Guo Q, Liu J G(2014). Improved hybrid information filtering based on limited time window. Physica A: Statistical Mechanics and its Applications, vol.416, p.192-197.
[7] Liu J, Li J, Chen Y, et al(2019). Modeling complex networks with accelerating growth and aging effect[J]. Physics Letters A, vol.383, no.13, p.1396-1400.
[8] Medo M, Cimini G, Gualdi S(2011). Temporal effects in the growth of networks. Physical review letters, vol.107, no.23, e238701.
[9] Wang Y, Han L(2020). Personalized recommendation via network-based inference with time. Physica A: Statistical Mechanics and its Applications, vol.550, e123917.
[10] Li W J, Dong Q, Shi Y B, et al(2017). Effect of recent popularity on heat-conduction based recommendation models. Physica A: Statistical Mechanics and its Applications, vol.474, p.334-343.