A Semi-Supervised User Group Identification Based on Synergetic Neural Network and Information Entropy

• Main Authors: Zhehuang HUANG, Yidong CHEN
• Format: Article
• Language: English
• Published: IFSA Publishing, S.L. 2013-10-01
• Series: Sensors & Transducers
• Subjects: User group identification; SNN; Information entropy; PSO; Semi-supervised learning.
• Online Access: http://www.sensorsportal.com/HTML/DIGEST/october_2013/P_1376.pdf

User group identification is an important task in intelligent personalized information service. A key problem of intelligence user server model is how to classify and indentify the user groups. Only when the user group can be effectively identified, the desired service can be offered. At present, it is difficult to obtain a large number of labeled corpuses which takes a certain amount of human and material resources. How to improve the comprehensive utilization of a small amount of labeled sample and a large number of unlabeled samples is an important task. To solve the problem, we propose novel semi-supervised user group identification based on SNN and information entropy in this paper. This paper has two main works. Firstly, a user group identification using synergetic neural network (SNN) is presented, which can effectively identify user groups; Secondly, we propose a noise filter based on information entropy to reduce the noise of expand data. The experiment results show the proposed model in this paper has a higher performance for user group identification, and provide a good practicability and a promising future for other tasks.