Information Entropy-Based Two-Stage Clustering Recommendation Algorithm

The accuracy of final recommendation results is always affected by less active evaluating information, which comes from users to the webpage text. The paper presents a two-stage clustering recommendation algorithm based on information entropy. By extracting the characteristic words and giving different weights, the paper calculates the information entropy value of each webpage text browsed by users and according to the threshold value of the nearest entropy difference which is an interval scale measurement, to measure whether the two pages with the nearest information entropy should be divided into different clusters. Moreover, the paper obtains the content of the recommendation results by two-stage clustering, and combine the continuous random variable of uniform distribution with the approaches of the average entropy value approximation and the logarithmic function fitting, etc. The experimental results imply that the new algorithm is stable during the real system operation and improves the accuracy of final recommendation results comparing to the traditional algorithms.