Improving the Performance of Wikipedia Based on the Entry Relationship between Articles

Improving the Performance of Wikipedia Based on the Entry Relationship between Articles

Wikipedia is the largest online encyclopedia in the world. It is free to access by anyone and its main advantage is that it can also be edited by any person at any time. On the one hand, this caused a rapid growth to its number of available articles and languages. It is likely to cause that most users are difficult to differentiate various synonymy and polysemy terms from the millions of articles in Wikipedia. On the other hand, traditional semantic analysis models are mainly focus on to deal with the semantic relationships between terms, or terms and documents. However, these models are lacking to deal with the semantic relationships between documents.

In this paper, to enhance the semantic relationships between documents, we use the entry relationship between any two Wikipedia articles to design our Latent Entry Analysis (LEA) model. The advantages of LEA have the following several aspects: (1) it can effectively deal with the problems of synonymy and polysemy; (2) it is a good model to find the semantic relationships between terms, terms and documents, or documents; (3) it is a good model with a high-performance and low-cost compared to other semantic analysis models; (4) it is a suitable model to effectively handle big data sets in Wikipedia.