Web News Media Retrieval Analysis Integrating with Knowledge Recognition of Semantic Grouping Vector Space Model

Abstract

Traditional Web news media retrieval technology can only meet the specific requirements of customers. Because of its universal characteristics, it cannot meet the needs of different environments, different purposes and different times simultaneously. Researchers have proposed a search method for online news media, which is used for computing the semantic grouping vector space model. The customer's interest model is analyzed through the characteristics of the user's different classification areas. In this paper, we propose a vector space model that performs semantic grouping based on feature words. The model divides four groups that are relatively independent in the meaning of feature words in a news report: time, place, person, and event, and then form four vector spaces, and calculate the weight value and similarity of each vector space. Theoretical analysis and experimental results show that the improved model is suitable for searching Web news information, and improves the calibration rate, calibration rate and query speed.