User model construction method based on ontology

Abstract

The invention discloses an ontology-based user model construction method, which includes the following steps: (1) obtaining relevant information that can reflect user characteristics, preferences and needs; (2) constructing a tourism field ontology; (3) using a similarity algorithm to Classify users; (4) Construct an interest model for each user, and use ontology to represent the user model; (5) Update the user model by adjusting the degree of interest transmission. The present invention collects the basic information of users through web usage logs based on the ontology concept hierarchy tree, utilizes these information and the domain ontology database, uses ontology editing tools to establish domain ontology, classifies users through improved similarity algorithm, and establishes domain ontology for each type of user. User model. Finally, the user interest model is updated by combining the method of interest degree and transmission adjustment, which reduces the difficulty of model construction and has high credibility and accuracy for describing user interest.

Description

technical field [0001] The invention relates to a modeling method in the field of tourism, more specifically, to a user model construction method based on ontology. Background technique [0002] Tourism is a comprehensive industry that integrates transportation, sightseeing, accommodation, catering, shopping, entertainment and other links. With the rapid development of network technology, the tourism information that people can obtain on the Internet shows an explosive growth trend. While these network information resources bring people rich knowledge and great convenience, they will also cause personal harm. information fatigue and information stress. In order to quickly and accurately locate the required travel information and push information to people in a personalized way, research on user interest models has become a hot and difficult point at present. At present, some common user model construction methods have appeared: keyword-based, vector space model-based, neur...

AI Technical Summary

Problems solved by technology

Although domain knowledge is applied, the basic technology is still based on the description of keywords and the application of traditional machine learning algorithms, and most of the established user models are just a conceptual hierarchy. User models are static and cannot be followed. Changes in user interests are dynamically updated in a timely manner, and the stability of the user model is low

The user model cannot be improved in a timely manner with changes in user interests (including content changes, decay of old interests, and generation of new interests), so it cannot reflect the process of interest changes well

Images