Water conservancy portal information hybrid recommendation method

Abstract

The invention discloses a water conservancy portal information hybrid recommendation method. The method comprises: defining a water conservancy portal information classification system in advance, then crawling water conservancy portal information from all large water conservancy portal websites through a web crawler, conducting manual labeling according to water conservancy portal information classification, establishing a water conservancy portal information classifier, and automatically classifying a large amount of water conservancy portal information in a recommendation data set; respectively constructing a hot spot information recommendation model based on space-time sensitivity, a recommendation model based on information content and a collaborative filtering recommendation model based on users and information categories by utilizing the recommendation data set and the user behavior data; and linearly mixing the three models into old user recommendation information according toa certain proportion, and recommending information to a new user by using a hot spot information recommendation model based on time-space sensitivity. The accuracy and specialty of water conservancy information recommendation are improved, and innovative application of the recommendation technology to the water conservancy portal website is achieved.

Description

technical field [0001] The invention belongs to the technical field of information processing, and in particular relates to a mixed recommendation method for water conservancy portal information. Background technique [0002] The water conservancy portal displays the information on the website according to the chronological order of information release. However, this information display method has the following disadvantages: [0003] (1) The release form is single and lacks flexibility. The website provides the same information to all users according to the chronological order in which the information is published, without considering user preferences and lacking flexibility. [0004] (2) It is difficult for users to find information, and the user experience is poor. When users browse the water conservancy portal website, they usually look for portal information on the website according to their needs. However, with the continuous increase of information on water conserv...

AI Technical Summary

Problems solved by technology

[0010] Purpose of the invention: In order to overcome the disadvantages of the existing water conservancy portal website, such as the single form of information release, lack of attractiveness, difficulty in finding information for users, poor experience, low satisfaction, and easy for users to miss important water conservancy portal information, and to apply recommendation technology to water conservancy For the cold start problem, data sparsity and adaptability problems in the portal website, the present invention provides a mixed recommendation method for water conservancy portal information, which classifies the water conservancy portal information according to the water conservancy business and its own characteristics, and considers user interests on this basis Attenuation, water conservancy portal information timeliness, and changes in water conservancy business hotspots in different time and space, build a hybrid recommendation model, improve the accuracy and professionalism of water conservancy information recommendation, and realize the innovative application of recommendation technology on water conservancy portal websites

Images