Knowledge graph-based industry recommendation method and system

Abstract

The invention provides a knowledge graph-based industry recommendation method and system, and relates to the technical field of knowledge graphs, the method comprises the following steps: a data reporting step: collecting and uploading a data source, the data source comprising user portrait data, user real-time behavior data and real-time article data; a content risk control step: filtering information data in the data source; a user feature step: extracting user features from the filtered data source; a text feature step: carrying out text feature extraction on all information articles in the data source; a multi-path recall step: according to the text data features, performing personalized recall on the to-be-selected information to form a recall set; and a recall layer sorting step: enabling the system to use a financial knowledge graph and a deep learning algorithm to sort the recall set, and select the most interested content of the user. The user stickiness can be effectively improved, the improvement of various business indexes is promoted, and the user experience is improved.

Description

technical field [0001] The present invention relates to the technical field of knowledge graphs, in particular to an industry recommendation method and system based on knowledge graphs. Background technique [0002] In order to promote industrial upgrading, innovation drives development. Various regions and departments have launched policies to promote industrial development, such as industrial development funds. As an individual in industrial development, how to quickly and accurately enjoy policy services has always been a difficult problem for the enterprise service department. Policy matching matches relevant industrial policy items based on basic information such as the company's registration place, registered capital, business scope, and industry. The intelligent matching based on the knowledge map realizes the intelligent matching of enterprises from finding policy projects to policy projects, improving the accuracy of service enterprises in the enterprise service d...

AI Technical Summary

Problems solved by technology

[0004] In the general policy recommendation system, enterprises need to check and understand projects one by one when they need to choose projects that match themselves, which reduces the willingness of enterprises to declare industrial funds, and also prevents industrial funds from fully playing their due role. However, the existing The calculation reliability of the recommendation results is relatively low, the interpretability of the recommendation results is insufficient, it is difficult to convince users, it is impossible to accurately explore user preferences, and it is impossible to ensure that each recommendation is in line with the user's favorite effect, and the user experience is low

Images