Intelligent recommendation method based on user questions and answers

Abstract

The invention relates to an intelligent recommendation method based on user questions and answers. The method comprises the following steps: step one, obtaining questions and answers of a user through a user terminal by providing an input box; step two, carrying analyzing by a back-end service system on data of a current user through NLP word segmentation and a data burying point interface; step three, judging whether the current word segmentation part-of-speech, data tags, and knowledge base content are matched or not; if yes, obtaining recommendation content from a recommendation resource pool from a matched resource pool according to algorithm capability of big data, and returning the recommendation content to the user terminal in cooperation with the knowledge base content; if not, directly returning a question and answer knowledge base of the user to the user terminal. The method has the beneficial effect that the recommendation is relatively accurate.

Description

technical field [0001] The invention relates to an intelligent recommendation method based on user questions and answers. Background technique [0002] In the current travel scene, the user intelligent recommendation of the online system is mostly based on the user's behavior data and preference tag data as the important data model of the user recommendation system. In the business system, the user's behavior data is collected, stored, and analyzed through embedded technology, combined with the actual business of the online system, and through the existing capabilities of big data, the online recommendation of data content that is helpful to the user allows the user to perceive the system intelligence of information. [0003] However, this method still has shortcomings: on the one hand, intelligent recommendation is completely dependent on the user's online behavior, including clicks, browsing and other events. When the big data field needs to store a large enough amount of...

AI Technical Summary

Problems solved by technology

[0003] However, this method still has shortcomings: on the one hand, intelligent recommendation is completely dependent on the user's online behavior, including clicks, browsing and other events. When the big data field needs to store a large enough amount of data, the underlying data algorithm can be accurate based on user behavior data. recommend

On the other hand, the real-time and accuracy of this type of intelligent recommendation cannot be applied to every public user. Different users need to generate recommendations based on their own behavior data. The underlying algorithm model is not universal to a certain extent.

Images