Interest point recommendation method based on LBSN and multi-graph fusion

Abstract

The invention discloses an interest point recommendation method based on LBSN and multi-graph fusion. The method comprises the steps: user and interest point internal feature modeling: dividing a user-interest point scoring matrix into a product of a user matrix and an interest point matrix through a matrix decomposition algorithm, and enabling the product to serve as an internal potential vector of a user and an interest point; and user and the interest point external feature modeling: learning feature vectors of the user in an interest point space and a social space and feature vectors of the interest point in a user space and a position space through multi-graph fusion and an improved k-means clustering algorithm, and further obtaining external representation vectors of the user and the interest point; and inputting the final vectors of the user and the interest points into a neural network for learning, and recommending the first k interest points with the highest scores to the user according to the scores. According to the method, the user-interest point interaction graph and the user social relation graph are learned in a multi-graph fusion mode, and a new thought is provided for interest point recommendation; and recommendation errors can be effectively reduced, and the accuracy of recommendation results is improved.

Description

technical field [0001] The invention belongs to the technical field of natural language processing and geographic information, and in particular relates to an interest point recommendation method based on LBSN and multi-image fusion. Background technique [0002] The development of geographic information systems and mobile networks has promoted the rapid development of location-aware social media. With the increasing number of spatial Web objects (also known as POIs), POI recommendation, as one of the important services of location-based social networks, has gradually become an important part of the current Web query, natural language processing and location-based social network (LBSN) A hot topic of analysis. Collaborative filtering is the earliest point-of-interest recommendation method. By learning the potential characteristics of users and points of interest, users and points of interest are represented in the form of vectors, and then the user's preference for points o...

AI Technical Summary

Problems solved by technology

However, such methods cannot capture the collaborative information in user-POI interaction records

Images