A method for constructing multi-dimensional information portraits of travel agency users

Abstract

The invention discloses a method for constructing multi-dimensional information portraits of travel agency users, comprising the following steps: obtaining feature vectors and user portrait information of sample users; assigning weights to each label of sample users; setting the number of clusters k and Similarity threshold θ; according to the label vector of the sample user, calculate the similarity between each two sample users; calculate the neighbor matrix A; calculate the link matrix; calculate the similarity measure of each two sample users, and the similarity measure is the highest Merge two sample users of ; use the objective function to obtain the optimal clustering result; repeat iterations until k clusters are formed; classify user portrait groups. The present invention uses a clustering algorithm to construct user portraits of travel agencies and classifies them into groups, which is beneficial to sorting out the commonality between users, so as to better understand user needs, further develop new users, and determine the model of tourism products.

Description

Technical field[0001]The invention belongs to the technical field of data processing, and specifically relates to a method for constructing a multi-dimensional information portrait of a travel agency user.Background technique[0002]With the development and application of big data technology, more and more companies and government agencies have an increasing demand for precise positioning of a person’s characteristics. The most important thing in the process of integrating big data projects is to target different types of people. Demand for precision marketing and personalized solutions. On the one hand, users need to filter out the products they are interested in from a large amount of product information; on the other hand, companies need to accurately sell suitable products to the right users through the modification of product characteristic parameters. Currently commonly used recommendation algorithms are: content-based recommendation, user behavior-based recommendation, hybrid m...

AI Technical Summary

Problems solved by technology

Currently commonly used recommendation algorithms include: content-based recommendation, user behavior-based recommendation, hybrid model-based recommendation, label-based recommendation, etc.; for the operation model with relatively small product volume and large user volume, precise matching is required In , these commonly used recommended methods are not very practical

Images