O2O service recommendation method and system based on preference spatio-temporal evolution analysis

By constructing a service score and category matrix, it is estimated that the dynamic similarity between consumers has solved the problem of low accuracy of O2O services in existing technologies, and achieved more efficient consumer relationship evaluation and recommendation performance.

CN119379393AActive Publication Date: 2025-01-28RENMIN UNIVERSITY OF CHINA
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411513259.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-28
Publication Date
2025-01-28
Estimated Expiration
2044-10-28

AI Technical Summary

Technical Problem

In the prior art, preference similarity estimates based on static panel data lead to low accuracy in O2O service recommendations.

Method used

By obtaining the data of O2O consumers, services and services, build a service scoring matrix and service category matrix within the historical period of history, and count the service selection, scores and position changes between consumers. Dynamic similarity to obtain dynamic preference similarity.

Benefits of technology

Improve the accuracy of the recommendation of O2O services, and improve the recommendation performance by more accurately measure the relationship between consumers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119379393A_ABST
    Figure CN119379393A_ABST
Patent Text Reader

Abstract

The invention provides an O2O service recommendation method and system based on preference spatio-temporal evolution analysis, a storage medium and electronic equipment, and relates to the field of O2O service recommendation. The method comprises the following steps: firstly, acquiring an O2O consumer set, an O2O service set and a service category set, and constructing a service score matrix and a service category matrix; then, three different dynamic similarities are obtained based on preference changes of consumers in service selection, evaluation and geographic positions of the selected services; fusing different dynamic similarities to obtain dynamic preference similarities; and finally, for a target consumer, sorting neighbors of the target consumer by using dynamic preference similarity, and carrying out priority recommendation of the O2O service. Based on the dynamic similarity proposed by the preference spatio-temporal evolution of the consumers, the relationship between the consumers is measured more accurately, so that the recommendation performance of the O2O service is improved. Experimental results show that the proposed dynamic similarity has high accuracy and can effectively reflect powerful consumer relationships.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method and system for recommendation based on O2O data

    CN108022150A

  • Recommendation system and method based on multi-dimensional similarity

    CN111026957A

  • Recommendation method and system for fusing user request and service preference of Internet of Vehicles

    CN112989194A

  • Personalized product service scheme recommendation method based on credibility and forgetting function

    CN113407863A

  • Position recommendation method based on position timeliness characteristics and time perception dynamic similarity

    CN115687801A