A Relevance Evaluation Method for Internet of Things Item Information Search Ranking

Abstract

The invention relates to a correlation evaluation method for searching and sorting item information in the Internet of Things, comprising the steps of: 1) establishing a user personalized interest model based on user historical query information; 2) establishing an item quality scoring mechanism based on user historical query information; 3) Construct a tripartite graph based on the above two contents, and obtain a correlation score through a random walk algorithm. The beneficial effects of the present invention are: the present invention mainly establishes a user's personalized interest model, establishes a quality scoring mechanism for item attributes based on the user's historical query information, and establishes a model for evaluating the relevance of candidate items. After the model is established, the user-item-item attribute tripartite graph is constructed based on the user's preference weight for item attributes and the item attribute quality score, and the correlation score between the target user and different candidate items is obtained through the random walk propagation algorithm, so as to realize the Relevance assessment.

Description

technical field [0001] The invention relates to a correlation evaluation method, in particular to a correlation evaluation method for searching and sorting item information in the Internet of Things. Background technique [0002] With the rapid development and popularization of cheap miniature sensors, wireless communication technology and mobile computing technology, many real-world objects with perception, computing and communication capabilities (such as wearable devices, smart meters, key chains and seats, etc.) will widely exist in the Internet of Things. It can be predicted that, just as the current use of web search engines to find Internet information, with the increasing demand for searching and browsing real-world item information in various application fields (such as environmental monitoring, smart home, smart logistics, and smart construction sites, etc.), it is very likely that People will be able to search and browse information on various items in the real w...

AI Technical Summary

Problems solved by technology

In existing research, due to the heterogeneity of Internet of Things items in terms of functions, models, materials, and manufacturers, users have different requirements for different attributes of items when querying items. For example, when searching for environmental monitoring sensors, some Users like fast collection frequency, and some users prefer low collection cost. Therefore, how to establish a correlation evaluation method for candidate items that varies from person to person is a major problem in the search and ranking of IoT items.

Images