A User Information Retrieval Method Based on Vector Similarity

Abstract

The invention provides a vector similarity-based user information retrieval method. The method comprises the following steps of S100, receiving input data; S200, obtaining Xth stored data; S300, obtaining first similarity S1 between the input data and the Xth stored data; if the first similarity S1 is greater than or equal to a first threshold D1, executing the step S300; S400, obtaining second similarity S2 between the input data and the Xth stored data; if the second similarity S2 is greater than or equal to a second threshold D2, executing the step S500; S500, taking the Xth stored data asa user information retrieval result; and S600, repeatedly executing the step S200 for (X+1)th stored data. According to the method, existing user information similar to user information can be effectively obtained; and the power dispatching accuracy and efficiency are improved.

Description

technical field [0001] The invention belongs to the field of big data mining and information retrieval, and in particular relates to a user information retrieval method based on vector similarity. Background technique [0002] With the development of power business, the scale of power users has expanded rapidly, and the power system has accumulated a large amount of data related to power consumption of users, including basic information of users, historical power consumption information and other data. Using these data, the similarity of medium and large power users can be estimated, so as to provide a prediction basis for the configuration of power facilities and the regulation of power energy. In a simple example, for a newly built residential area, based on the accumulated data, the configuration of power facilities and historical power consumption of similar residential areas can be retrieved, so as to predict the configuration of power facilities that should be based on...

AI Technical Summary

Problems solved by technology

[0003] However, the existing estimation methods for user similarity are relatively extensive, and the utilization of accumulated data is not particularly comprehensive, and the utilization rate is low, resulting in inaccurate retrieval results of user similarity, which cannot meet the requirements of detailed power facility configuration and power deployment. requirements

Images