User perception evaluation method and electronic equipment based on backtracking search algorithm

Abstract

The invention discloses a user perception evaluation method based on a retrospective search algorithm, including: S1, defining a QoE variable structure model, the input variable is the KQI index of the service to be evaluated, and the output variable is the comprehensive user perception of the service; S2, Binary tree encoding relationship prediction model is adopted; S3, the improved BSA algorithm is adapted to the operator's crossover mutation strategy, including the mutation method based on variable length, providing single-point factor and combined factor combined mutation method; S4, realizing crossover based on the mapping value Map Strategies, providing single-point and multi-point crossover factor retention strategies. Research on the construction of KQI-QoE numerical relationship model to improve the quality and construction efficiency of the relationship model, and apply it to the user experience evaluation of the business scenario of the device connected to the Internet of Things platform. The invention also provides corresponding electronic equipment.

Description

technical field [0001] The invention belongs to the technical field of communications, and more specifically relates to a method for evaluating user perception based on a backtracking search algorithm and electronic equipment. Background technique [0002] There are usually two methods of user perception measurement, one is to actually investigate the user's application experience quality, and the other is to infer the application's user experience quality by analyzing business indicators. The two methods of actual investigation and analysis of business indicators are not enough to obtain user perception in real time and intuitively. At present, with the wide application of communication technology and IoT platform, various services and applications urgently need to improve their users' application experience based on a large amount of high-quality and useful data collected and stored. Due to the diversity of users, actual scenarios, and performance parameters, solving the ...

AI Technical Summary

Problems solved by technology

At present, most of them use a fixed model or a multiple linear regression to fit the prediction, resulting in a large deviation between the predicted perception and the actual perception

In addition, some scholars use GA (Genetic Algorithm, Genetic Algorithm) to construct complex fitting functions, but the algorithm's crossover mutation strategy is weak in ensuring solution diversity, and has inherent defects such as precocity and low efficiency.

Images