We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Design and Use of Deep Confidence Network Based on Crayfish Optimization Algorithm in Automatic Assessment Method of Hearing Effectiveness.
- Authors
Ying Cheng
- Abstract
INTRODCTION: Listening strategy analysis and assessment not only need objective and fair sound listening strategy analysis, but also need high-precision and high real-time assessment model, and even more need in-depth analysis and feature extraction of the influencing factors of listening assessment. OBJECTIVES: To address the problems of current automatic assessment methods, such as non-specific application, poor generalization, low assessment accuracy, and poor real-time performance. METHODS: This paper proposes an automatic assessment method based on a deep confidence network based on crawfish optimization algorithm. First, the multi-dimensional listening strategy evaluation system is constructed by analyzing the listening improvement strategy; then, the depth confidence network is improved by the crayfish optimization algorithm to construct the automatic evaluation model; finally, through the analysis of simulation experiments. RESLUTS: The proposed method improves the evaluation accuracy, robustness, and real-time performance. The absolute value of the relative error of the automatic evaluation value of the proposed method is controlled in the range of 0.011, and the evaluation time is less than 0.005 s. The method is based on a deep confidence network based on the crayfish optimization algorithm. CONCLUSION: The problems of non-specific application of automated assessment methods, poor generalization, low assessment accuracy, and poor real-time performance are addressed.
- Subjects
OPTIMIZATION algorithms; CRAYFISH; FEATURE extraction; ABSOLUTE value; CONFIDENCE
- Publication
EAI Endorsed Transactions on Scalable Information Systems, 2024, Vol 11, Issue 3, p1
- ISSN
2032-9407
- Publication type
Article
- DOI
10.4108/eetsis.4847