Research on the Effects of the Intelligent Marking Algorithm Based on the Analytic Hierarchy Process
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This paper compares two intelligent marking algorithms developed by a laboratory, establishing a multidimensional evaluation system for scoring effectiveness. Three first-level indicators are defined: accuracy (measured by Spearman coefficient, MSE, RMSE, MAE), stability (evaluated via coefficient of variation and variance), and consistency (assessed by consistency rate). Using the Analytic Hierarchy Process (AHP), the evaluation framework is stratified into target, criterion (including 1st/2nd-level indicators), and scheme layers. Judgment matrices are constructed via 1-9 scale pairwise comparisons, with indicator weights determined by eigenvalue method after consistency testing (CR < 0.1). Key weights include accuracy (63.70%), consistency (25.83%), and stability (10.47%). Comprehensive scores incorporate data standardization for negative indicators. Experimental results from 5 subjects show Intelligent Algorithm 1 outperforms Algorithm 2 in most scenarios, particularly in accuracy (Spearman 0.9027–0.9823 vs. 0.8595–0.9794) and consistency (0.5295–0.8922 vs. 0.4676–0.8862). While Algorithm 2 excels in Subjects 2–4, Algorithm 1 demonstrates more balanced performance. This study provides a replicable AHP-based framework for evaluating intelligent marking algorithms, supporting rational selection in educational assessment and intelligent decision-making domains.
Fields of Study & Subjects
Educational Technology and Assessment
Multi-Criteria Decision Making
Educational Technology and Pedagogy
Publication details
- Journal / source
- IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
- Pages
- 316–320
- Publication date
- 28 January 2026
- Language
- EN