ESG accounting disclosure quality research based on accelerated genetic algorithms

With the development of computer technology, current research on the quality of ESG accounting disclosure faces issues such as unreasonable evaluation indicators and difficulties in high-level data processing. To address this shortcoming, this study proposes an accounting disclosure quality evaluation model based on an accelerated genetic algorithm. By combining genetic search and projection pursuit methods, it achieves effective dimensionality reduction and feature extraction of high-dimensional ESG data. The results show that the model has the highest decision sensitivity of 28.53%, and it is also closest to the recorded data in the evaluation of comprehensive indicators, which proves its reliability. Not only that, it can still maintain good performance in case of data distortion, which proves its stability in accounting disclosure quality. The above results show that the research model has great advantages in the research of accounting disclosure quality and provides new ideas and methods for the research field of accounting disclosure quality.