Design of innovation ability evaluation model based on IPSO-LSTM in intelligent teaching

PeerJ Comput Sci. 2023 Nov 20:9:e1679. doi: 10.7717/peerj-cs.1679. eCollection 2023.

Abstract

Guided by the development of an innovative economy, students' innovative education has also become the focus of talent training. This research aims to realize the intelligent evaluation of students' innovation ability. In this article, we proposed an innovation ability framework that integrates students' psychological state and innovation evaluation indicators. Firstly, the qualitative description of psychological data is quantified using the Delphi method. Secondly, this article proposes an improved particle swarm optimization-long short-term memory (IPSO-LSTM) model to achieve high-precision evaluation and classification of innovation capabilities. The classification accuracy of this model for excellent, general and failed innovation capabilities is up to 95.3%. Finally, the characteristic contribution analysis of psychological and innovative ability characteristics is carried out. The results show that the evaluation of creative ability contributes more than 50% to the psychological aspects of excellent students. This shows the importance of psychological status on creative ability and provides a theoretical basis for integrating innovative education and psychological education in the future.

Keywords: IPSO-LSTM; Innovation ability evaluation; Intelligent evaluation; Psychological analysis.

Grants and funding

This work was supported by the Key Project of Guangdong Primary and Secondary School Teachers’ Education and Scientific Research Ability Improvement in 2022 “Research on the Practice Path of Future School Construction Oriented to the Cultivation of Innovative Talents” (No: 2022ZQJK108). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.