In order to improve the development and construction level of colleges and universities and help teachers improve their scientific and technological abilities, a model based on deep learning model for the extraction and analysis of factors affecting the improvement of college teachers scientific and technological abilities was proposed. This article analyzes the data of teaching evaluation and finds that the text contains students subjective understanding of teachers teaching quality defects. By extracting key words from students teaching evaluation texts and combining with the teaching evaluation indexes, a teaching evaluation label system integrating teaching evaluation texts is designed. In order to find the defects of teachers teaching ability, this article, based on the principle of data portrait, combines the characteristics of teachers personal basic information, curriculum information, teaching evaluation information, and social relations to portrait teachers. The experimental results show that the F1 value extracted from the evaluation labels fluctuates in the evaluation text data of different colleges, with the lowest value of 91.7 in the School of Statistics and the highest value of 95.8 in the School of Foreign Languages. The algorithm in this article has a higher F1 value of the evaluation label vector extracted from the evaluation text of different grades. F1 values showed a trend of gradient decline with the increase of grade, and the decreasing range became bigger and bigger. Conclusion. The constructed teacher portraits are more accurate and effective, and provide a comprehensive and effective data model for teaching ability improvement method recommendation strategy.

A scientific and reasonable student evaluation system plays a crucial guiding role in the development of higher education. However, for the current student evaluation theories and models, their operability is relatively weak. The constructed models have poor adaptability (weak robustness), lack cross-scenario research on guiding policies, and the evaluation results of the models are also relatively single. Therefore, this study proposes a hierarchical student evaluation enhanced decision-making model which is based on a zero-order fuzzy classifier as the construction unit. By introducing the administrative decision-making characteristics of the educational administration department (such as policy orientation indicators, teaching intervention suggestions, etc.), combined with a hierarchical fuzzy system and an improved ridge regression algorithm, it achieves the collaborative optimization of evaluation efficiency and interpretability. The experimental results show that the model demonstrates excellent classification performance and semantic interpretability on the degree student evaluation dataset, and can accurately predict students’ academic performance and support personalized educational decisions.