Design and Implementation of a Mind Map Generation System Based on TextCNN

DOI: https://doi.org/10.62517/jbdc.202601317

Author(s)

Ying Tian1, Yang Zhao1, Wanyue Liu1,2,*

Affiliation(s)

1School of Artificial Intelligence and Big Data, Henan University of Technology, Zhengzhou, Henan, China 2iFLYTEK Co., Ltd., Hefei, Anhui, China *Corresponding Author

Abstract

In the process of organizing course materials, it is necessary to extract operation steps, core concepts, review points and other contents manually, which requires a lot of time and may lead to problems such as repetition and omission. In this study, in order to solve this problem, the TextCNN system is used to create a mind map. First, the system processes the CSV data and the course text, and inputs the processed data into the model in the form of tensors, so as to obtain a series of numerical information. Secondly, the system uses one-dimensional convolution kernel to extract local semantic features and complete text classification. Finally, the system generates a variety of forms of maps, including timelines and tree maps, and can also generate mind maps. This system has a simple structure, fast training speed, and does not require high-performance hardware, which can meet the needs of visual presentation of course texts.

Keywords

TextCNN; Text Classification; Knowledge Visualization; Mapdocument; Classification Mind Map Generation System.

References

[1] Reales D, Manrique R, Grévisse C. Core concept identification in educational resources via knowledge graphs and large language models. SN Computer Science, 2024, 5(8): 1029. [2] Zhang Y, Wallace B C. A sensitivity analysis of (and practitioners’ guide to) convolutional neural networks for sentence classification//Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2017: 253-263. [3] Xu L, Hu H, Zhang X, et al. CLUE: A Chinese language understanding evaluation benchmark//Proceedings of the 28th international conference on computational linguistics. 2020: 4762-4772. [4] Lin C, Lin Y J, Yeh C J, et al. Improving multi-criteria Chinese word segmentation through learning sentence representation//Findings of the Association for Computational Linguistics: EMNLP 2023. 2023: 12756-12763. [5] Kim Y. Convolutional neural networks for sentence classification//Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). 2014: 1746-1751. [6] Nair V, Hinton G E. Rectified linear units improve restricted boltzmann machines//Proceedings of the 27th international conference on machine learning (ICML-10). 2010: 807-814. [7] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 2014, 15(1): 1929-1958. [8] Zeng G. Invariance properties and evaluation metrics derived from the confusion matrix in multiclass classification. Mathematics, 2025, 13(16): 2609. [9] Lu J, Dong X, Yang C. Weakly supervised concept map generation through task-guided graph translation. IEEE transactions on knowledge and data engineering, 2023, 35(10): 10871-10883. [10]Stamper J, Gaind B, Thankachan K, et al. Hierarchical concept map generation from course data//AAAI 2023 Workshop on Artificial Intelligence in Education (AI4Edu). 2023.