Article Details

Abstract

To serve the high-quality development of the regional digital economy and accurately identify the talent competence requirements of data-related positions, this study collects recruitment information of data-related positions in Zhengzhou from Zhilian Recruitment and BOSS Zhipin through web crawling, and obtains 639 valid "position requirement" texts after data cleaning. Text mining analysis is conducted using Chinese word segmentation, word frequency statistics, and TF-IDF keyword extraction. The number of topics is determined according to perplexity and topic coherence; an LDA topic model is then applied to extract topics, and the pyLDAvis tool is used for visual presentation. The results show that the demand for data-related positions mainly concentrates on two major themes: decision-making and technology-oriented. The former emphasizes business understanding, management, coordination, and communication skills, while the latter focuses on programming skills, tool application, and practical project experience. The findings indicate that enterprises' demand for data talent presents a compound "technology + business" feature, providing a reference for curriculum optimization in universities and career planning for job seekers.

Keywords

LDA topic model; text mining; talent demand analysis; recruitment data; data-related positions

References

1. [1] Xinhua News Agency Outline of the 15th Five Year Plan for National Economic and Social Development of the People's Republic of China [R/OL]. (2026-03-13) https://www.gov.cn/yaowen/liebiao/202603/content_7062633.htm?enable_bottom_share_style = 1&hybrid_event_param = HybridEventP.
2. [2] Zhengzhou Municipal Committee of the Communist Party of China Proposal of the Zhengzhou Municipal Committee of the Communist Party of China on Formulating the 15th Five Year Plan for National Economic and Social Development of Zhengzhou City [R/OL]. (2025-12-17) https://fgw.zhengzhou.gov.cn/work/9794078.jhtml.
3. [3] Wang Yimei A Data based Job Analysis and Recommendation System Based on LDA and Improved Apriori Algorithm [D]. Shanghai: Shanghai Normal University, 2024
4. [4] Liu Xing, Zhang Haotian, Tian Xuan Artificial Intelligence, Human Capital Structure, and Flexible Employment: Empirical Evidence from Online Recruitment Big Data [J]. Economic Research, 2025, 60 (7): 240-260
5. [5] Tan Ying, Liu Yingjie, Zhang Xun Statistical measurement of the degree of digital transformation in enterprises: from the perspective of demand for digital technology talents [J]. Economic Research, 2025, 60 (4): 122-138
6. [6] Blei D M，Ng A Y，Jordan M I. Latent Dirichlet Allocation [J].The Journal of Machine Learning Research，2003，3：993-1022.
7. [7] Wang Meng, Su Chengcheng, Chen Zhide Topic mining and analysis of thesis review comments based on LDA and Word2Vec models [J]. Journal of Fujian Normal University (Natural Science Edition), 2024, 40 (5): 41-51
8. [8] Nasah S A，Dong Y X. Envelope Dimension Reduction with Application to Binary Classification [J].Journal of Systems Science & Complexity，2026，39（1）：79-87.
9. [9] Wang Hui Research on Text Clustering Based on LDA Theme Model [D]. Lanzhou: Lanzhou University, 2018
10. [10] Cheng Shujia, Liu Linlin, Xia Fang Analysis of the Evolution of Research Themes in Vocational Education in China Based on LDA Theme Model [J]. Vocational and Technical Education, 2025, 46 (36): 64-71

Funding

2025 Zhengzhou Social Science Research Project (0557)

About the authors

Song Yanli (1997-), female, Han ethnicity, from Ruzhou, Henan, teaching assistant, master's student, research interests include big data technologies.