Modern methods of collecting and processing information about the labour market and the directions of their use in the practice of human resource management

Abstract

The article presents the results of a study on methods for obtaining labour market data from web sources. The relevance of the research stems from the need to explore innovative approaches to collecting and analyzing labour market data, ensuring comprehensive, reliable, and timely information in the context of the digital transformation of society. The purpose of the study is to compare the methods of collecting and analyzing information, to identify their advantages and disadvantages. Methodologically, an interdisciplinary approach combines the methods of data science, machine learning, natural language processing and human resource management. Empirically, the study draws from expert surveys conducted with heads of Belarusian IT companies and data sourced from the Artificial Analysis platform. Key findings reveal that large language models (LLM) outperform other methods in terms of speed, quality and ease of use. Among the top 10 models, GPT-4o is recognized as the most effective, prioritized for business tasks and economic analysis. The advantages and disadvantages of data collection and analysis methods are revealed. The scientific novelty of the research consists in obtaining new theoretical and empirical results that reveal the potential of using LLM to obtain data on the labour market, as well as in identifying priority areas for the development of methodology and tools in this area. The practical significance lies in the development of a methodology for comparative analysis of the effectiveness of methods for collecting and analyzing information about the labour market from web sources; substantiation of promising directions for the development of information collection methods that open up opportunities for creating innovative solutions in the field of human resource management based on artificial intelligence and big data processing. Financial support for the research was provided by the Belarusian Republican Foundation for Basic Research within the framework of the scientific project No Г24-013 “Theoretical approaches and methodological support for labour market analysis in the Republic of Belarus using big data”.

Keywords

• api
• large language models
• web page parsing
• rss feeds
• data collection
• labour market analysis