Scholarlink

In the modern world, artificial intelligence technologies are actively implemented in various fields of activity, including personnel management. In the article, this method of using large language models (LLM) to automate the recruitment process through resume analysis. Based on the aisale.kz product, ByteMachine conducted a study on the effectiveness of using LLM in the parliament of candidate qualifications taking into account the requirements of vacancies. Experiments were conducted using models based on the BERT design, during which real results and vacancies were analyzed. The results of measuring the accuracy and completeness indicators, indicating the capabilities of the models, effectively assess the suitability of candidates. Comparison with conservative recruitment methods revealed the importance of taking into account processing time and increasing the objectivity of assessments. The limitations of models associated with taking into account the qualitative characteristics of candidates are discussed and ways to overcome them are provided. The need for further results of the search for algorithms and the provision of additional data to clarify accuracy estimates is noted. A conclusion is made about the significant potential of LLM in the transformational processes of recruiting and the prospects for their development in the field of personnel management.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on the Application of Large Language Models (LLM) for Improving Recruitment Efficiency and Accuracy

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Alimzhan Yerkebulan,
  • Madina Mansurova,
  • Assel Abdildayeva,
  • Olzhas Sharip

摘要

In the modern world, artificial intelligence technologies are actively implemented in various fields of activity, including personnel management. In the article, this method of using large language models (LLM) to automate the recruitment process through resume analysis. Based on the aisale.kz product, ByteMachine conducted a study on the effectiveness of using LLM in the parliament of candidate qualifications taking into account the requirements of vacancies. Experiments were conducted using models based on the BERT design, during which real results and vacancies were analyzed. The results of measuring the accuracy and completeness indicators, indicating the capabilities of the models, effectively assess the suitability of candidates. Comparison with conservative recruitment methods revealed the importance of taking into account processing time and increasing the objectivity of assessments. The limitations of models associated with taking into account the qualitative characteristics of candidates are discussed and ways to overcome them are provided. The need for further results of the search for algorithms and the provision of additional data to clarify accuracy estimates is noted. A conclusion is made about the significant potential of LLM in the transformational processes of recruiting and the prospects for their development in the field of personnel management.