Scholarlink

<p>Employee attrition is one of the most critical problems faced by the leading IT organizations across the world. In this research, we have undertaken a data-driven approach to predict the duration of employment for any employee in an IT organization within India. Unlike most extant research in this domain, we collected the employment and professional data about IT employees across multiple organizations directly from the LinkedIn website and later cleaned it and used it to build various machine learning based models. The results show that the decision tree regressor and the neural network models predict the tenure most accurately. We have also created clusters using our dataset and have done the risk profile analysis for each cluster. These investigations reveal that total experience, seniority, education, number of technology certifications, and professional connections, are the most significant factors in predicting duration of work for an employee within an organization. Based on the findings of this research, we have also proposed a framework which organizations can adopt for hiring and retaining employees.</p>

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

Employment tenure prediction for Indian IT professionals using linkedin network data and machine learning based techniques

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年12月12日

查看全文
  • Prajas Naik,
  • Swapnajit Chakraborti

摘要

Employee attrition is one of the most critical problems faced by the leading IT organizations across the world. In this research, we have undertaken a data-driven approach to predict the duration of employment for any employee in an IT organization within India. Unlike most extant research in this domain, we collected the employment and professional data about IT employees across multiple organizations directly from the LinkedIn website and later cleaned it and used it to build various machine learning based models. The results show that the decision tree regressor and the neural network models predict the tenure most accurately. We have also created clusters using our dataset and have done the risk profile analysis for each cluster. These investigations reveal that total experience, seniority, education, number of technology certifications, and professional connections, are the most significant factors in predicting duration of work for an employee within an organization. Based on the findings of this research, we have also proposed a framework which organizations can adopt for hiring and retaining employees.