CAREER: Mining Career, Education and Job Data to Bridge the Talent Gap between Demand and Supply

人才差距是雇主需要的工人与劳动力市场提供的劳动力之间的不匹配，在劳动力市场的许多部门都存在，对与STEM相关的劳动力市场和少数群体来说更是如此。迫切需要了解和研究人才差距，并为包括员工、雇主和教育机构在内的不同利益攸关方制定有用和可行的建议，以弥合这一差距。该项目收集和深入分析了大量的职业、教育和工作(CEJ)数据，并开发了衡量微观和宏观层面人才缺口的量化方法和一套填补缺口的推荐技术。这项研究对CEJ数据分析领域有限的知识库做出了重大贡献。这些发现、工具和文件将帮助教育和经济学等领域的研究人员更好地研究美国劳动力市场的人才缺口挑战。这个项目通过课程和论文项目，让研究生和本科生提高他们解决实际问题的知识和技能。该项目还让K-12的学生参加夏令营，并鼓励他们更早地为自己的教育和职业发展做准备。该项目专注于三项主要任务：使用机器学习方法收集和建模异质CEJ数据，测量和解释人才差距，以及开发推荐解决方案来弥补差距。为了解决第一项任务，该项目从在线专业网络等多种来源收集了CEJ的各种数据，并开发了先进的嵌入方法。为了解决第二个任务，该项目开发了基于嵌入结果的新的定量测量和定制的可视化技术。为了处理第三项任务，该项目开发了复杂的推荐方法，为员工、雇主和教育机构提供可行的建议。该项目的结果将以同行评议出版物、开源软件、教程、研讨会和工作室的形式传播。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The talent gap, a mismatch between the workers that employers need and the workforces that labor markets provide, exists across many sectors of labor markets, and is even worse for STEM-related labor markets and minority groups. There is a critical need to understand and study the talent gap and develop helpful and actionable recommendations for different stakeholders including employees, employers, and education institutes to bridge the gap. This project collects and deeply analyzes the large wealth of career, education and job (CEJ) data, and develops quantitative methods for measuring the micro and macro level talent gap and a suite of recommendation techniques for filling the gap. This study contributes significantly to the limited knowledge base in the area of CEJ data analytics. The findings, tools and documents will help researchers in fields such as education and economics better study the talent gap challenge of U.S. labor markets. This project involves, through courses and thesis projects, graduate and undergraduate students to enhance their knowledge and skills in solving real problems. This project also involves K-12 students through summer camps and encourages them to earlier prepare their education and career development.This project focuses on three major tasks: collecting and modeling heterogenous CEJ data with machine learning methods, measuring and interpreting the talent gap, and developing recommendation solutions to bridge the gap. To tackle the first task, this project collects a variety of CEJ data from multiple sources such as online professional networks and develops advanced embedding methods. To solve the second task, this project develops novel quantitative measurements and customized visualization techniques based on the embedding results. To handle the third task, the project develops sophisticated recommendation methods for generating actionable suggestions for employees, employers and education institutes. The results of this project will be disseminated in the form of peer-reviewed publications, open-source software, tutorials, seminars, and workshops.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.