III: Small: Collaborative Research: Harnessing Big Data for Improving Career Mobility

III：小：协作研究：利用大数据提高职业流动性

• 批准号: 2006387
• 负责人: Xiaodong Lin
• 金额: $ 25万
• 依托单位: Rutgers University Newark
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2006387&HistoricalAwards=false
• 关键词: III; Small; Collaborative; Research; Harnessing

U.S. college students are facing critical challenges for their career development and job mobility, which is vital for their long-term career success, especially during global pandemic times. Indeed, the questions that often puzzle students include what career choices to choose next, how to update skills for future new jobs, and which learning opportunities to take. These challenges have been increasingly observed among different groups of students in different majors and socioeconomic statuses at many universities. This project collects and analyzes academic curriculum and student career data, discovers useful patterns about college curriculum and students’ career development, studies students’ career choices, and develops sophisticated solutions to improve their career mobility. This study makes significant contributions to the fields of data mining, machine learning, and education and career data analytics. The results of this project can bring new ways for understanding and improving college graduates’ career success, provide useful insights and tools for students to make their decisions on career development, and augment the service capability of college career and academic advising offices. This project integrates the research with education through new course module development, involving graduate and undergraduate students in research, and research showcases for local K-12 students. This project focuses on the following three specific aims (SA): mining and informing useful semantics and patterns about college curriculum and graduates’ career development; studying the career choices of college graduates; and developing sophisticated solutions to improve career mobility of college graduates. To achieve the first SA, this project develops a novel context-aware deep learning method for mining semantics from heterogenous textual data and discovers insightful horizontal and vertical patterns. To solve the second SA, this project mines multiple-scale career path patterns and develops a new hierarchical neural network method to model and predict graduates’ career choices. To achieve the third SA, this project develops novel reinforcement learning methods to recommend learning items for both graduated students and enrolled ones. The results of this project will be disseminated in the form of peer-reviewed publications, publicly available data set, 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.

美国大学生正面临着职业发展和工作流动性的严峻挑战，这对于他们的长期职业成功至关重要，尤其是在全球大流行时期。事实上，经常困扰学生的问题包括下一步选择什么职业，如何选择。更新未来新工作的技能，以及在许多大学的不同专业和社会经济地位的不同学生群体中越来越多地观察到这些挑战，该项目收集并汇总了学术课程和学生职业数据，发现了有用的模式。关于大学课程和学生的职业发展、学习这项研究为数据挖掘、机器学习、教育和职业数据分析领域做出了重大贡献，可以为理解和改进提供新的方法。大学毕业生的职业成功，为学生做出职业发展决策提供有用的见解和工具，并增强整合大学职业和学术咨询办公室的服务能力。该项目通过新课程模块的开发，涉及研究生和教育。本科生进行研究，并为当地提供研究展示该项目侧重于以下三个具体目标（SA）：挖掘和告知有关大学课程和毕业生职业发展的有用语义和模式；研究大学毕业生的职业选择；以及开发复杂的解决方案以改善职业流动性；为了实现第一个 SA，该项目开发了一种新颖的上下文感知深度学习方法，用于从异构文本数据中挖掘语义，并发现有洞察力的水平和垂直模式。为了解决第二个 SA，该项目挖掘多尺度的职业路径。模式并开发新的层次模式为了实现第三个SA，该项目开发了新颖的强化学习方法来为毕业生和在校学生推荐学习项目，该项目的结果将以同伴的形式传播。 -审查出版物、公开数据集、教程、研讨会和讲习班。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Scalable Heterogeneous Graph Neural Networks for Predicting High-potential Early-stage Startups

用于预测高潜力早期初创企业的可扩展异构图神经网络

• DOI: 10.1145/3447548.3467383
• 发表时间: 2021-08
• 期刊: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining
• 作者: Zhang, Shengming;Zhong, Hao;Yuan, Zixuan;Xiong, Hui
• 通讯作者: Xiong, Hui

Multi-Faceted Knowledge-Driven Pre-Training for Product Representation Learning

产品表征学习的多方位知识驱动预训练

• DOI: 10.1109/tkde.2022.3200921
• 发表时间: 2023-07-01
• 期刊: IEEE Transactions on Knowledge and Data Engineering
• 影响因子: 8.9
• 作者: Denghui Zhang;Yanchi Liu;Zixuan Yuan;Yanjie Fu;Haifeng Chen;Hui Xiong
• 通讯作者: Hui Xiong

MANE: Organizational Network Embedding with Multiplex Attentive Neural Networks

MANE：具有多重注意力神经网络的组织网络嵌入

• DOI: 10.1109/tkde.2022.3140866
• 发表时间: 2022-01
• 期刊: IEEE Transactions on Knowledge and Data Engineering
• 影响因子: 8.9
• 作者: Ye, Yuyang;Dong, Zheng;Zhu, Hengshu;Xu, Tong;Song, Xin;Yu, Runlong;Xiong, Hui
• 通讯作者: Xiong, Hui