Vertical Search Engine and Graph Homomorphism for Enhancing the Cybersecurity Workforce

当一个新的或快速变化的劳动力市场出现时，求职者在证明自己的能力和雇主在评估证书方面往往存在信息差距。这对于快速扩张的网络安全人员来说是一个特别的挑战。该项目将以SFS项目为测试平台，创造实用工具，弥合雇主和对网络安全职业感兴趣的个人之间的差距。该项目还将支持网络安全教育者的实践社区，以鼓励采用这些创新工具。这个探索性项目可能会面临收集和整合来自许多网站的招聘信息数据以及从非结构化文本创建数据结构的挑战。然而，如果成功的话，这个项目可以通过帮助网络安全求职者产生重大影响。它也可以作为一个试点，为其他劳动力市场开发类似的工具。该项目将开发两个重要且实用的工具，以支持未来和当前的网络安全工作人员：(1)一个名为Bruce的垂直搜索引擎，一个寻找综合网络安全学习资源的信息交换所；(2)网络安全职位与求职者技能和胜任力的匹配工具。该项目将开展研究，以创建一种准确、可扩展、细粒度的命名实体识别（NER）方法，该方法可以从网络安全学习资源中提取各种实体，包括课程名称、讲师姓名、教科书和考试信息。此外，除了简单的语法/语义匹配之外，本研究还将设计一种不同结构和词汇的职位与求职者之间的映射算法。网络安全职位发布将映射到NIST NICE网络安全框架中最相关的功能和类别，这是网络安全相关活动、角色、技能和概况的现代分类法。该项目的研究方法是使用理论计算机科学中的图同态技术和机器学习领域的图嵌入技术。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

When there is a new or rapidly changing labor market, there is often an information gap between job seekers on documenting their competencies and employers in evaluating credentials. This is particular a challenge in the rapidly expanding cybersecurity workforce. This project will create practical tools to bridge the gap between employers and individuals interested in cybersecurity careers, using the SFS program as a test bed. The project will also support a Community of Practice of cybersecurity educators to encourage the adoption of these innovative tools. This exploratory project may face challenges with collecting and integrating job posting data from many sites and with creating data structures from unstructured texts. However, if successful, this project could have a significant impact by helping cybersecurity career seekers. It may also serve as a pilot to develop similar tools for other labor markets. This project will develop two important and practical tools to support future and current cybersecurity workforce: (1) a vertical search engine, called Bruce, a clearinghouse for finding comprehensive cybersecurity learning resources; and (2) a matchmaking tool between cybersecurity job postings and job seekers' skills and competencies. The project will conduct research to create an accurate, scalable, fine-grained Named Entity Recognition (NER) method that can extract various entities from cybersecurity learning resources, including course title, instructor name, textbook, and exam information. In addition, the proposed research will design a mapping algorithm between posted jobs and job seekers with varying structures and vocabulary beyond a simple syntactic/semantic matching. Cybersecurity job postings will be mapped to the most relevant functions and categories in the NIST NICE Cybersecurity Framework, a modern taxonomy of cybersecurity related activities, roles, skills, and profiles. The project's research approach is to use techniques in graph homomorphism from theoretical computer science and graph embedding from machine learning fields.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.