SBIR Phase II: A Dynamic Real-Time Analytics Recruiting Platform

SBIR第二阶段项目旨在通过动态采购和分析平台为招聘行业带来透明度和效率。目前，在为一个需要确切标准的职位进行征聘时，很难找到理想的候选人。拟议的平台将使招聘人员能够快速识别最佳候选人，并了解这些候选人的地理位置、工作和接受教育的地点。这种数据驱动的透明度将改善招聘人员/招聘经理的互动，并允许更快地填补职位，减少员工流动造成的生产力损失。该平台将突出最有资格的候选人，无论先入为主的偏见，以帮助发现被忽视的候选人。这些效率将有利于招聘人员，招聘公司，个人候选人，大学和社会。鉴于该行业的规模（1600亿美元）和效率低下的规模，该项目具有巨大的商业影响潜力。第二阶段的研究和开发将主要集中在机器学习技术上，与强大的计算框架一起使用，并应用于包含数十亿数据点的大量数据集。该技术将用于驱动实时动态分析，从而通过交互式仪表板提供强大的招聘分析和理想的求职者。第二阶段将以第一阶段在机器学习分类系统、实时处理大量数据的并行计算环境和用户友好的可视化等领域取得的进展为基础。第二阶段的目标包括提高数据体系结构的处理能力，对输入属性进行建模，提高建模算法的准确性，以及提高整体界面性能。该奖项反映了NSF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评审标准进行评估，被认为值得支持。

This SBIR Phase II project intends to bring transparency and efficiency to the recruiting industry through a dynamic sourcing and analytics platform. Currently, it is exceedingly difficult to identify ideal candidates when recruiting for a position requiring exact criteria. The proposed platform will enable recruiters to rapidly identify optimal candidates and understand where these candidates are geographically concentrated, working, and being educated. This data-driven transparency will improve recruiter/hiring manager interactions and allow for positions to be filled more rapidly, with less productivity lost by employee turnover. The platform will highlight the most qualified candidates for a position, regardless of preconceived bias, to help uncover overlooked candidates. These efficiencies will benefit recruiters, hiring companies, individual candidates, universities, and society. Given the size of the industry ($160 billion) and the scale of inefficiencies, the project has vast commercial impact potential. Phase II research and development will be primarily focused around machine learning techniques, leveraged alongside a powerful computing framework, and applied to a substantial dataset containing billions of data points. This technology will be used to drive real-time dynamic analysis resulting in powerful recruiting analytics and ideal job candidates via interactive dashboards. Phase II will build upon the progress achieved in Phase I in the areas of machine-learning classification systems, parallel computing environments to process large quantities of data in real-time, and user-friendly visualizations. The goals of Phase II include increasing the processing power of the data architecture, modeling imputed attributes, improving the accuracy of modeling algorithms, and increasing overall interface performance.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.