SBIR Phase II: An Innovative and More Effective Means to Manage the Communication Process Between Colleges and Prospective Students

这个小型企业创新研究（SBIR）第二阶段项目提出将预测建模技术商业化，该技术可以自动适应高等教育提供者（学院和大学）与消费者（未来学生）之间不断变化的互动模式。目前的方法只产生回顾性的静态模型，由于高等教育招聘周期的特殊性，数据采集和应用到新的前景之间需要至少一年的滞后。因此，数据挖掘技术在大学招聘中只获得了有限的普及。这里提出的方法采用了一个专有的自适应建模引擎（AME）来利用来自CRM系统的实时交易数据，并动态更新评分算法来预测结果。AME依赖于逻辑接口和统一的维度数据模型来提取可分析的记录集，这些记录集准确地表示系统有效日期范围内任何时间片的底层事务数据的状态。这些关键技术的集成允许在招聘过程中识别关系模式，并允许评分算法在单个招聘周期内动态适应不断变化的模式。据信，不断变化的大学生人口结构将对传统的大学商业模式构成一些重大障碍，并可能危及该县许多高等教育机构未来的财务健康。十年来每年需求增长的趋势，以进入大学的新生人数为代表，将在2009年结束。与过去10年24%的市场增长形成鲜明对比的是，未来的入学人数总体上将保持停滞，在许多地方，大学入学人数实际上将下降。此外，为高等教育市场提供资金的高中毕业生的地理人口、种族和文化组合正在发生巨大变化。随着学生竞争的急剧增加，面对不断上升的出勤成本，不断减少的捐赠，不断变化的人口统计数据，以及大学学生的减少，每所大学的生存能力，更不用说繁荣了，将直接取决于其识别能力，理解，并以更有效和更具成本效益的方式与学生沟通。那些能够通过使用AME等创新工具来适应这一新形势的学校将蓬勃发展，而那些无法适应的学校将面临入学率下降和财务不稳定的不确定未来。

This Small Business Innovation Research (SBIR) Phase II project proposes to commercialize a predictive modeling technology that automatically adapts to changing interaction patterns between providers of higher education (colleges and universities) and consumers (prospective students). Current methods produce only retrospective static models which, due to peculiarities of the higher education recruitment cycle, require at least a one-year lag between data acquisition and application to new prospects. As a result, data mining techniques have gained only limited popularity in college recruiting. The approach proposed here employs a proprietary adaptive modeling engine (AME) to leverage real-time transactional data from a CRM system and dynamically update scoring algorithms to predict outcomes. AME relies on a logical interface and unified dimensional data model to extract analyzable record-sets accurately representing the state of underlying transactional data at any time-slice within the system's effective-dated range. The integration of these key technologies allow relationship patterns to be identified in the recruitment process as they occur and scoring algorithms to dynamically adapt to changing patterns within a single recruitment cycle. It is believed that the changing demographics of college-going students will present a number of significant obstacles to the traditional college business model and could jeopardize the future financial health of many higher education providers in this county. The decade-long trend of yearly increases in demand, as represented by the number of new students entering college, comes to an end in 2009. In stark contrast to the 24% growth the market has experienced over the past decade, future enrollment numbers will remain stagnant overall, and in many localities college enrollment will actually decline. Furthermore, dramatic shifts are coming in the geodemographic, ethnic, and cultural mix of high school graduates that feed the higher education market. As competition for students increases dramatically in the face of rising attendance costs, dwindling endowments, changing demographics, and a decline in college-bound students, each college's ability to survive, much less prosper, will depend directly on its ability to identify, understand, and communicate with students in a more efficient and cost-effective manner. Those that are able to adapt this new landscape through the use of innovative tools like AME will flourish, and those who are unable to adapt will face an uncertain future of declining enrollments and financial instability.