SGER: Finding Interesting Patterns through Analysis of Complex Prediction Models

利用数据挖掘技术，可以为大型高维数据训练准确的预测模型。不幸的是，复杂的预测模型本身并不容易理解。为了使它们“易于理解”，分析人员需要更简单的模式来总结模型提取的复杂功能。这样的函数摘要的数量是压倒性的。原始数据空间的低维子空间的每个切片都可以包含一个有趣的函数摘要。这个项目的目标是开发自动高效地找到最“有趣”的函数摘要的技术。这分三步完成。首先，通过形式化各种模式类型的趣味性概念。其次，通过开发一种声明性语言来指定这些有趣的度量。使用声明性语言，分析人员定义他们感兴趣的内容，但他们不需要指定如何有效地找到它。第三，针对小语言片段的优化编译器处理性能效率。一个主要的研究挑战是在语言的表达性和使其适应有效的查询优化之间取得适当的平衡。这个项目的结果将为强大的探索性分析工具铺平道路。它们还将支持对声明性语言的优化器和用户友好界面的未来研究。该方法将利用鸟类知识网络（AKN）中鸟类学界组织的丰富数据资源进行验证。这将对识别影响地球生物多样性的最重要的环境变量的能力产生巨大影响。例如，土地管理者可以发现他们的决定对生态系统健康可能产生的影响。该语言的一个组成部分将通过AKN网站（http://www.avianknowledge.net/content）的Web服务向公众提供。其他结果将通过项目网站（http://www.cs.cornell.edu/~mirek/Projects/FunctionSummaries）传播。这将使从研究人员到土地管理人员或鸟类观察者、教师或学生等广泛的受众能够从收集的数据资源中获得新的知识。

With data mining techniques it is possible to train accurate prediction models for large high-dimensional data. Unfortunately, complex prediction models per se are not easy to understand. To make them 'digestible', analysts need simpler patterns that summarize the complex functions extracted by the model. The number of such function summaries is overwhelming. Each slice of a lower-dimensional subspace of the original data space could contain an interesting function summary.The goal of this project is to develop techniques for finding the most 'interesting' function summaries automatically and efficiently. This is done in three steps. First, by formalizing the notion of interestingness for a wide variety of pattern types. Second, by developing a declarative language for specifying these interestingness measures. With a declarative language analysts define what they find interesting, but they need not specify how to find it efficiently. Third, an optimizing compiler for a small language fragment handles the performance efficiency. A major research challenge is to strike the right balance between expressiveness of the language and making it amenable to effective query optimization.The results of this project will pave the way for powerful exploratory analysis tools. They will also enable future research on optimizers and user-friendly interfaces for the declarative language. The approach will be validated using the rich data resources being organized by the ornithological community in the Avian Knowledge Network (AKN). This will have a tremendous impact on the ability to identify the most significant environmental variables that affect biodiversity on the planet. For example, land managers could discover the possible impact of their decisions on an ecosystem's health.A component of the language will be available to the public through Web services on the AKN Web site (http://www.avianknowledge.net/content). Additional results will be disseminated through the project Web site (http://www.cs.cornell.edu/~mirek/Projects/FunctionSummaries). This will enable a broad audience, from researchers to land managers or bird watchers, teachers or school children to derive novel knowledge from the data resources gathered.