From Computer Data to Human Knowledge: A Cognitive Approach to Knowlege Discovery and Data Mining

这项研究是与IIS-9732562一起完成的，该研究由佐治亚理工学院心理学系的多丽特·比尔曼完成。这项研究关注的是可以通过数据挖掘技术开发的智能决策辅助工具。经验已经确定，这种系统可以学习准确的模型，但在决策辅助中使用这些模型的领域的专家往往不愿信任它们，因为它们不使用相同的测试，例如，专家已经逐渐信任的中间结论或抽象。或者他们根本不使用专家认为相关的某些因素。专家们还希望模型在被分析数据的微小变化下是稳定的。心理学家已经发现了一些因素，可以简化人类对类别和信息的学习、理解和交流。这项研究试图根据现有KDD算法的输出来探索心理学原理，然后开发和评估新的KDD算法，这些算法将提供便于人们学习、使用和与他人交流的输出。有了这项研究的结果，应该可以使这种辅助决策变得更以人为中心，以便在实践中更频繁和更有效地使用它们。http://www.ics.uci.edu/~pazzani

This research is being done in conjunction with IIS-9732562, which is beingperformed by Dorrit Billman in the Psychology Department at GeorgiaInstitute of Technology. The research is concerned with intelligentdecision aids that can be developed by data mining techniques. Experiencehas determined that such systems can learn accurate models, but thatexperts in areas where those models are used in decision aids are oftenreluctant to trust them because they do not, for instance, use the sametests intermediate conclusions or abstractions that the experts have grownto trust. Or they do not use certain factors at all that experts feel tobe relevant. Experts also want models that are stable under small changesin the data being analyzed. Psychologists have discovered factors thatsimplify the learning, understanding, and communication of category andprocess information by humans. This research seeks to explore thesepsychological principles in light of the output of existing KDD algorithmsand then go on to develop and evaluate new KDD algorithms that will provideoutput that is easy for people to learn , use, and communicate to others.With the results of this research, it should be possible to make suchdecision aids more "human centered", so that they will be used more oftenand more effectively in practice.http://www.ics.uci.edu/~pazzani