We introduce a new method of sign language subtitling aimed at young deaf children who have not acquired reading skills yet, and can communicate only via signs. The method is based on: 1) the recently developed concept of “semantroidTM” (an animated 3D avatar limited to head and hands); 2) the design, development, and psychophysical evaluation of a highly comprehensible model of the semantroid;...

In text classification, dictionaries can be used to define human-comprehensible features. We propose an improvement to dictionary features called smoothed dictionary features. These features recognize document contexts instead of ngrams. We describe a principled methodology to solicit dictionary features from a teacher, and present results showing that models built using these human-comprehensi...

Predictive modeling applications have a number of concerns. While predictive accuracy is often the primary concern, other issues such as efficiency and robustness are often considered. It is my position that comprehensibility is another concern in predictive modeling applications, and is often an important one. However, without an understanding of comprehensibility in predictive modeling, it is...

The ability to intelligently collect, manage and analyze information about customers and sellers is a key source of competitive advantage for an e-business. This ability provides an opportunity to deliver real time marketing or services that strengthen customer relationships. This also enables an organization to gather business intelligence about a customer that can be used for future planning ...

This thesis is concerned with the problem of learning comprehensible theories from structured data and covers primarily classification and regression learning. The basic knowledge representation language is set around a polymorphically-typed, higherorder logic. The general setup is closely related to the learning from propositionalized knowledge and learning from interpretations settings in Ind...

Although neural networks have been used to develop highly accurate classifiers in numerous real-world problem domains, the models they learn are notoriously difficult to understand. This thesis investigates the task of extracting comprehensible models from trained neural networks, thereby alleviating this limitation. The primary contribution of the thesis is an algorithm that overcomes the sign...