July 2005 @ Amsterdam

Kai-min Kevin Chang Watermelon Ph.D. student, Language Technologies Institute, School of Computer Science, Carnegie Mellon University. Affiliated with Center for Cognitive Brain Imaging, Center for the Neural Basis of Cognition Profile:CV, resume Email: kaimin dot chang at gmail dot com Phone: +1-412-268-3414 Fax: +1-412-268-6298 Office: Baker Hall 327B Address: Language Technologies Institute Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213-3891 USA

Research Interests

My research interests include using mathematical methodologies and machine learning techniques to investigate and model various human cognitive processes. In particular, I have studied semantic presentation of objects using functional Magnetic Resonance Imaging, knowledge representation in the context of an Intelligent Tutoring System (ie, student models) and language processing in the connectionist framework. I am currently working in Center for Cognitive Brain Imaging, where I am supervised by Dr. Marcel Just and Dr. Tom Mitchell.

Supervised by Dr. Marcel Just and Dr. Tom Mitchell, I used functional Magnetic Resonance Imaging to study the cortical systems that underpin semantic representation of object knowledge. In a picture-naming task, participants were presented with black and white line drawings of 60 objects from a range of categories (e.g. tools, dwellings, animals, etc.) and were instructed to think of the same properties consistently during each presentation. I modeled the picture-naming task with a probabilistic generative model (i.e. Bayes nets). The proposed Bayes nets assume the observed cortical activation pattern is generated by both the object stimulus, and a set of hidden processes that represent the semantic knowledge of object. The semantic representation of object knowledge was derived from a set of distinguishing sensory/functional features (e.g. visual-motion, visual-parts, function, etc.) according to Cree & McRae (2003)'s semantic feature norming studies. For example, the word "hammer" has a higher value in the "function" feature, compared to the word "cat," because people tend to recall more properties that are related to function when they think about hammers. I trained the Bayes net using the Expectation Maximization algorithm and cross-validated across 9 subjects. The performance of my model is evaluated in two folds. In discriminative evaluation, the 60-way exemplar classification accuracy is computed given the observed cortical activation (i.e., reading the mind). An accuracy of 0.35 is obtained for the best performing subject and 0.2167 across subjects on average. Both classification accuracies are significantly different from a random guessing of 0.0167. In generative evaluation, we use the model to generate predicted activation pattern for reading a word (i.e., predict brain activation) and compute the correlation between the predicted cortical activation pattern and human's cortical activation pattern reading the same word. A correlation of 0.7219 is obtained for the best performing subject and 0.5228 across subjects on average.

Intelligent tutoring systems derive much of their power from having a student model that describes the learner's competencies. However, constructing a student model is challenging for computer tutors that use automated speech recognition (ASR) as input, due to inherent inaccuracies in ASR. Under the supervision of Dr. Jack Mostow and Dr. Joseph Beck, I proposed two models of developing word decoding skills and demonstrated that sufficient information existed in ASR output to determine which model better fits student performance and under what circumstances (Chang et al., 2005). Moreover, we found modeling individual learners' proficiencies may enable improved speech recognition in a computer tutor (Beck et al., 2005). In the above work, we used Knowledge Tracing, a derivative of Atkinson's model (1972) of human memory, to trace student's knowledge across different skills. We then followed Reye's work (1998), which proved that Knowledge Tracing is a special case of a Bayesian network, and implemented a generic Bayesian network toolkit (BNT-SM; Chang et al., 2006) for student modeling. BNT-SM inputs a data set and a compact XML specification of a (dynamic) Bayes net model hypothesized by a researcher to describe causal relationships among student knowledge and observed behavior. It generates and executes the code to train and test the model using the Bayes Net Toolbox (Murphy, 1998). BNT-SM allows researchers to easily explore different hypothesis with respect to the knowledge representation in a student model. For example, by varying the graphical structure of a Bayesian network, we examined how tutoring intervention can affect students' knowledge state - whether the intervention is likely to scaffold or to help students to learn.

I have also worked on low-level computational models of reading, in which cognitive processes are implemented in terms of cooperative and competitive interactions among large numbers of simple, neuron-like processing units. Contemporary leaders in computational model of reading are divided in whether a localist or distributed representation is more appropriate. For example, localist representation assumes activation of a word to be a corresponding lexical unit, whereas distributed representation assumes such to be a pattern of activation distributed over a number of primitive representational units. Fortunately, I have the opportunity to work in both camps. For the localist representation, I worked with Dr. Derek Besner to challenge one of the fundamental assumption of DRC, a leading computational model of reading which assumes a Dual Route model and that information processing occurs in a Cascaded fashion within the model. In order to have the model correctly reproduces the joint effects of letter length and stimulus quality seen in skilled readers, I implemented a threshold at the letter level in DRC, as part of my undergraduate thesis. For the distributed representation, I worked with Dr. David Plaut to address two tasks in the reading model of parallel distributed processing (PDP) framework: 1) learning static representations of variable-length strings, and 2) generating continuous articulatory trajectories as output. These two tasks are fundamental extensions of PDP modeling of word reading, enabling models to process multi-syllabic words and to generate more realistic analogues of human response time.

Software Tools

Publication

Mitchell, T.M., Shinkareva, S.V., Carlson, A., Chang, K.M., Malave, V.L., Mason, R.A., & Just, M.A. (2008). Predicting Human Brain Activity Associated with Noun Meanings. Science (pdf)

Beck, J.E., Chang, K.M., Mostow, J., & Corbett, A. (2008). Does help help? A comparison of three evaluation frameworks. Intelligent Tutoring Systems. Best Paper Award

Chang, K.M., Malave V., Shinkareva, S., Mitchell, T.M., & Just M.A. (2007). What functional brain imaging reveals about neuroarchitecture of object knowledge. Poster Presenation at the 1st Okinawa Instistute of Science and Technology Workshop on Cognitive Neurobiology. (ppt)

Beck, J.E. & Chang, K.M. (2007). Identifiability: A Fundamental Problem of Student Modeling. Proceedings of the 11th International Conference on User Modeling. Corfu, Greece. (pdf)

Chang, K.M., Beck, J.E., Mostow, J., & Corbett, A. (2006). Does Help Help? A Bayes Net Approach to Modeling Tutor Interventions. Proceedings of the AAAI2006 Workshop on Educational Data Mining, 21st National Conference on Artificial Intelligence, Boston, MA. (pdf)

Chang, K.M., Beck, J.E., Mostow, J., & Corbett, A. (2006). A Bayes Net Toolkit for Student Modeling in Intelligent Tutoring Systems. Proceedings of the 8th International Conference on Intelligent Tutoring Systems, Jhongli, Taiwan. p. 104-113. (pdf)

Chang, K.M., Beck, J.E., Mostow, J., & Corbett, A. (2005). Using Speech Recognition to Evaluate Two Student Models for a Reading Tutor. Proceedings of the AIED 2005 Workshop on Student Modeling for Language Tutors, 12th International Conference on Artificial Intelligence in Education, Amsterdam, p. 12-21. (.pdf)

Beck, J.E., Chang, K.M., Mostow, J., & Corbett, A. (2005). Using a Student Model to Improve a Computer Tutor's Speech Recognition. Proceedings of the AIED 2005 Workshop on Student Modeling for Language Tutors, 12th International Conference on Artificial Intelligence in Education, Amsterdam, p. 2-11. (.pdf)

Chang, K.M. (2003). Evalution of the Dual Route Cascaded Model of Reading Aloud. Undergraduate Thesis. (.pdf)

Project

Carnegie Mellon University
10-701: Applying Fourier Transform and Linear Time Invariant Analysis to Brain Imaging Data (.rtf)
11-761: Distinguishing Tri-gram Generated Articles from True Articles (.rtf)
10-751: An Acoustic Model of the Water Noise in the Dolphin Project (.doc)
11-791: User Interface Prototype (MedlineKLMY)

University of Waterloo
CS 499: An Application of Neural Networks in Learning of RoboCup Soccer Playing Team Stragety (.pdf)
Stat 468: A Real-Coded Genetic Algorithm Approach to Data Segmentation Problem (.pdf)
CS 444: WCC Ada/CS Compiler Documentation (.pdf)

Professional

Biographical Sketch

Personal Interest

Some people write their diaries with words, some record them with pictures. I mark mine with food! Yes, I love to eat! My plan is taste all the savoury dishes in the world and mark them on my Savoury Google Maps! Still a long way to go, but I am getting there! :p I like to read the Sladhdot, the tw.bbs.talk.joke newsgroup, and watch Comedy Central on TV. Three comic strips that I frequently visit are Piled Higher and Deeper, Dilbert, and River's 543. For leisure, I enjoy playing chess and nine ball. I used to play snooker during my undergraduate years, though I think my aim has rotten since then. Finally, I treasure freedom in speech, thoughts, codes, and am an advocate of Open Source software. PS, If you noticed my title of Watermelon Ph.D. Student, kudos to you! But it is true! I was named the Watermelon Ph.D. Student according to this news article, originally published by University of Waterloo school officials on Apr 1, 2003. ;) Quite frankly, I joined Carnegie Mellon University later and that indeed made me a student of watermelon.