I recently founded the company, Yottamine Analytics, to provide software products and consulting services on predictive analytics. My research interests are machine learning and their application to real world problems such as traffic predictions,  digital advertisement, text classification, time series analysis, handwritten recognition, and bioinformatics. Prior to Yottamine Analytics, I was a research scientist at Microsoft and the senior scientist at INRIX where I was specialized in applying my research result to commercial applications, in particular large-scale web classification and real-time traffic prediction. Specifically,  I have been developing algorithms for support vector machines and graph-based semi-supervised learning for solving large-scale problems.  If you want to know more about me, my full CV is available in PDF format.  I am the first author of the monograph  Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning which is published by Springer. To the best of my knowledge, it was the first book that treats the fields of supervised, semi-supervised and unsupervised machine learning in a unifying way. If you would like to know more about it, you can also visit the site of the book http://www.learning-from-data.com/.

Software

• SemiL: The software SemiL is the first program that implements graph-based semi-supervised learning techniques for large-scale problems.
• ISDA: A  support vector machines tool with a nice GUI for solving large-scale classification and regression problems. 
• ^new  LinearSVM: The newest extremely fast machine learning (data mining) algorithm for solving multiclass classification problems from ultra large data sets that implements an original proprietary version of a cutting plane algorithm for designing a linear support vector machine. LinearSVM is a linearly scalable routine meaning that it creates an SVM model in a CPU time which scales linearly with the size of the training data set. 

Publications

Book

1. Huang, T.-M., Kecman, V. and Kopriva, I., Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning, Series: Studies in Computational Intelligence Vol. 17, Springer, 2006.

Book Chapters

2. Kecman, V., Huang T.-M., Vogt M., Iterative Single Data Algorithm for Training Kernel Machines from Huge Data Sets: Theory and Performance, Chapter in 'Support Vector Machines: Theory and Applications, Ed. Wang, L., Series: Studies in Fuzziness and Soft Computing, Springer Verlag, Vol. 177, pp.255-274, 2005.

Journal Papers

3. Huang, T.-M., Kecman, V., Gene Extraction for Cancer Diagnosis using Support Vector Machines- An Improvement,  Artificial Intelligence in Medicine (2005) 35, pp.185-194, Special Issue on Computational Intelligence Techniques in Bioinformatics, 2005. Top 25 Articles (Jul -Sep 2005) within AI in Medicine.

4. Huang, T.-M., Kecman, V., Semi-supervised Learning from Unbalanced Labeled Data – An Improvement, 'International Journal of Knowledge-Based and Intelligent  Engineering Systems', Special Issue: Innovational Soft Computing,  IOS Press,  Vol 10., No. 1, pp. 21-27, 2006.

Conference Papers

5. Huang, T.-M., Kecman, V., Performance Comparisons of Semi-Supervised Learning Algorithms. Proceedings of the Workshop on Learning with Partially Classified Training Data, at the 22^nd International Conference on Machine Learning, ICML 2005, W5,pp 45-49, Germany, 2005.

6. Huang, T.-M., Kecman, V., Gene Extraction for Cancer Diagnosis using Support Vector Machines: An improvement and comparison with nearest shrunken centroid method. Lecture Notes in Computer Science 3696, pp. 617-624, 2005.

7. Huang, T.-M., Kecman, V., Semi-supervised Learning from Unbalanced Labeled Data – An Improvement, 'Knowledge Based and Emergent Technologies Relied Intelligent Information and Engineering Systems', Eds. Negoita, M. Gh., at al., Lecture Notes in Computer Science 3215, pp. 765-771, Springer Verlag, Heidelberg, 2004. Best Paper Award and Best Student Contribution Paper Award.

8. Huang, T.-M. Kecman, V., Bias Term b in SVMs Again, 12^th European Symposium on Artificial Neural Network, ESANN 2004, pp. 441-448, Bruges, Belgium, April 28-30, 2004.

9. Kecman, V., Vogt, M., Huang, T.-M., On the Equality of Kernel AdaTron and Sequential Minimal Optimization in Classification and Regression Tasks and Alike Algorithm for Kernel Machines, 11^th European Symposium on Artificial Neural Networks, ESANN 2003, pp. 215-222, Bruges, Belgium, April 23-25, 2003. 

10.  Huang, T.-M. and Flay, R. G. J., A Comparison of Recursive Filter and Spectral Methods for Digital Corrected Pressure Measurements Distorted by Tubing Response, 5^th UK Wind Engineering Society Conference Nottingham, 2002.  

 For any question or comment, please feel free to contact me via email:

Copyright Huang © 2005-2006 All Rights Reserved