Page 55 - untitled

ੵcూ

Chang, Fu ਿ ͉ ༟ ࣘ Research Description
Research Description
ਿ ͉ ༟ ࣘ
Machine learning is the major line of my re- effects than almost all instance-based learning algo-
ᔖcc၈j ਓ޼Ӻࡰ
Associate Research Fellow (1990/6--) search in recent years. The new methods that I have rithms. Among the applications of machine learning
௰৷ኪዝj Ph.D., Mathematical Statistics, Columbia achieved in this field include adaptive prototype methods, we have obtained either new or improved
University (1983) learning (APL) algorithms, acceleration of large- results in various problems, including recognition
ཥcc༑j+886-2-2788-3799 ext. 1819 scale vector matching by means of various decom- of machine-printed or handwritten characters, bi-

ෂccॆj+886-2-2782-4814 position techniques, and data reduction methods. narization of document images with non-uniform Research Fellows
The APL algorithms prove to be very effective luminance, document layout analysis, classifi cation
ཥɿڦᇌjfchang@iis.sinica.edu.tw
methods for multiclass classification benchmarks of ethnic background by means of genetic markers,
ၣccࠫjhttp://www.iis.sinica.edu.tw/pages/fchang

and applications, and can also be used as data reduc- classification of neural signals, facial attractiveness,
tion methods, achieving significantly better accuracy etc.

• AT&T Bell Laboratories (1983~1990),
Research Fellows
• Ph.D., Mathematical Statistics, Columbia
Selected Publications
University (1983), ޼Ӻᔊʧ Selected Publications
޼Ӻᔊʧ
• M.S., Mathematics, North Carolina State University 1. J. J. W. Chen, R. Wu, P. C. Yang, J. Y. Huang, Y. P. Sher, M. H. Han, on Machine Learning and Cybernetics, Xian, 2003.
(1978), ዚኜኪ୦݊ҢڐϋԸ˴ࠅٙ޼Ӻ˙Σfίவ W. C. Kao, P. J. Lee, T. R. Chiu, F. Chang, Y. W. Chu, C. W. Wu, and 13. F. Chang, C.-H. Chou, C.-C. Lin, and C.-J. Chen, A Prototype Classifi -
K. Peck, Profi ling Expression Patterns and Isolating Differentially Ex- cation Method and Its Application to Handwritten Character Recogni-
• B.A., Philosophy, National Taiwan University ࡈჯਹ༁dҢ௴ிٙอ˙جܼ̍jሜቇ׌ࡡۨኪ୦
pressed Genes by cDNA Microarray System with Colorimetry Detec- tion, IEEE SMC, pp. 4738-4743, Hague, 2004. (EI)
(1973). tion, Genomics, Vol. 51, pp. 313-324, 1998. (SCI) 14. F. Chang, C.-C. Lin, and C.-J. Chen, Applying A Hybrid Method To
ၑۆdɽۨΣඎˢ࿁ٙ̋஺˙جd˸ʿ༟ࣘᏀᐵ˙
2. F. Chang, Y. C. Lu, and T. Pavlidis, Feature Analysis Using Line Handwritten Character Recognition, Intern. Conf. Pattern Recognition
Sweep Thinning Algorithm, IEEE Trans. Pattern Analysis and Ma- 2004, vol. 2, pp. 529-532, Cambridge, 2004. (EI)
جfሜቇ׌ࡡۨኪ୦ၑۆίεᗳйʱᗳٙᏐ͜ၾ಻
chine Intelligence, Vol. 21, No. 2, pp. 145-158, 1999. (SCI, EI) 15. F. Chang, C.-J. Chen, and C.-J. Lu, A Linear-Time Component-La-
3. F. Chang, K. H. Liang, T. M. Tan, and W. L. Huang, Document Image beling Algorithm Using Contour Tracing Technique, Computer Vision
༊ɪᗇ׼݊ڢ੬Ϟɢٙ˙جdϾ˲̙˸͜Ը຅Ъ༟
Binarization Using Hadamard Multiresolution Analysis, Intern. Conf. and Image Understanding, vol. 93, no. 2, pp. 206-220, 2004. (SCI, EI)
Document Analysis and Recognition, Banglore, 1999. (EI) 16. Y.-H. Liu, C.-C. Lin, and F. Chang, Language Identification of Char-

ࣘᏀᐵٙ˙جdίၚᆽٙࣖ؈ˢɽεᅰ˸ࡈԷމਿ
4. F. Chang and T.-R. Chou, Problems and Solutions for a Content Cap- acter Images Using Machine Learning Techniques, Inter. Conf. Docu-
th
ture and Information Retrieval System, 4 IAPR International Work- ment Analysis and Recognition, Seoul, 2005. (EI)
ᓾٙኪ୦˙جމԳfίዚኜኪ୦ٙᏐ͜ɪdҢࡁί
shop on Document Analysis Systems, Rio de Janeiro, 2000. 17. C.-Y. Su, A. Lo, C.-C. Lin, F. Chang, and W.-L. Hsu, "A Novel Ap-
5. F. Chang, Retrieving Information from Document Images: Problems proach for Prediction of Multi-Labeled Protein Subcellular Localiza-
஢εਪᕚɪᐏ՟ə࿃อאҷԄٙഐ؈dܼ̍ΙՏא
and Solutions, Intern. J. Document Analysis and Recognition, Special tion for Prokaryotic Bacteria," IEEE The Computational Systems
Issues on Document Analysis for Office Systems, vol. 4, no. 1, pp. Bioinformatics Conference, Stanford, 2005.
˓ᄳ˖οٙ፫ᗆdΈܓʔɓߧٙ˖΁ᅂ྅ʘɚЍ
46-55, 2001. 18. C.-H. Chou, C.-C. Huang, W.-H. Lin, and F. Chang, Learning to Bina-
6. F. Chang, Document Layout Analysis as Solving Multiple Constraints rize Document Images Using a Decision Cascade, IEEE Intern. Conf.
ʷd˖΁રوʱؓd˸፲ෂᅺႦԸਜʱૄ໊ߠ౻d
Problem, Workshop for Document Layout Interpretation and Its Ap- Image Processing, Genova, 2005. (EI)
plications, Seattle, 2001. 19. Y.-H. Liu, C.-C. Lin, W.-H. Lin, and F. Chang, Fast Vector Matching
ग़຾ৃ໮ٙʱᗳdᑕˆቾɢdഃഃf
7. F. Chang and J. D. Luo, Network Catastrophe: A Distributive Model Methods and Their Applications to Handwriting Recognition, Inter.
for Collective Phenomena, Journal of Mathematical Sociology, vol. Conf. Document Analysis and Recognition, Seoul, 2005. (EI)
26, no. 1-2, pp. 93-110, 2002. (SSCI) 20. F. Chang, S.-Y. Chu, and C.-Y. Chen, Chinese document layout analy-
8. F. Chang, C.-C. Huang, S.-Y. Chu, Recognition of Punctuation in Tra- sis using adaptive regrouping strategy, Pattern Recognition, vol. 38,
ditional Chinese Documents, IPPR Conference on Computer Vision, pp. 261-271, 2005. (SCI, EI)
Graphics, and Image Processing, Kinmen, Taiwan, 2003. 21. F. Chang and G.-C. Chen, C.-C. Lin and W.-H. Lin, Caption analysis
9. S.-L. Yeh and F. Chang, Role of Configurational Structure in Chinese and recognition for building video indexing systems, ACM Multimedia

Character Recognition, ICCS/ASCS-2003 Joint International Confer- Systems Journal, vol. 10, no. 4, pp. 344-355, 2005. (SCI, EI)
ence on Cognitive Science, Sydney, 2003. 22. C.-H. Chou, C.-C. Lin, Y.-H. Liu., and F. Chang, A Prototype Classifi -
10. F. Chang, C.-C. Lin, and W.-H. Lin, Learning to Identify Patterns, cation Method and Its Use in A Hybrid Solution for Multiclass Pattern
ICCS/ASCS-2003 Joint International Conference on Cognitive Sci- Recognition, Pattern Recognition, vol. 39, no. 4 , pp. 624-634, 2006.
ence, Sydney, 2003. (SCI, EI)
11. F. Chang and C.-J. Chen, A Component-Labeling Algorithm Using 23. C.-H. Chou, S.-Y. Chu, and F. Chang, Estimation of skew angles for
th
Contour Tracing Technique, 7 International Conference on Document scanned documents based on piecewise covering by parallelograms,
Analysis and Recognition, Edinburgh, 2003. (EI) Pattern Recognition, to appear. (SCI, EI)
12. F. Chang, C.-C. Lin, and W.-H. Lin, A Two-Stage Classification
nd
Method for Vector Pattern Matching Problems, The 2 Intern. Conf.

44 45