Table of Content

  1. Publication
  2. Programs
  3. Scripts
  4. Misc

Publication

Academic

  1. Chen, Ding-Yi, Xue Li, Zhaoyang Dong, and Xia Chen,  Chapter VIII: Incremental Learning for Interactive E-Mail Filtering, Agent Technologies and Web Engineering: Applications and Systems , edited by Ghazi Alkhatib and David Rine, IGI Publishing, pp. 134-152., 2009
  2. Song, Dawei, Raymond Y.K. Lau, Peter D. Bruza, K.F. Wong, Ding-Yi Chen Sequential Pattern Mining and Nonmonotonic Reasoning for Intelligent Information Agents. International Journal of Pattern Recognition and Artificial Intelligence. Vol.21, No.4, pages 773-790. World Scientific Publishing., 2007
  3. Chen, Ding-Yi, Xue Li, Zhao Yang Dong and Xia Chen. Fitness Assessment of Document Model, IJSS: International Journal of Systems Science, Vol 37. No 13, p. 893-904, Taylor & Francis Group, 2006.
  4. Chen, Ding-Yi, Xue Li, Zhao Yang Dong and Xia Chen. Incremental Learning for Interactive Email Filtering, IJITWE International Journal of Information Technology and Web Engineering, April-June, pp. 60-78, Idea Group Publishing Inc., 2006.
  5. Chen, Ding-Yi, Xue Li, Zhao Yang Dong and Xia Chen. 2005. Effectiveness of Document Representation for Classification. In DaWaK 2005: 7th International Conference on Data Warehousing and Knowledge Discovery. Copenhagen, Denmark: Springer-Verlag.
  6. Chen, Ding-Yi, Xue Li, Zhao Yang Dong and Xia Chen. 2005. Determining the fitness of a document model by using conflict instances. In ADC 2005: 16th Australasian Database Conference. Newcastle, Australia:125--134: Australian Computer Society Inc.
  7. Chen, Ding-Yi, Xue Li, Zhao Yang Dong, and Paul Smith. 2004. Interactive email filtering - learning from misclassified examples. In CIS 2004: IEEE Conference on Cybernetics and Intelligent Systems. Singapore:1060--1065: IEEE.
  8. Chen, Ding-Yi and Xue Li. 2004. PLD: A distillation algorithm of misclassified documents. In WAIM 2004: 5th International Conference on Advances in Web-Age Information Management, Lecture Note in Computer Science (LNCS) 3219:499--508: Springer-Verlag.
  9. Chang, Chin-Lung, Ding-Yi Chen, and Tyng-Ruey Chuang. 2002. Browsing newsgroups with a social network analyzer. In International Conference on Information Visualisation (Web Graphics and Visualisation).
  10. Chen, Ding-Yi, Tyng-Ruey Chuang, and Shi-Chun Tsai. 2001. JGAP: A java-based graph algorithms platform. Software-Practice and Experience 31, no. 7: 615-635.

News Paper

  • Ding-Yi Chen (2003), "Treasure Finding in Internet- A proper way to use search engines", Christian Tribune News. (Taiwan) (Traditional Chinese Encoding)
  • Ding-Yi Chen (2002), "Can you really make money on mobile commerce?", Network Technology. (Taiwan)(Traditional Chinese Encoding)

Blog

Note: My livejournal blogis now deprecated.

Programs

ucs4_sort (2008-09-05) in C

sort in Unix system is a wonderful tool, however, it does not handles utf8 files properly, because sort tends to treat them as ASCII strings. There is some solution like msort, but I never get it work.

I therefore write a small program, ucs4_sort, that sorts lines of text files by their UCS-4 encoding. This program DOES NOT implement the Unicode collation algorithm, so do not use the on locale sensitive data.

Installation

You need to have glib >= 2.0, gcc and pkg-config to be able to compile it.
  1. Download ucs4_sort.c.
  2. Compile it by
    gcc ucs4_sort.c -o ucs4_sort.c `pkgconfig --cflags --libs glib-2.0`
  3. Copy the executable ucs4_sort to the desired directory such as ~/bin

Usage

Name

ucs4_sort - sort lines of text files according to their UCS-4 representation

Synopsis

ucs4_sort [-u] <INPUT_FILE>

Description

Write sorted concatnation of the INPUT_FILE to standard output.

Examples

Sort the lines, and output to stdout.

ucs4_sort <input-file>

Sort the lines, and output only the unique lines to stdout

ucs4_sort -u <input-file>

Copyright

This program is release under the BSD licence.


Text Mining Package

See the here for program lists. Sorry, no further documentation.

Document Classifier Examination Package

The package contains 2 parts, one is the program itself, the other is a set of corpuses.

Installation Instruction

For Windows (we use c:\ as example of installation path):

  1. Extract DocumentClassifier*.zip to c:\
  2. Extract corpus.zip to c:\
  3. add c:\DocumentClassifier\bin to environment variable CLASSPATH

Usage

java au.edu.uq.itee.Classifier.Framework.LearningFromMistakeFrameworkTester corpusPath corpusName coreAlgorithm

corpusPath
Path of document corpus
corpusName
Name of document corpus. Available corpus: reuters,ohsumed
coreAlgorithm
Assign classifier to be the core algorithm. Current available classifier: SVM,Winnow

Please note corpusName and coreAlgorithm are case sensitive.

Output files

The package will generate multiple output files in the directory corpusPath. Their names are formated as follows:

corpusName-resultK-coreAlgorithm-diligenceLevel.txt

For example, if we apply SVM on reuters corpus, then following files will be generated:

  • reuters-resultK-SVM-0.2.txt
  • reuters-resultK-SVM-0.4.txt
  • reuters-resultK-SVM-0.6.txt
  • reuters-resultK-SVM-0.8.txt
  • reuters-resultK-SVM-1.0.txt

Scripts

Web Page Menu (Javascript)

Are you interested the menu I am using?
Come to get these files (use right click to download):

Steps:

  1. Modify the menuDef for your own usage. Please, keep my name in those files as reference.
  2. Use one of my html files as example in order to learn how to apply the menu system.

Hopefully it should work.

Too simple, you are saying? Ok, I will write the document LATER. Comments are welcome, click here to mail me.

Sort algorithm demo applet (1998)

Misc

Presentation Slides

Fedora/RHEL kickstart files

My kickstart files are for my personal Fedora live CD, DVD, USB, and virtualization image.
  1. f14-dchen-base.ks
  2. f14-dchen.ks
For live CD and virtualization, only the first one is required; and for liveDVD, both files are required. The packages in live CD/DVD includes:
  • XFCE
  • Chinese input/display support
  • OpenOffice default install
  • joe, vim-X11
The packages in live DVD also includes:
  • GNOME, KDE, LXDE
  • packages from rpmfusion, both free and non-free
  • Japanese/Korean input/display support
  • OpenOffice-extra, Eclipse
  • Media players: audacious and plugins, vlc
  • opera
  • and many more

Use livecd-creator -c <kickstartFile> to make live DVD iso. If you also want live USB, use liveuse-creator and follow the UI.

Yum repo

My custom made repos includes: Put them in your /etc/yum.repo.d to use them.