Non-invasive data analysis (NiDA) Resource Page

Under construction!

Data models and model assumptions

• Data Mining and Warehousing Survey
• Data Mining Software (includes Rough Set Software)
• Zadeh's programmatic paper about soft modeling

Rough Set Data Analysis

• Electronic Bulletin of Rough Set Community
• International Rough Set Society (IRSS)
• ROSETTA: A software system for Rough Set Analysis

  ---

• Algebraic aspects of attribute dependencies in information systems, Fundamenta Informaticae, 29(1997), 119 - 133, 1997.
  As the titel says.
• GROBIAN. In: L. Polkowski & A. Skowron (Eds.) Rough sets in knowledge discovery. Vol 2. Heidelberg: Physika-Verlag, 555 - 557, 1998.
  Introducts GROBIAN, a software for basic RSDA.
• Rough Approximation Quality Revisited, submitted.
  There is more than one way to define reasonable approximation qualities measures in RSDA!

Rule Significance

• Statistical Evaluation of Rough Set Dependency Analysis, International Journal of Human-Computer Studies, 46 (1997), 589 - 604, 1997.
  The casuality problem is discussed and it is shown how to test rules against casuality.
• Statistical tools for rule based data analysis. In L. Polkowski, S. Tsumoto & T.Y.Lin (Eds.) Rough Set Methods and applications. New Developments in Knowledge Discovery and Applications. Chap. 10. Heidelberg: Springer, 2000.
  It is neccessary to treat results of RSDA statistically. The paper shows how and why.

Discretisation

• Classificatory Filtering in Decision Systems, International Journal of Approximate Reasoning, 23, 111-136, 2000.
  One can call it a lattice machine which filter the data in an optimal way.
• Simple data filtering in rough set systems, International Journal of Approximate Reasoning, 18, 93 - 106, 1998.
  It says it's a simple approach - and simple is sometimes best!

Model selection

• Uncertainty measures of rough set prediction, Artificial Intelligence, 106, 109 - 137, 1998.
  If you want to know what Rough Entropy looks like, you should consult the paper.

Probabilistic granule analysis

• Statistical tools for rule based data analysis. In L. Polkowski, S. Tsumoto & T.Y.Lin (Eds.) Rough Set Methods and applications. New Developments in Knowledge Discovery and Applications. Chap. 10. Heidelberg: Springer, 2000.
  It is neccessary to treat results of RSDA statistically. The paper shows how and why.

Imputation

• Maximum consistency of incomplete data via non-invasive imputation, submitted.
  A Rough Set Approach to estimate missing data.

Beyond Rough Sets

• Relational attribute systems, submitted.
  Shows that spatial resoning and the evaluation of observer agreement have something in common.
• Approximation quality for sorting rules. Submitted.
  Extends the approach of RSDA to ordinal attributes.
• Knowledge structures and their applications in CALL In: Language Teaching and Language Technology , Sake Jager, John Nerbonne and Arthur van Essen (Eds.). Lisse: Swets and Zeitlinger, 177 - 186, 1998.
  A lattice based approach to represent knowledge.

Relevant Journals

• AutoSoft
• International Journal of Approximate Reasoning

Active Researchers (we hope, they don't mind!)

• Ivo Düntsch (University of Ulster; Northern Ireland)
• Günther Gediga (Institute for Evaluation & Market Research; Germany)
• Tsau Young Lin (San Jose State University; USA)
• Andrej Skowron (Institute of Mathematics, Warshaw; Poland)
• Roman Slowinski (Poznan University of Technology; Poland)
• Yiyu Yao (University of Regina; Canada)
• Hui Wang (University of Ulster; Northern Ireland)
• Wojciech P. Ziarko (University of Regina; Canada)

Further Düntsch-Gediga contributions to NiDA

• Archetypal psychiatric patients: An Application of Rough Information Analysis. Submitted.
  An application study in the field of clinical psychology.
• Logical tools for rule based data analysis. In L. Polkowski, S. Tsumoto & T.Y.Lin (Eds.) Rough Set Methods and applications. New Developments in Knowledge Discovery and Applications. Chap. 9. Heidelberg: Springer, 2000.
  One certainly should know what kind of logics are connected to the Rough Set Approach.
• Rough set data analysis: A road to non-invasive knowledge discovery. Bangor: Methodos, 2000.
  An overview of the ID/GG approach to RSDA.
• Sets, Relations, Functions. Bangor: Methodos, 2000.
  Nothing about RSDA, but the book offers a modern approach to the basic tools of mathematical thinking.
• IRIS revisited: A comparison of discriminant and enhanced rough set data analysis. In: L. Polkowski & A. Skowron (Eds.) Rough sets in knowledge discovery. Vol 2. Heidelberg: Physika-Verlag, 345 - 368, 1998.
  The Fisherian IRIS data can be used for RSDA as well - results are not bad!
• Relation restricted prediction analysis, In: A. Sydow (Ed.) 15th IMACS World Congress, Vol. 4: Artificial Intelligence and Computer Science, Wissenschaft & Technik Verlag, Berlin, 619 - 624, 1997.
  A first attempt to ordinal predicition with RSDA.