1998 Fall EE 380L Data Mining

Unique # 15250


News

16 Nov 1998 Added Financial Time Series Data link and EXPO SE (Windows 95) time-series tool.
9 Nov 1998 Visualization Techniques for Mining Large Databases: A Comparison (local copy) is now free of errors.
29 Oct 1998 Updated How to create postscript files to include info on creating .ps files in ENS 507 PC

Course

Papers

  1. Primary List of Papers (References ONLY)
  2. Secondary List of Papers (Downloadable)
  3. Presentations

Term Paper

  1. Presentations

Informational Links

  1. White paper on OLAP
  2. Data Warehouse basics and links
  3. AUAI Tutorials
  4. Visualization

Free Software/Libraries

Links (original) Local download? (faster)
EXPO SE (Windows 95/NT)
MLC++ (NT version available) Yes 8 MB
3 MB (NT)
TOOLDIAG (DOS version available) Yes 376 KB
245 KB (DOS)
WEKA (includes some UCI datasets) Yes 4 MB
MOBAL Yes 2.5 MB
DBMiner 1.0e (NT/Win95 ONLY) Yes
11 MB (Win95)
12 MB (NT)
KDNuggets

Datasets

Links (original) Local download? (faster)
  1. Delve
  2. ELENA
  3. PRNN
  4. PROBEN1
  5. StatLib
  6. Statlog
  7. UCI Machine Learning databse
LANS benchmarks Datasets
KDD Sisyphus I Yes
KDNuggets
Financial Time Series Data
The Data Mine

Misc. Topics

  1. Principal Component Analysis (PCA)
  2. EM Algorithm (2-pg notes)
  3. GTM (Generative Topological Mapping)
  4. Hierachical Probabilistic PCA

Tools

  1. (Mac/PC) Expander for decompressing gzip files (.gz)
  2. Ghostview for reading postscript files (.ps)
  3. Acrobat Reader for reading acrobat files (.pdf)
  4. (Mac/PC) How to create postscript files