AbstraktyExperiences with KDD projects
across the EU (anglicky, bez překladu) The market for knowledge discovery in Europe is just starting to pick up speed. In this talk I will give an admittedly incomplete (due to the sensitive nature of most data mining projects) and somewhat research-oriented overview of successful data mining projects conducted in European countries. Data mining applications range from mailing campaign optimizations to the discovery of carcogenic compounds. It is interesting to note that despite the huge variety of available data mining tools, most users employ only a few selected methods and strongly prefer relatively simple methods. Furthermore, many users knowingly ignore or overrule basic rules of statistics or machine learning in their quest for knowledge. I will conclude the talk with an overview of specialized KDD tools that reflect one of the basic lessons learned by tool vendors: KDD tools need to be integrated into business processes, and must be tailored to specific needs since there is no, and will be no, generic KDD tool that satisfies the needs of all applications. Data Mining a Oracle
|
||
Stránky pro vás připravil Jirka Kosek. Připomínky a dotazy ke stránkám směřujte na adresu badame@vse.cz. Poslední modifikace . |