Move for Standard Data Mining Process Gathers Pace

First CRISP-DM SIG Workshop Held

Key players in the data mining market met in Amsterdam last week as the move towards a standard process model for data mining gained momentum.

The CRISP-DM initiative - "CRoss-Industry Standard Process for Data Mining" – is partially funded by the European Commission. Members include NCR, ISL, Daimler-Benz and OHRA.

Key to CRISP-DM is a Special Interest Group (SIG) of data mining tool and service suppliers and large-scale commercial users.

At the Amsterdam workshop, SIG members from eight countries heard the CRISP consortium outline the industry’s need for a common methodology and describe initial work towards a standard data mining process model. SIG members were invited to present their views. Speakers represented data mining vendors (including Syllogic, Data Distilleries and Attar Software), systems suppliers and integrators (including Cap Gemini and ICL Retail), management consultancies (including Deloitte & Touche and Price Waterhouse) and large-scale industrials (including British Telecom and ABB).

In the SIG presentations and discussions afterwards, several key points were stressed repeatedly. These were that:

  • users will not always be technology experts,

  • a methodology should provide a framework for documenting and re-using the experience gained on data mining projects,

  • business issues are as important as technology, and must be addressed by any process model.

The CRISP partners were delighted to find such a high level of agreement between the SIG members, and support for the aims of the initiative.

"This workshop has confirmed the need for a standard, cross-industry process model," said Jens Hejlesen of NCR, CRISP-DM project manager. "There was overwhelming agreement that data mining needs a common process model, and from the SIG members’ input we are confident that our work is going in the right direction. Most importantly, there is a recognition that the data mining market needs such a standard now if we are to see this technology adopted as infrastructure by Global 2000 companies."

Members are still being recruited for the CRISP SIG, and further workshops are planned during the next few months. An email discussion forum will also be established, and a newsletter will be published.