Continuing the series of Computer Business Review’s profiles of Europe’s high flyers we look at the prospects for DataMind.
DataMind Corp is a company borne out of the frustration of its founder Dr Khai Minh Pham. Pham was conducting medical research at the University of Paris into the success of transplant operations, but was being hindered by a lack of AI artificial intelligence tools, including decision trees and neural networks, to help him in his work. So, he went back to school, got a second degree in software development and created his own agent network technology. The result was promptly rejected by AI groups at the university who argued it did not fit into established AI models. But, without aiming to do so, Pham had invented the tool that would later become the core technology of DataMind, the company of which he is now chairman. Incorporated in 1994, DataMind was founded with the idea of helping business professionals to understand corporate data in more depth and help them make better business decisions. It was officially launched in April 1996 with $4.7m in venture capital backing, while a second round of backing brought in a further $7.1m in September 1996. Investors includes Atlas, Draper Fisher, Sofinnova and Sequoia. Unlike first- generation, rules-based software which uses a data verification model to implement system rules based on experience, DataMind’s products allow an automatic discovery process, fusing technologies like AI, induction and neural networks to predict tendencies. DataMind DataCruncher, the company claims, is the first client-server data mining engine able to analyse unlimited volumes of highly-complex data. The technology is in use at over 40 corporations worldwide including Chase Manhattan Bank and BellSouth.
Create demand for data mining
Currently, DataMind is focusing its efforts on delivering data mining systems to telecoms companies which experience customer revenue rates of 20% each year at a cost of more than $4bn. We believe mainstream business professionals will create the greatest market demand for data mining, says Eric Archambeau, the company’s president and chief executive, and a former executive of Sofinnova, one of DataMind’s investors. Critical to success in the data mining market, he says, will be the ability to integrate with and extend the functionality of companies’ existing data warehouse solutions. The Meta Group expects the market for data mining software and services will rocket from $60m in 1996 to $800m by the year 2000. As a result, DataMind is commanding the respect of the likes of Hewlett-Packard Co and Arthur Andersen. It also has strategic partnerships with Arbor Software Corp, Business Objects SA and Red Brick Systems Inc. DataMind’s key competitors are IBM Corp and NeoVista Corp, a California-based data mining software start-up. Hewlett-Packard Co, which has adopted DataMind into its OpenWarehouse program, calls DataMind’s cross-breeding of technology a ‘fresh approach’. While the company, which has its headquarters in Redwood City, California and regional sales and support centers in the US and France, refuses to disclose any revenue figures, it says an initial public offering is planned in the two years.
Content from our partners
