Redwood City, California-based startup DataMind Corp has its promised data mining products ready to roll (CI No 2,837). The process of data mining is the discovery of new relationships by sifting through huge amounts of data stored in a data warehouse. The idea is that organizations can make and recommend decisions based on the relationships they find. DataMind claims its tools are designed for use by business users, not statisticians and technicians and come with a graphical user interface that guides users through the mining process by first creating a study which defines the scope of a search or automated data modelling. Excel spreadsheet and graphics charts or Word documents are used to display the model and predictions Web browser and PowerPoint front-ends will come later. DataMind Professional Edition runs on Windows95 and 3.1 clients, includes an Object DataBase Connectivity interface and is designed to run against a data warehouse with millions, rather than billions of records. It costs $5,000. A DataMind Solo Lite version costs $2,500 and is designed for organizations with up to 20 users that have data stored in local spreadsheets and text files. It has fewer variables and no Object DataBase Connectivity interface. The $25,000 DataMind DataCruncher server enables users to offload desktop data mining functions and automatically batch process regular data mining models, supporting Unix and NT. DataMind will sell direct and via partners it is now looking to recruit.
