U.P.B. Sci. Bull., Series C, Vol. 71, Iss. 2, 2009 ISSN 1454-234x ON-LINE ANALYTICAL MINING FOR ADVANCED BUSINESS INTELLIGENCE SOLUTION Marius GURAN 1 , Aref MEHANNA 2 , Bilal HUSSEIN 3 Mineritul de date (DM) se referă la procesul de căutare şi descoperire de paternuri în colecţiile de date pentru a le prezenta într-o formă cât mai utilă şi predictibilă pentru procesele decizionale. Prin DM se poate descoperi cunoaşterea care nu se poate obţine prin metode clasice de tip OLAP. În lucrare se propune un model OLAP pentru DM, numit OLAM (OLAP Mining), util în exploatarea unor Baze de Date (DB) şi Depozite de Date (DW) pentru aplicaţii de tipul BI (Business Intelligence). Data mining (DM) deals with the proactive process of searching and discovering the patterns in the data and present it in a more predictable and useful format. It discovers the hidden knowledge within the information which is not possible with traditional relational or on-line analytical processing (OLAP) technologies. OLAM (OLAP Mining) model, which is based on the influence domain, is proposed in this paper. An exploratory analysis can be made in large DBs or DWs, based on OLAM, in BI applications. Keywords: Business Intelligence (BI), Data Base (DB), Data Warehouse (DW), Data Mining (DM), On-Line Analytical Processing (OLAP), On-Line Analytical Mining (OLAM) 1. Introduction Identifying the patterns in the business data is a required step for the analysts to take right decisions at the right time for business applications. For this reason there is more insight on the need for DM for completing OLAP in BI applications. Relational and OLAP tools are good for finding and reporting information that is already existing and visible within the data. Hidden knowledge that represents patterns and regularities in data can’t be easily found by traditional reporting. In Relational and OLAP cases the selection and grouping criteria are known in advance. DM provides a powerful solution when the specific selection and grouping criteria are not known in advance, but are derived from the 1 Professor, University POLITEHNICA of Bucharest, ROMANIA, e-mail: mguran@mix.mmi.pub.ro 2 PhD student, University POLITEHNICA of Bucharest, Romania. 3 PhD student, Université de Valenciennes-LAMIH, France