Telecommunication Case Modelling – Call Center C.Chudzian, J.Granat, W.Traczyk Decision Support Systems Laboratory National Institute of Telecommunications Warsaw, Poland

2 Data Mining in Practice, Dortmund, 18th February 2003 The goal Selecting prospective clients for targeting a marketing campaign based on the existing data (call center data, billing data and others).

3 Data Mining in Practice, Dortmund, 18th February 2003 Model of the call center

4 Data Mining in Practice, Dortmund, 18th February 2003 The data is distributed Call center Switch Target group of clients Analysis

5 Data Mining in Practice, Dortmund, 18th February 2003 Process of building a table for data mining

6 Data Mining in Practice, Dortmund, 18th February 2003 Billing Call Center Classification of clients Switch data No List of clients YES

7 Data Mining in Practice, Dortmund, 18th February 2003 SAS Enterprise Miner

8 Data Mining in Practice, Dortmund, 18th February 2003 MiningMart requirements

9 Data Mining in Practice, Dortmund, 18th February 2003 The set of operators

10 Data Mining in Practice, Dortmund, 18th February 2003 Client Features III Client Features II Preprocessing (part I) Process details for all clients UnSegment Process details for a client SpecifiedStatistics Segmentation Features (I,II,III) for all clients Client Features I RowSelectionByQuery SpecifiedStatistics TimeIntervalManualDiscretization RowSelectionByQuery SpecifiedStatistics UnSegment JoinByKey Features I for all clients Features II for all clients Features III for all clients

11 Data Mining in Practice, Dortmund, 18th February 2003 Preprocessing (part II) Features(I,II,III) for all clients GenericFeatureConstruction All features for all clients Call Center Data Service users UnionByKey Service info GenericFeatureConstruction Service info (switch prefered) UnionByKey Mining Concept

12 Data Mining in Practice, Dortmund, 18th February 2003 NIT case - HCI view

13 Data Mining in Practice, Dortmund, 18th February 2003 Business data

14 Data Mining in Practice, Dortmund, 18th February 2003 NIT case – preprocessing steps

15 Data Mining in Practice, Dortmund, 18th February 2003 Conclusions Conceptual modeling improves: the understanding of the data preparation process maintenance of the data preparation process Knowledge transfer for other people We do not need to use programming language

16 Data Mining in Practice, Dortmund, 18th February 2003 Questions & discussion ?