Icon

Basic Customer Segmentation Use Case

Basic Customer Segmentation
Customer SegmentationCustomer segmentation is the sub-division of a market into discrete different groups of customers, where each group shares similar characteristics. This workflow illustrates how to build a basiccustomer segmentation model, using a clustering procedure.Task Build a basic customer segmentation using a clustering procedure. For a more extensive analysis, check the whitepaper "Customer Segmentation comfortably from a Web Browser" at the following URL: https://www.knime.org/files/white-papers/customer_segmentation.pdf Pre-processing - Join contract data and behavioral data - Convert Churn values to String to be used as class in upcomingclassification - Normalize all numerical columns in [0,1] ClusteringClustering is performed with k-Means. Other Learner nodes train other models. MostLearner nodes output a PMML model (blue square output port). Input data withassigned cluster Cluster centers Data Reading 2 files: - contract data - behavioral (calls) dataBoth files are located in TheData/Customers Try this:Clustering Algorithms1) Choose your clustering technique and placeit here. k-Means here is just an example. 2) Open the view of the k-Means node to seecluster prototypes.Denormalization1) After executing the k-Means, denormalize thedata to see values in their original range. 10 clusterson all numericalinputsNormalize allnumerical columnsto fall in [0,1]Back to originaldata rangeBack to originaldata rangeconverting a numerical column to Stringexcludes it from the clustering procedureExclude area code and churn from subsequent clusteringReadingContractData.csvReadingCallsData.xlsJoin calls dataand contract data k-Means Normalizer (PMML) Denormalizer (PMML) Denormalizer (PMML) Number To String CSV Reader Excel Reader Joiner Customer SegmentationCustomer segmentation is the sub-division of a market into discrete different groups of customers, where each group shares similar characteristics. This workflow illustrates how to build a basiccustomer segmentation model, using a clustering procedure.Task Build a basic customer segmentation using a clustering procedure. For a more extensive analysis, check the whitepaper "Customer Segmentation comfortably from a Web Browser" at the following URL: https://www.knime.org/files/white-papers/customer_segmentation.pdf Pre-processing - Join contract data and behavioral data - Convert Churn values to String to be used as class in upcomingclassification - Normalize all numerical columns in [0,1] ClusteringClustering is performed with k-Means. Other Learner nodes train other models. MostLearner nodes output a PMML model (blue square output port). Input data withassigned cluster Cluster centers Data Reading 2 files: - contract data - behavioral (calls) dataBoth files are located in TheData/Customers Try this:Clustering Algorithms1) Choose your clustering technique and placeit here. k-Means here is just an example. 2) Open the view of the k-Means node to seecluster prototypes.Denormalization1) After executing the k-Means, denormalize thedata to see values in their original range. 10 clusterson all numericalinputsNormalize allnumerical columnsto fall in [0,1]Back to originaldata rangeBack to originaldata rangeconverting a numerical column to Stringexcludes it from the clustering procedureExclude area code and churn from subsequent clusteringReadingContractData.csvReadingCallsData.xlsJoin calls dataand contract data k-Means Normalizer (PMML) Denormalizer (PMML) Denormalizer (PMML) Number To String CSV Reader Excel Reader Joiner

Nodes

Extensions

Links