Organizations and business are overwhelmed by the flood of data continuously collected into their data warehouses and arriving from external sources – the Web above all. Traditional exploratory techniques may fail to make sense of the data, due to its inherent complexity and size. Data mining and knowledge discovery techniques emerged as an alternative approach, aimed at revealing patterns, rules and models hidden in the data, and at supporting the analytical user to develop descriptive and predictive models for a number of business problems. This short course focusses on the main applications scenarios of data mining to challenging problems in the broad CRM domain - Customer Relationship Management.
|0.||11.04.2014 - 09:00-18:00||Introduction to data mining and big data analytics||intro dm slides|
|1.||03.06.2014 - 09:00-18:00||Pattern and association rule mining & market basket analysis + Exercises||pattern mining slides|
|2.||04.06.2014 - 09:00-18:00||Clustering analysis & customer segmentation + Exercises||clustering slides air miles slides l'oreal slides|
|3.||05.06.2014 - 09:00-18:00||Prediction models for promotion performance and churn analysis + Exercises||classification slides slides slides slides slides|
|4.||09.06.2014 - 09:00-18:00||Mobility data mining & analysis of human movement behavior. Fraud detection||paper paper|
|5.||10.06.2014 - 09:00-18:00||Social network analysis: fundamentals. Network metrics. Small world. Strength of weak ties. Centrality. Preferential attachment. + Exercises||slides link|
|6.||11.06.2014 - 09:00-18:00||Models of social contagion, spreading, diffusion and epidemics, with applications to viral marketing. Analysis of innovators/early adopters. + Privacy & Data Mining||slides slides slides slides|
|7.||Tutorial on Knime||KNIME slides Workflows: Data Manipulation; Clustering FP and AR; Decision Trees|
1. Market Basket Analysis. Problem: given a database of transactions of customers of a supermarket, find the set of frequent items co-purchased and analyse the association rules that is possible to derive from the frequent patterns. Provide a short document (max three pages in pdf, excluding figures/plots) which illustrates the input dataset, the adopted frequent pattern algorithm and the association rule analysis.
Guidelines for the report: The report has to illustrate the input dataset, the adopted frequent pattern algorithm and the association rule analysis discussing your findings related to the most interesting rules by using the different measure introduced in the course.
2. Customer segmentation with k-means. Problem: given the dataset of RFM (Recency, Frequency and Monetary value) measurements of a set of customers of a supermarket, find a high-quality clustering using K-means and discuss the profile of each found cluster (in terms of the purchasing behavior of the customers of each cluster). Provide a short document (max three pages in pdf, excluding figures/plots) which illustrates the input dataset, the adopted clustering methodology and the cluster interpretation. Dataset legend: for each customer, the dataset contains the recency, frequency and monetary value variables (relative to all purchases, to purchases of fresh food articles, to canned food articles and no-food articles; the variables are present both with original and normalized values):
Guidelines for the report:
* Data Understanding: useful as a preliminary step to capture some data property that can help the clustering analysis (Distribution analysis, statistics computation, suitable transformation of variables and Elimination of redundant variables by correlation analysis);
* Clustering Analysis by K-means: Identification of the best value of k and Characterization of the obtained clusters by using both analysis of the k centroids and comparison of the statistics of variables within the clusters and that in the whole dataset.
3. Churn analysis with decision trees. Problem: given a dataset of measurements over a set of customers of an e-commenrce site, find a high-quality classifier, using decision trees, which predicts whether each customers will place only one or more orders to the shop. The explanation of the available variables is here. Provide a short document (max three pages in pdf, excluding figures/plots) which illustrates the input dataset, the adopted classification methodology and the decision tree validation and interpretation.
Guidelines for the report: The report has to illustrate the input dataset, the adopted classification methodology and the decision tree validation and interpretation. Describe the process adopted to select the proposed tree, together with its quality evaluation.
Deadline: the three documents must be sent email to all instructors within 15 July 2014. Specify [MAINS] in the subject of the email.
The exam of the CRM module consists in the evaluation of the reports of the proposed exercises.