Strumenti Utente

Strumenti Sito


dm:mains.santanna.dm4crm.2012

Differenze

Queste sono le differenze tra la revisione selezionata e la versione attuale della pagina.

Link a questa pagina di confronto

Entrambe le parti precedenti la revisione Revisione precedente
Prossima revisione
Revisione precedente
dm:mains.santanna.dm4crm.2012 [19/05/2015 alle 15:17 (9 anni fa)]
Dino Pedreschi [Exercises]
dm:mains.santanna.dm4crm.2012 [18/04/2016 alle 19:45 (8 anni fa)] (versione attuale)
Anna Monreale [Calendar]
Linea 43: Linea 43:
  
 ^ ^ Date ^ Topic ^ Learning material ^Instructor ^  ^ ^ Date ^ Topic ^ Learning material ^Instructor ^ 
-|01.   13.05.2015 - 09:00-13:00  | Introduction to data mining and big data analytics | {{:dm:1.dm_ml_introduction.pdf| slides: intro}} {{:dm:2.dm_ml-casestudies.ppt.pdf| slides: case studies}} | Giannotti | +|01.   11.05.2016 - 09:00-13:00  | Introduction to data mining and big data analytics | {{:dm:1.dm_ml_introduction.pdf| slides: intro}} {{:dm:2.dm_ml-casestudies.ppt.pdf| slides: case studies}} | Giannotti | 
-|02.   13.05.2015 - 14:00-18:00  | Data understanding; data preparation; Knime tutorial | {{:dm:4.dm_ml_data_preparation.pdf| slides}} {{:dm:04_dataunderstanding.pdf| slides data understanding}} {{:dm:knime_slides_mains.pdf| Tutorial Knime}}{{:dm:du-iris.zip|Knime su Iris}} | Pedreschi, Monreale | +|02.   11.05.2016 - 14:00-18:00  | Data understanding; data preparation; Knime tutorial | {{:dm:4.dm_ml_data_preparation.pdf| slides}} {{:dm:04_dataunderstanding.pdf| slides data understanding}} {{:dm:knime_slides_mains.pdf| Tutorial Knime}}{{:dm:du-iris.zip|Knime su Iris}} | Pedreschi, Monreale | 
-|03.   14.05.2015 - 09:00-13:00  | Pattern and association rule mining & market basket analysis | | Giannotti | +|03.   12.05.2016 - 09:00-13:00  | Pattern and association rule mining & market basket analysis | | Giannotti | 
-|04.   14.05.2015 - 14:00-18:00  | Clustering analysis & customer segmentation | {{:dm:dm.pedreschi.clustering.2015.pdf| slides clustering}} {{:dm:customersegmentation.pdf| slides customer segmentation}} | Pedreschi | +|04.   12.05.2016 - 14:00-18:00  | Pattern and association rule mining: esercizi con Knime | | Giannotti, Monreale|  
-|05.   15.05.2015 - 09:00-13:00  | Pattern and association rule mining: esercizi con Knime | | Giannotti, Monreale | +|05.   | 13.05.2016 - 09:00-13:00  | Clustering analysis & customer segmentation | {{:dm:dm.pedreschi.clustering.2015.pdf| slides clustering}} {{:dm:customersegmentation.pdf| slides customer segmentation}} | Pedreschi | 
-|06.   | 15.05.2015 - 14:00-18:00  | Clustering analysis: esercizi con Knime | | Pedreschi, Monreale | +|06.   | 13.05.2016 - 14:00-18:00  | Clustering analysis: esercizi con Knime | | Pedreschi, Monreale | 
-|07.   18.05.2015 - 09:00-13:00  | Classification & prediction | {{:dm:dm.giannotti.pedreschi.classification.2015.pdf| slides classification}} | Pedreschi | +|07.   16.05.2016 - 09:00-13:00  | Classification & prediction | {{:dm:dm.giannotti.pedreschi.classification.2015.pdf| slides classification}} | Pedreschi | 
-|08.   18.05.2015 - 14:00-18:00  | Prediction models for promotion performance and churn analysis | | Giannotti | +|08.   16.05.2016 - 14:00-18:00  | Prediction models for promotion performance and churn analysis | {{:dm:5.dml-ml-crm-redemption-churn-promozioni-profili-innovatori.pptx.pdf| slides}} | Giannotti | 
-|09.   19.05.2015 - 09:00-13:00  | Classification & prediction: esercizi con Knime | | Pedreschi, Monreale | +|09.   18.05.2016 - 09:00-13:00  | Classification & prediction: esercizi con Knime | | Pedreschi, Monreale | 
-|10.   19.05.2015 - 14:00-18:00  | Social network analysis: fundamentals | | Pedreschi | +|10.   18.05.2016 - 14:00-18:00  | Social network analysis: fundamentals | {{:dm:pedreschi_sna_crash_course_mains.pptx.pdf| slides}} | Pedreschi | 
-|11.   | 20.05.2015 - 09:00-13:00  | Mobility data mining & big data analytics | | Giannotti | +|11.   | 20.05.2016 - 09:00-13:00  | Mobility data mining & big data analytics | | Giannotti | 
-|12.   | 20.05.2015 - 14:00-18:00  | Big Data Analytics: Privacy awareness | | Giannotti, Monreale | +|12.   | 20.05.2016 - 14:00-18:00  | Big Data Analytics: Privacy awareness | | Giannotti, Monreale |
 ===== Datasets ===== ===== Datasets =====
  
Linea 66: Linea 65:
 ===== Exercises ===== ===== Exercises =====
  
-** DSB-Churn Dataset: ** The dataset consists of 20,000 examples (lines, rows) over 12 variables (fields, columns). The dataset constitutes two-class supervised learning problem The class variable, LEAVE, is the last variable on each line, and its legal values are LEAVE and STAY.  The header of churn.arff describes the legal values of each variable.  Informally, in the following we list their meanings:+** DSB-Churn Dataset: ** The dataset consists of 20,000 examples (lines, rows) over 12 variables (fields, columns) describing features of customers of mobile phone provider, including the class variable LEAVE representing whether e customer decided to quit the company or not. The class variable, LEAVE, is the last variable on each line, and its legal values are LEAVE and STAY.  The header of churn.arff describes the legal values of each variable.  Informally, in the following we list their meanings:
  
 COLLEGE : Is the customer college educated? COLLEGE : Is the customer college educated?
Linea 97: Linea 96:
 **Guidelines:** **Guidelines:**
  
-Each group of 2-3 people has to produce a report (max 10 pagesincluding the discussion of the following tasks:+Each group (2-3 people) is required to deliver a report (max 10 pages including all figures) describing the methods adopted and the discussion of achieved results with reference to the tasks listed below. Assume that the report is targeted to a //marketing strategist//, who is interested to learn the story inferred in the various data mining analyses and to receive suggestions on how to take appropriate actions as a consequence.
  
-**1. Data Understanding**: useful as a preliminary step to capture some data property that can help the next step and especially the clustering analysis (Distribution analysis, statistics computation, suitable transformation of variables and Elimination of redundant variables by correlation analysismanaging of missing values and so on);+**1. Data Understanding**: useful as a preliminary step to capture basic data propertyDistribution analysis, statistical exploration, correlation analysis, suitable transformation of variables and elimination of redundant variables, management of missing values.
  
-**2. Market Basket Analysis. ** Problem: the above dataset prepare the data for the extraction of interesting association rules that is possible to derive from the frequent patterns.  The report should be discuss the parameters used for the analyses, the adopted frequent pattern algorithm and the association rule analysis justifying your findings related to the most interesting rules by using the different measure introduced in the course.+**2. Market Basket Analysis. ** Problem: prepare data and extract interesting association rules and frequent patterns.  The report should discuss the parameters used for the analyses, justifying your findings related to the most interesting rules according to the different measure introduced in the course.
  
-**3. Customer segmentation with k-means.** Problem: given the above dataset, find a high-quality clustering using K-means and discuss the profile of each found cluster (in terms of the properties that describe the behaviour of the customers of each cluster). The report should illustrate the adopted clustering methodology and the cluster interpretation. In particular, it is necessary to discuss the identification of the best value of k and the characterisation 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. Customer segmentation with k-means.** Problem: find a high-quality clustering using K-means and discuss the profile of each found cluster (in terms of the properties that describe the properties of the customers of each cluster). The report should illustrate the adopted clustering methodology and the cluster interpretation. In particular, it is necessary to discuss the identification of the best value of k and the characterisation of the obtained clusters by using both analysis of the k centroids and comparison of the statistics of variables within the clusters with that in the whole dataset.
  
  
-**4. Churn analysis with decision trees. ** Problem: given a above dataset, find a high-quality classifier, using decision trees, which predicts whether each customer will STAY or LEAVE. The report should  illustrate the adopted classification methodology and the decision tree validation and interpretation, describing also the process adopted to select the proposed tree, together with its quality evaluation.+**4. Churn analysis with decision trees. ** Problem: find a high-quality decision tree that predicts whether each customer will STAY or LEAVE. The report should  illustrate the adopted classification methodology and the decision tree validation and interpretation, describing also 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 **1 July 2015**. Specify [MAINS] in the subject of the email. +**Deadline**: send the report by email to all instructors within **1 July 2015**. Specify [MAINS] in the subject of the email. 
 ====== Exams ====== ====== Exams ======
  
-The exam of the CRM module consists in the evaluation of the reports of the proposed exercises.+The exam consists in the evaluation of the report of the proposed mining exercises.
  
  
dm/mains.santanna.dm4crm.2012.1432048629.txt.gz · Ultima modifica: 19/05/2015 alle 15:17 (9 anni fa) da Dino Pedreschi