Data Warehouse e Data Mining (Modulo 2)
Informazioni sul corso
Nome Docente: Pasquale Rullo.
Orario di ricevimento: su appuntamento.
Assistente: Ettore Ritacco
Orario di ricevimento: su appuntamento.
Avvisi
* 07/05/2015
Lab info: on 12/05/2015 h11.30-13.30, Mario Ettorre, Marketing & Sales Director of Exeura (http://www.exeura.eu/), will hold a seminar titled From BI to Big Data Analytics: market evolution, technologies, methodologies and solutions for vertical markets. The lab lesson will be postponed to Friday, 15th April h11.30.
TITOLO: Dalla BI ai Big Data Analytics: evoluzione del mercato, tecnologie, metodologie e soluzioni per i vertical market
ABSTRACT: L’avvento dei Big Data ha rivoluzionato il mondo dell’analisi dei dati richiedendo tecnologie, approcci e metodologie “disruptive”. Nel corso del seminario si illustrerà l’evoluzione rilevata nel corso degli ultimi venti anni nel panorama della “Data Analysis" che, partendo dai data warehouse, è giunta oggi ai “Big Data Analytics attraversando l’importante tappa della Business Intelligence.In particolare si mostreranno gli impatti e le opportunità offerte dei Big Data nei vai mercati verticali focalizzando l’attenzione sul tema del Customer Lifecycle Management. Il seminario prevede la condivisione di demo live di prodotti e soluzioni industriali costruite sul prodotto Rialto.
* 21/04/2015
- Lab info: today's lesson will be postponed to Friday, 24th April h11.30
Teaching material
Slides
Lesson 1 - Introduction [ rullo - lesson1.pdf ]
Lesson 2 - Concept Learning [ Lesson2 - Concept Learning.pdf ]
Lesson 3 - Beyond Candidate Elimination [ rullo - lesson 3 - beyond CE-1.pdf ]
Lesson 4/5 - Decision Tress [ Less5-Decision Trees.pdf ]
Lesson 6 - Naive Bayes [ Less7 - NB Classifiers.pdf ]
Lesson 7 - Rule-Based Classification [ Less6- RuleBased Classifiers v1.pdf ]
Lesson 8 - Instance-Based Classification [ Less8-InstanceBasedClassifiers v1.pdf ]
Lesson 9 - Text Classification - Clustering (K-Means) [ Less9 - Text Classification.pdf ] [ Less12-Clustering-Kmeans.pdf ]
Lesson 10 - Association Rules [ Less13-Association Rules - A priori v1.pdf ]
Lab Activites
Lesson 1 [ 01.Introduction.pdf ]
- Introduction to the Data Mining
- DIKW model
- CRISP-DM Methodology
Lesson 2 [ 2. Data Understanding.pdf ] , [ 3. Data Preparation.pdf ]
- Data analysis
- Data manipulation
Lesson 3 [ 4. Study case - Drug.pdf ], [ drug.arff ]
- Weka
- A study case for data understanding and manipulation - Drug
Lesson 4 [ 5. Study case - Churn Analysis.pdf ], [ churn.arff ], [ sick.arff ]
- A study case for data understanding and manipulation - Churn
- A study case for data understanding and manipulation - Sick
Lesson 5 [ 06. Study case - Image Segmentation.pdf ], [ segment.arff ]
- Exam simulation - A study case for data understanding and manipulation - Image Segmentation
Lesson 6 [ 07. Evaluation - part 1.pdf ], [ 08. Study case - Intrusion Detection.pdf ], [ kddcup99-sample.zip ]
- Evaluation (part 1)
- Exam simulation - A study case for data understanding and manipulation - Image Segmentation
Lesson 7 [ 07. Evaluation.pptx ], [ 09. Exercises - DT & NB.pptx ]
- Evaluation (final)
- Exercises: Decision Trees and Naive Bayes
Lesson 8 [ Project.pdf ], [ Training set.zip ]
- Project description
- Project's training data
Lesson 9 [ 10. Exercises - RL & IBC - Errata Corrige.pptx ]
- Exercises: Rule Learning and Instance-Based Classification
Updated version of the lesson (18th May 2015)
Lesson 10 [ Test set.zip ]
- Final evaluation: in attachment the test set