Risorse bibliografiche
Risorsa bibliografica obbligatoria
Risorsa bibliografica facoltativa
Scheda Riassuntiva
Anno Accademico 2022/2023
Scuola Scuola di Ingegneria Industriale e dell'Informazione
Insegnamento 057054 - ANALYTICS FOR BUSINESS LAB
Docente Secchi Piercesare
Cfu 5.00 Tipo insegnamento Modulo Di Corso Strutturato

Corso di Studi Codice Piano di Studio preventivamente approvato Da (compreso) A (escluso) Insegnamento

Obiettivi dell'insegnamento

The course intends to allow students to work hands-on on top-notch, concrete issues in analytics for business. Through a strong collaboration with companies and the possibility to work on real cases, the intended objectives of the lab are to:

- Introduce advanced methods for managing analytics for marketing and performance management;

- Allow students to apply these methods and the others studied in the stream and the study course in real contexts in order to identify effective, viable and reliable solutions to actual problems;

- Favor an understanding of the actual needs and problems of companies when approaching issues related to data-powered decision making in different aspects of their business.


Risultati di apprendimento attesi

Understand challenges, functions, processes in a business and industrial environment and their mutual effects on business, economy, environment and society.

Identify trends,technologies, key methodologies and stakeholder needs in analytics for business.

Interact in a professional, responsible, inclusive, effective and constructive way in a working environment, also motivating group members.


Argomenti trattati

The course will consist of three main parts:

1.     Advanced analytical methods: seminars to introduce top-notch methods and advanced tools to address marketing and performance management real problems, combining a technical introduction and a discussion of their applications, complementarily to what has already been analyzed in the other stream courses. A particular emphasis shall be put on the investigation of the causal nature of the relationship between variables, boing beyond mere correlation

2.     Seminars with practitioners: testimonials from the business community will take part in the classes to show their experiences and discuss with participants the real issues they deal with in their day-to-day job

3.     Lab: students will form groups and be provided with a detailed outline of a business scenario, a dataset to ground their analysis and will be required to develop, in team, a concrete, viable and effective solution applying analytical methods encountered in the stream. During the lab, tutors will deepen the tools and methods required for the solution, also by running ad hoc deep dives and exercises. In the final part of the lab, groups will present their result to the representatives of the companies that will provide the cases and the datasets in order to get also their feedbacks

With respect to the advanced analytical methods, the topics addressed will be:

  • Statistical Natural Language Processing (SNLP) for text classification:
    • Text representations and distributional semantics
    • Singular Value Decomposition and Non-negative Matrix Factorization
    • Latent Semantic Analysis
    • Probabilistic LSA
    • Sentiment Evaluation
    • Sentiment Classification
  • Anomaly Detection Methods (methods for identifying outliers, frauds or other non-standard situations)
    • Traditional statistical approaches: z-scores, IQR, IQR-alpha, …
    • Advanced statistical methods: Gaussian Mixture Models, Independent Component Analysis (ICA), Regression-model based
    • Ensemble Methods: Isolation Forests
    • Reconstruction and subspace-based Methods: AE, PCA
    • Classification Methods: One-class SVM
  • Advanced Market basket analysis (beyond support, confidence and lift, to dig into the underlying association rules)
    • Frequent Itemset Mining
    • Association Rule mining
    • Statistical evaluation of Association Rules, other measures of "interestingness" for AR
    • Contrast set learning
    • Sequence Analysis
  • Advanced Graph/Network theory (for social network analysis, recommendation engines, etc.)
    • Types of Graphs and Graphs as Data representations
    • Node Classification: label propagation
    • Link Prediction: topology-based (common neighbors, Katz measure, preferential attachment, …), probabilistic methods (hierarchical structure model)
    • Network Clustering: stochastic block models
    • Graph Embedding: factorization-based (laplacian eigenmaps, inner-product methods), random walk-based methods



Solid background in statistics is particularly useful. Basic knowledge of SQL and R or similar languages and/or of statistical tools and software (e.g., SPSS, Stata, etc.) may facilitate the learning process.The contents of the courses “Advanced performance measurement” and “Marketing Analytics” will be considered as pre-requisite

Modalità di valutazione
  • 80%: evaluation of the project work (group-based)
  • 20%: test on the theoretical methods discussed in the course (individual)


Software utilizzato
Nessun software richiesto

Forme didattiche
Tipo Forma Didattica Ore di attività svolte in aula
Ore di studio autonome
Laboratorio Informatico
Laboratorio Sperimentale
Laboratorio Di Progetto
Totale 50:00 75:00

Informazioni in lingua inglese a supporto dell'internazionalizzazione
Insegnamento erogato in lingua Inglese
Disponibilità di materiale didattico/slides in lingua inglese
Disponibilità di libri di testo/bibliografia in lingua inglese
Possibilità di sostenere l'esame in lingua inglese
Disponibilità di supporto didattico in lingua inglese
schedaincarico v. 1.8.0 / 1.8.0
Area Servizi ICT