Educational Structure
The Master has a year-long duration; it requires to acquire 60 CFU (university credits) accumulating 1500 hours of work split into frontal teaching, individual study and internship. Students are expected to take a final test consisting in the discussion of a Master Thesis.
The 1500 hours are divided as follows:
- 310 hours of frontal teaching and 690 hours of individual study corresponding to 40 CFU
- 400 hours of internship corresponding to 16 CFU
- 100 hours for the Final Test corresponding to 4 CFU
The 2023/2024 teaching plan is structured into the following courses (click on the name of each course to see the detailed program):
BUSINESS INTELLIGENCE SOFTWARE AND PROGRAMMING METHODOLOGIES
- Master’s agenda and presentation of the topics
- Business Intelligence, Data Science: excursus on Master’s modules
- Backend: Tools for data archiving and updating, flat tables, DW and DataMart
- SAS Basic and SAS Guide: Tools for data collection and data quality – Access Data
- SAS Basic and SAS Guide: Tools for data collection and data quality – Manage Data
- SAS Basic and SAS Guide: Tools for data collection and data quality – Analyze Data
- SAS Basic and SAS Guide: Tools for data collection and data quality – Reporting
- SQL language
- Geographic information and QGIS
- Qlik View
- Data visualization and Dashboard, Tableau
- R
- Python
- Tableau
- Module’s final examination
DATA MANAGEMENT AND UNSTRUCTURED DATA PROCESSING
- Exploring internal and external information sources in the public administration area (Government and digital agendas, Open Data, Data Catalogues)
- Building applications and information bases starting from knowledge needs (Data Design Thinking and Service Design techniques)
- How to build decision databases (modelling decision data, building of decision database)
- Introduction and Data Integration
- Data Integration
- Knime Seminar
- Final examination
DATA COLLECTION AND DATA QUALITY
- Sensors, IoT, Big Data Platforms
- Geographic information and Qgis
- Management of personal data for the recognition and customer contactability
- Final examination
FOUNDATION OF STATISTICS
The detailed program will be available soon.
MINING I – MACHINE LEARNING AND TEXT MINING
- NLP introduction and levels of analysis (syntax, semantics)
- Systems and approaches for knowledge representation and linguistic processing
- Representation of concepts and notion of semantic distance. Text analytics and word sense disambiguation
- Representations of senses and representation of sentence semantics: WordNet and FrameNet
- Search for textual documents based on keywords; Vector Space Model and Document Similarity
- Introduction, preprocessing, data types, similarity and dissimilarity measures
- Classification (introduction, k-Nearest Neighbors, Naive Bayesian networks, decision trees, random forests, performance evaluation: accuracy, F-measure, cross validation, ROC curves)
- Clustering (introduction, K-means, hierarchical Clustering: Single link and Complete link, DBScan, hints on performance evaluation)
- Association rules
- Exercises with Knime on classification and clustering
- Final examination
MINING II – Multivariate Analysis
- Identification of different types of customers: cluster analysis
- Synthesis of multidimensional phenomena: factorial analysis and correspondence analysis
- Examples of synthesis of statistical information in the public administration framework (performance indicators and dashboards, data and text mining, information display)
- Knowing customers by building their profile through Behavioral Segmentation: identification of the objective and setting of the analysis
- Interpretation of results, transition to production and support to company's marketing strategies
ECONOMETRIC MODELS
- Linear regression and logistic regression models
- Diagnostics of a logistic model
- Model choice (linear, non-linear), Correspondence Analysis
- Models: description and examples. Use of conditional probabilities. Forecasting and formalization
- Context in which numbers are produced and "theoretical" models from which they originate: aims and data processing in the models
- Structural models: Path Analysis
- Use of SAS Enterprise Miner
- Final Examination
SIMULATION TECHNICS
- Model tuning
- Model estimation
- Examples of simulation: power, Pareto and Zipf distributions
- Forecast: what-if analysis
- Tutorial on structural models building
- Final Examination
INTERPRETATION AND COMMUNICATION OF STATISTICAL RELATIONS
- Information for Business Intelligence in companies
- Legal responsibility for research, data ownership of the funder, copyright and open access in its dissemination
- Data Display and Dashboard
- The communication of scientific data to stakeholders: models and critical issues of communication, the case of social budgets
- Effective ways to communicate results and numbers in Business Intelligence
- Psychological fundamentals for graphic representations, graphic tools, dashboards
- Journalistic communication of numbers and other graphic representations
- Final examination
Attendance is mandatory. Classroom lessons are subject to the fulfilment of the health safety conditions required by the laws in force at the time of the course start. If the conditions for face-to-face teaching are not satisfied, the lessons and activities of the Master will be held in blended mode (i.e. integrating face-to-face and remote activities) or exclusively in telematic mode, in full compliance with the measures that will be provided for the provision of teaching within University Masters.
Additionally, some hours of in-depth studying are being planned, with attendance being voluntary. Their emphasis will be on the application of specific software (such as SAS), laboratory practice, and real-world case scenarios. Generally, these lessons will be planned for Tuesday afternoons.
Click here to view the list of softwares used in each of the Master's courses.
Any updates will be published regularly on this section of the site.
In order to verify learning, ongoing checks will be carried out during the educational path with an evaluation expressed out of 30 with 18 as the minimum mark for passing. A final test is scheduled at the end of the whole Master, evaluated on a scale out of 110 with 66 as the minimum mark and consisting in the discussion of Master's thesis linked to the internship experience.
All tests must be passed successfully in order to obtain the Master's Diploma.