90733 - Data Analytics

Learning outcomes

By the end of the course, students will: (i) be familiar with the main enabling Machine Learning techniques for data analytics; (ii) be able to implement a data pipeline, from data acquisition to model training; and (iii) understand the main applications of data analytics.

Course contents

This course is part of the integrated course STATISTICAL LEARNING AND DATA ANALYTICS.

Prerequisites: Calculus, Linear Algebra, and Python Programming.

The course introduces concepts, techniques, and tools for designing and implementing processes aimed at analysing data. In particular, it provides a comprehensive overview of all the key components of a data analysis pipeline, from data acquisition and preprocessing to knowledge extraction through Machine Learning techniques and performance evaluation.

The course also discusses use cases of data analytics in business and corporate management contexts. A brief overview of the topics covered is provided below:

• Review of Python programming and data analysis libraries, including Pandas, NumPy, and others.
• Introduction to data: data acquisition, data types, data cleaning, data preprocessing, and dimensionality reduction.
• Introduction to learning: prediction and regression, data splitting, interpretability, and the bias–variance trade-off.
• Supervised data analysis techniques: k-Nearest Neighbours, Support Vector Machines, Decision Trees, Random Forests, and Neural Networks.
• Unsupervised data analysis techniques: K-Means clustering and Gaussian Mixture Models.

Theoretical lectures are complemented by laboratory sessions. The PyTorch framework will be used for Neural Networks, while the Scikit-learn library will be used for all other Machine Learning techniques. Both libraries will be explored in detail during lectures, for example through sample code, and during laboratory sessions, through practical exercises.

Readings/Bibliography

All course slides are made available on the Virtuale platform.
There is no single required textbook; for specific sections of the course (indicated by the instructors as needed), the following readings are recommended:

• Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani, An Introduction to Statistical Learning, Springer, 2013

• Christopher Bishop, Pattern Recognition and Machine Learning, Springer, 2016

• Jiawei Han, Micheline Kamber, Jian Pei, Data Mining: Concepts and Techniques, Morgan Kaufmann Series in Data Management Systems

Additional readings may be suggested by the instructors throughout the course, depending on the topics covered.

Teaching methods

Teaching methods include taught lessons and exercises. The latter will be implemented by (mainly) using the Python language; the lecturers will provide the datasets, the code snippets and solutions on the Virtuale platform. Moreover, business seminaries will be scheduled in the last week of the course.

Assessment methods

This course unit is part of an integrated course. A single final grade will be calculated as the weighted average of the grades obtained in the two course units. In particular, the grade for Statistical Learning accounts for 40% of the final grade, while the grade for Data Analytics accounts for 60%. The final grade will be officially recorded once the student has completed the assessments for both course units.

Further information on the assessment method for the Data Analytics course unit is provided below. For the assessment method of the Statistical Learning course unit, please refer to the relevant course page.

Assessment for the Data Analytics course unit consists of two components: a theoretical written examination, which is mandatory, and a project, which is optional.

The mandatory theoretical written assessment requires students to answer two or more open-response questions on the topics covered during the course lectures. The maximum grade for the written exam is 24. The dates of the theoretical written exam will be published on ALMAESAMI, and students must register for the session through the same portal. The grade for the written exam can be rejected only once.

The use of AI during the theoretical written exam is prohibited. Any use of AI constitutes a violation of academic integrity.

The optional project consists of implementing a data analytics process covering all stages of the pipeline presented in class and using the tools introduced during the lectures, namely PyTorch and Scikit-learn. The dataset to be used for the project will be presented by the lecturer by the end of the course. The project must be completed in groups of no more than two or three students. The maximum grade for the project is 8. The dates for the project presentations will be published on ALMAESAMI, and all members of each group must register through the portal. The project must be submitted through the VIRTUALE platform no later than one week before the date indicated on ALMAESAMI.

For the project, limited and non-substantial use of AI is permitted for support activities, such as code checking and documentation. Substantial use of AI to complete parts of the assessment is not permitted. All students in the group must explicitly declare whether they have used generative AI tools for the project. In addition, students must always be able to answer questions about the code used in their project.

Students who intend to complete the project must inform the lecturer by the end of the course and provide the names of the group members.

Students who complete both the mandatory theoretical written exam and the optional project will receive a final grade calculated as the sum of the grades obtained in the two assessments.

The two assessments may be completed independently. For example, a student may take the theoretical written exam first and complete the project afterwards, or vice versa.

Students with specific learning disabilities or temporary or permanent disabilities are advised to contact the relevant University office well in advance: https://site.unibo.it/studenti-con-disabilita-e-dsa/en. The office will propose any appropriate accommodations to the students concerned. These accommodations must, in any case, be submitted to the Professor for approval at least 15 days in advance. The Professor will assess their suitability, also in relation to the learning objectives of the course unit.

Teaching tools

The PDFs of the slides used in the course will be made available on the course page on Virtuale before each lecture.

The Python notebooks and datasets required for the laboratory sessions will be made available on the course page on Virtuale.

Office hours

See the website of Giuseppe Lisanti

See the website of Marco Di Felice