ARTIFICIAL INTELLIGENCE FOR DECISION-MAKING PROCESSES

ARTIFICIAL INTELLIGENCE FOR DECISION-MAKING PROCESSES

Vittorio Carlei

Instructional goals

The course aims to introduce the dynamics and megatrends of Artificial Intelligence technologies and to analyse the impacts they will have on socio-economic processes, especially those supporting information processing in general and decision-making in particular. During the course, concrete cases of projects in which AI-based technologies have innovated products and services offered by companies in a sometimes very strong way will be discussed.

Intended learning outcomes

Basic knowledge of the main predictive and generative AI technologies. Application areas of these technologies to real-world business decision-making processes (marketing, risk, finance, etc.). Ability to evaluate investments in AI-driven technologies in terms of economic and social impact, with the goal of integrating them into decision-making and/or production processes.

Course Contents

Introduction to Artificial Intelligence technologies. Brief historical background and future trends. Main types of AI algorithms and their possible applications on decision-making processes.

Reference Books

Lecturer's Handouts and Melanie Mitchell, "Artificial Intelligence: A Guide for Thinking Humans", (2019)

Teaching Methods

Lectures and case studies

Assessment Method

Oral examination

Thesis assignment criteria

None in particular

Week 1

Introduction and history of IA (1)

Week 2

Introduction and history of IA (2) Ongoing technological trends.

Week 3

Types of AI algorithms Supervised Algorithms: algorithms predicting and classifying all types of data (sounds, images, texts,...). Unsupervised algorithms: clustering and profiling algorithms.

Week 4

Reinforcement Learning Algorithms: k-armed bandit, Policy Evaluation Algorithms.

Week 5

Large Language Models (LLM), Generative AI. Introduction to the approach to artificial agents.

Week 6

Case studies of the application of algorithms to decision-making processes. Areas of application.

Week 7

Presentation of application cases to Marketing: behaviour-based profiling, segmentation and extrapolation algorithms through direct and indirect analysis.

Week 8

Presentation of application cases to risk management: behaviour-based profiling, segmentation and extrapolation algorithms through direct and indirect analysis.

Week 9

Presentation of application cases to financial markets: algorithms for analysing the financial statements of listed companies, analysis of price trends (1/3)

Week 10

Presentation of application cases to financial markets: algorithms for analysing the financial statements of listed companies, analysis of price trends (2/3).

Week 11

Presentation of application cases to financial markets: algorithms for analysing the financial statements of listed companies, analysis of price trends (3/3).

Week 12

Presentation of application cases in diagnostic screening: profiling, segmentation and extrapolation algorithms based on clinical and behavioural data.