This project focuses on the design and development of an intelligent Decision Support System (DSS) based on a Retrieval-Augmented Generation (RAG) architecture for university guidance. By integrating Large Language Models (LLMs) and extended-window In-Context Reasoning techniques, the system is capable of processing and correlating heterogeneous unstructured datasets—including AlmaLaurea employment indicators, OPIS teaching evaluations, and administrative regulations—to enable advanced comparative analyses between the various degree programs at the Polytechnic University of Bari. The software infrastructure follows a client-server architectural pattern, featuring a reactive front-end developed in Angular and a Python-based Flask back-end to manage inference logic and data retrieval. This provides a scalable, data-driven solution aimed at optimizing the student decision-making process and reducing educational mismatch.
Il presente progetto si occupa della progettazione e dello sviluppo di un Decision Support System (DSS) intelligente basato su architettura Retrieval-Augmented Generation (RAG) per l'orientamento universitario. Attraverso l'integrazione di Large Language Models (LLM) e tecniche di In-Context Reasoning a finestra estesa, il sistema è in grado di processare e correlare dataset eterogenei non strutturati — inclusi indicatori occupazionali AlmaLaurea, valutazioni didattiche OPIS e regolamenti amministrativi — per abilitare analisi comparative avanzate tra i diversi percorsi di laurea del Politecnico di Bari. L'infrastruttura software segue un pattern architetturale client-server, con un front-end reattivo sviluppato in Angular e un back-end Python basato su Flask per la gestione della logica di inferenza e il recupero dei dati, offrendo una soluzione scalabile e data-driven volta a ottimizzare il processo decisionale degli studenti e ridurre il mismatch formativo.
ANALISI COMPARATIVA DEI PERORSI DI LAUREA TRAMITE AI GENERATIVA: PROGETTAZIONE DI UN TUTOR DIGITALE PER LA SCELTA CONSAPEVOLE DEL PERCORSO DI STUDI
LANZOLLA, GIUSEPPE
2025/2026
Abstract
This project focuses on the design and development of an intelligent Decision Support System (DSS) based on a Retrieval-Augmented Generation (RAG) architecture for university guidance. By integrating Large Language Models (LLMs) and extended-window In-Context Reasoning techniques, the system is capable of processing and correlating heterogeneous unstructured datasets—including AlmaLaurea employment indicators, OPIS teaching evaluations, and administrative regulations—to enable advanced comparative analyses between the various degree programs at the Polytechnic University of Bari. The software infrastructure follows a client-server architectural pattern, featuring a reactive front-end developed in Angular and a Python-based Flask back-end to manage inference logic and data retrieval. This provides a scalable, data-driven solution aimed at optimizing the student decision-making process and reducing educational mismatch.I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/110441