Recommendation systems lie at the heart of many user-facing applications, providing valuable suggestions that improve the user experience. This thesis aims to develop and compare different content-based recommender systems using an existing science fiction book dataset. The implemented methods are based on Word2Vec, Doc2Vec, and BM25 algorithms, all of which rely on textual book descriptions in order to suggest the most similar titles. In addition to this, I have developed an interactive web application that puts the best performing recommender system into action in a user-friendly environment.

From Synopsis to Suggestion: Algorithmic Text-Based Recommendations for Sci-Fi Books

VALENTINI, CHIARA
2024/2025

Abstract

Recommendation systems lie at the heart of many user-facing applications, providing valuable suggestions that improve the user experience. This thesis aims to develop and compare different content-based recommender systems using an existing science fiction book dataset. The implemented methods are based on Word2Vec, Doc2Vec, and BM25 algorithms, all of which rely on textual book descriptions in order to suggest the most similar titles. In addition to this, I have developed an interactive web application that puts the best performing recommender system into action in a user-friendly environment.
Dipartimento di Scienze Statistiche
2024
From Synopsis to Suggestion: Algorithmic Text-Based Recommendations for Sci-Fi Books
Recommender Systems
Word Embeddings
Text Mining
Science Fiction
Lauree triennali
File in questo prodotto:
File Dimensione Formato  
Valentini_Chiara.pdf

accesso aperto

Dimensione 4.77 MB
Formato Adobe PDF
Visualizza/Apri
4.77 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/92984