A Hybrid Recommender System with Implicit Feedbacks in Fashion Retail

Ilaria Cestari, Luigi Portinale, Pier Luigi Riva

Risultato della ricerca: Capitolo in libro/report/atti di convegnoContributo a conferenzapeer review

Abstract

In the present paper we propose a hybrid recommender system dealing with implicit feedbacks in the domain of fashion retail. The proposed architecture is based on a collaborative-filtering module taking into account the fact that users feedbacks are not explicit scores about the items, but are obtained through user interactions with the products in terms of number of purchases; moreover, a second module provides a knowledge-based contextual post-filtering, based on both customer-oriented and business-oriented objectives. We finally present a case study where “look-oriented” recommendations have been implemented for a specific fashion retail brand.

Lingua originaleInglese
Titolo della pubblicazione ospiteAIxIA 2022 – Advances in Artificial Intelligence - XXIst International Conference of the Italian Association for Artificial Intelligence, AIxIA 2022, Proceedings
EditorAgostino Dovier, Angelo Montanari, Andrea Orlandini
EditoreSpringer Science and Business Media Deutschland GmbH
Pagine212-224
Numero di pagine13
ISBN (stampa)9783031271809
DOI
Stato di pubblicazionePubblicato - 2023
Evento21st International Conference of the Italian Association for Artificial Intelligence, AIxIA 2022 - Udine, Italy
Durata: 28 nov 20222 dic 2022

Serie di pubblicazioni

NomeLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13796 LNAI
ISSN (stampa)0302-9743
ISSN (elettronico)1611-3349

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???21st International Conference of the Italian Association for Artificial Intelligence, AIxIA 2022
Paese/TerritorioItaly
CittàUdine
Periodo28/11/222/12/22

Fingerprint

Entra nei temi di ricerca di 'A Hybrid Recommender System with Implicit Feedbacks in Fashion Retail'. Insieme formano una fingerprint unica.

Cita questo