Designing Multi-Modal Embedding Fusion-Based Recommender

Recommendation systems have lately been popularised globally. However, often they need to be adapted to particular data and the use case. We have developed a machine learning-based recommendation system, which can be easily applied to almost any items and/or actions domain. Contrary to existing reco...

Full description

Saved in:
Bibliographic Details
Published inElectronics (Basel) Vol. 11; no. 9; p. 1391
Main Authors Wróblewska, Anna, Dąbrowski, Jacek, Pastuszak, Michał, Michałowski, Andrzej, Daniluk, Michał, Rychalska, Barbara, Wieczorek, Mikołaj, Sysko-Romańczuk, Sylwia
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.05.2022
Subjects
Algorithms
Beverages
Collaboration
Deep learning
Electronic commerce
Feature selection
Footwear
Machine learning
Recommender systems
Representations
Online AccessGet full text

Cover

More Information
Summary:Recommendation systems have lately been popularised globally. However, often they need to be adapted to particular data and the use case. We have developed a machine learning-based recommendation system, which can be easily applied to almost any items and/or actions domain. Contrary to existing recommendation systems, our system supports multiple types of interaction data with various modalities of metadata through a multi-modal fusion of different data representations. We deployed the system into numerous e-commerce stores, e.g., food and beverages, shoes, fashion items, and telecom operators. We present our system and its main algorithms for data representations and multi-modal fusion. We show benchmark results on open datasets that outperform the state-of-the-art prior work. We also demonstrate use cases for different e-commerce sites.
ISSN:2079-9292
2079-9292
DOI:10.3390/electronics11091391