Autoencoders and recommender systems: COFILS approach

Barbieri, Julio; Alvim, Leandro G.M.; Braida, Filipe; Zimbro, Geraldo

doi:10.1016/j.eswa.2017.07.030

Cited by 30 publications

(28 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Julio Barbieriet al [48] proposed a technique named Autoencoder COFILS to minimize the data sparsity and to test varieties of supervised learning(SL). Stacked DenoisingAutoencoder (SDA) with switching SVD is used to extract the nonlinear features from data.This model is investigated on ML100k, 1M datasets where the accuracy is better as compared to other model-based CF techniques.…”

Section: Nnbased Modelmentioning

confidence: 99%

A Research on Collaborative Filtering Based Movie Recommendations: From Neighborhood to Deep Learning Based System

Behera*,

Das,

Swetanisha

2019

IJRTE

View full text Add to dashboard Cite

Recommender System or Recommendation Engine gaining popularity as it can tackle information overload problem. Initially it was considered as a domain of Information Retrieval system and was limited to few applications. With the advancement of different state-of-the-art modeling approaches recommender system can be applicable to many application domains. Movie Recommender System (MRS) is widely explored domain and used by many streaming service providers like Netflix, Amazon Prime, YouTube and many more. This system makes use of users’ data to explore and recommends personally as per their taste. In this paper a detailed study on recently published article related to movie recommendation is carried out. Popular techniques for MRS are commonlycategorized into collaborative filtering, content-based and hybridmethod. Neighborhood-based, latent factor based, neural network based and deep learning based techniques have been continuously evolved with application to MRS. Recently proposed models have been reviewed and it is found that hybrid method performs better as compared to individual model.

show abstract

Section: Nnbased Modelmentioning

confidence: 99%

A Research on Collaborative Filtering Based Movie Recommendations: From Neighborhood to Deep Learning Based System

Behera*,

Das,

Swetanisha

2019

IJRTE

View full text Add to dashboard Cite

show abstract

“…Wang et al used deep learning for collaborative filtering [40]. Another recent collaborative-filtering approach explicitly takes side information into account for autoencoders [1]. We include a similar model in our comparison, as it is one component of the adversarial autoencoder.…”

Section: Recommendation and Link Prediction Based On Deep Learningmentioning

confidence: 99%

Multi-Modal Adversarial Autoencoders for Recommendations of Citations and Subject Labels

Galke

Mai

Vagliano

et al. 2018

Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization

View full text Add to dashboard Cite

We present multi-modal adversarial autoencoders for recommendation and evaluate them on two different tasks: citation recommendation and subject label recommendation. We analyze the effects of adversarial regularization, sparsity, and different input modalities. By conducting 408 experiments, we show that adversarial regularization consistently improves the performance of autoencoders for recommendation. We demonstrate, however, that the two tasks differ in the semantics of item co-occurrence in the sense that item co-occurrence resembles relatedness in case of citations, yet implies diversity in case of subject labels. Our results reveal that supplying the partial item set as input is only helpful, when item co-occurrence resembles relatedness. When facing a new recommendation task it is therefore crucial to consider the semantics of item co-occurrence for the choice of an appropriate model.

show abstract

“…More precisely, we employ a TF-IDF weighted bag of embedded words representation which has proven to be useful for information retrieval [7]. The usage of side information in an undercomplete autoencoder is comparable to the approach by Barbieri et al [1]. A minor difference is that we supply the side information (titles, artists and albums) only to the decoder, yet use two hidden layers for both the encoder and the decoder to enable a fair comparison to the adversarial variant, which is described below.…”

Section: Our Approachmentioning

confidence: 99%

“…Given a set of playlist features and initial tracks, the system generates a list of recommended tracks that can "continue" that playlist. This is particularly interesting in services such as Spotify 1 . Automatic playlist continuation is becoming more and more important as user tend to prefer being recommended musical experiences rather than single tracks [15].…”

Section: Introductionmentioning

confidence: 99%

Using Adversarial Autoencoders for Multi-Modal Automatic Playlist Continuation

Vagliano

Galke

Mai

et al. 2018

Proceedings of the ACM Recommender Systems Challenge 2018

View full text Add to dashboard Cite

The task of automatic playlist continuation is generating a list of recommended tracks that can be added to an existing playlist. By suggesting appropriate tracks, i. e., songs to add to a playlist, a recommender system can increase the user engagement by making playlist creation easier, as well as extending listening beyond the end of current playlist. The ACM Recommender Systems Challenge 2018 focuses on such task. Spotify released a dataset of playlists, which includes a large number of playlists and associated track listings. Given a set of playlists from which a number of tracks have been withheld, the goal is predicting the missing tracks in those playlists. We participated in the challenge as the team Unconscious Bias and, in this paper, we present our approach. We extend adversarial autoencoders to the problem of automatic playlist continuation. We show how multiple input modalities, such as the playlist titles as well as track titles, artists and albums, can be incorporated in the playlist continuation task. CCS CONCEPTS • Information systems → Recommender systems; • Computing methodologies → Neural networks; Learning from implicit feedback;

show abstract

Autoencoders and recommender systems: COFILS approach

Cited by 30 publications

References 11 publications

A Research on Collaborative Filtering Based Movie Recommendations: From Neighborhood to Deep Learning Based System

A Research on Collaborative Filtering Based Movie Recommendations: From Neighborhood to Deep Learning Based System

Multi-Modal Adversarial Autoencoders for Recommendations of Citations and Subject Labels

Using Adversarial Autoencoders for Multi-Modal Automatic Playlist Continuation

Contact Info

Product

Resources

About