Attribute-aware non-linear co-embeddings of graph features

Rashed, Ahmed Nabih Zaki; Grabocka, Josif; Schmidt-Thieme, Lars

doi:10.1145/3298689.3346999

Cited by 31 publications

(23 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other works, taking a different approach, i.e. [24]- [26] propose the use of deep learning models for learning latent features from side information of users and items. Recommendation models using deep learning techniques have become a particularly popular area of research in recent years, following their success in multiple application domains such as computer vision and natural language processing [27].…”

Section: Related Work a Collaborative Filtering Enriched With Simentioning

confidence: 99%

FeatureMF: An Item Feature Enriched Matrix Factorization Model for Item Recommendation

et al. 2021

View full text Add to dashboard Cite

Matrix Factorization (MF) is one of the most successful Collaborative Filtering (CF) techniques used in recommender systems due to its effectiveness and ability to deal with very large useritem rating matrix. However, when the rating matrix sparseness increases its performance deteriorates. Expanding MF to include side-information of users and items has been shown by many both to improve general recommendation performance and to help alleviate the sparsity and cold-start issues in CF. In regard to item feature side-information, most schemes incorporate this information through a two stage process: intermediate results (e.g., on item similarity) are first computed based on item attributes; these are then combined with MF. In this paper, focussing on item side-information, we propose a model that directly incorporates item features with the MF framework in a single step process. The model, which we name FeatureMF, does this by projecting every available attribute datum in each of the item features into the same latent factor space with users and items, thereby in effect enriching item representation in the MF model. Results are presented of comparative performance experiments of the model against three state-ofthe-art item information enriched models, as well as against four reference benchmark models, using two public real-world datasets, Douban and Yelp, with four training:test ratio scenarios each. It is shown to yield the best recommendation performance over all these models across all contexts including data sparsity situations, in particular, achieving over 0.9% to over 6.5% MAE recommendation performance improvement over the next best model, HERec. The model is also found to alleviate cold start and to scale well, almost linearly, in regard to computational time, as a function of dataset size.

show abstract

Section: Related Work a Collaborative Filtering Enriched With Simentioning

confidence: 99%

FeatureMF: An Item Feature Enriched Matrix Factorization Model for Item Recommendation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In addition to the three baselines, we also compared UserReg against multiple state-of-the-art rating prediction models. GraphRec [18] and IGMC [17] are the two most recently proposed CF models for rating prediction task using ML-100K and ML-1M. Based on [18,17], we found a richer set of rating prediction models that present state-of-the-art results, including (1) graph neural networks (GNNs) based ones: GRALS [23], GC-MC [24], sRGCNN [25] and…”

Section: On Movielensmentioning

confidence: 99%

“…For ML-1M, we used λ = 0.1 and β = 8 for the 90/10 train/test split, and λ = 0.1, β = 5 for the 50/50 split. The results are depicted in Table 2, where the best performance figures are shown in bold (results of recent models were taken from [14,15,17,18]).…”

Section: On Movielensmentioning

confidence: 99%

“…On ML-1M, User-Reg cannot catch up with some of the NN-based models, such as I-AutoRec, CF-NADE, but still outperforms more than half of the remaining models. In particular, on both datasets UserReg achieves better performance than AutoSVD and AutoSVD++, which are powerful hybrid models [18] that integrate MF with a contractive autoencoder to tackle data sparsity in CF.…”

Section: On Movielensmentioning

confidence: 99%

“…In recent years, embedding based models using deep learning techniques have become particular popular, following their success in multiple application domains such as computer vision and natural language processing. Most of these models reported significant improvements over baselines [14,15,16,17,18]. However, some recent works have pointed out that models based on deep learning are not as strong as expected while also being computational complex [3,4].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

UserReg: A Simple but Strong Model for Rating Prediction

Zhang

Ганчев

Nikolov

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Collaborative filtering (CF) has achieved great success in the field of recommender systems. In recent years, many novel CF models, particularly those based on deep learning or graph techniques, have been proposed for a variety of recommendation tasks, such as rating prediction and item ranking. These newly published models usually demonstrate their performance in comparison to baselines or existing models in terms of accuracy improvements. However, others have pointed out that many newly proposed models are not as strong as expected and are outperformed by very simple baselines.This paper proposes a simple linear model based on Matrix Factorization (MF), called UserReg, which regularizes users' latent representations with explicit feedback information for rating prediction. We compare the effectiveness of UserReg with three linear CF models that are widely-used as baselines, and with a set of recently proposed complex models that are based on deep learning or graph techniques. Experimental results show that UserReg achieves overall better performance than the fine-tuned baselines considered and is highly competitive when compared with other recently proposed models. We conclude that UserReg can be used as a strong baseline for future CF research.

show abstract