Are we really making much progress? A worrying analysis of recent neural recommendation approaches

@article{Dacrema2019AreWR,
  title={Are we really making much progress? A worrying analysis of recent neural recommendation approaches},
  author={Maurizio Ferrari Dacrema and P. Cremonesi and D. Jannach},
  journal={Proceedings of the 13th ACM Conference on Recommender Systems},
  year={2019}
}
  • Maurizio Ferrari Dacrema, P. Cremonesi, D. Jannach
  • Published 2019
  • Computer Science
  • Proceedings of the 13th ACM Conference on Recommender Systems
    • Deep learning techniques have become the method of choice for researchers working on algorithmic aspects of recommender systems. [...] Key Result Overall, our work sheds light on a number of potential problems in today's machine learning scholarship and calls for improved scientific practices in this area.Expand Abstract
    View on ACM
    re.public.polimi.it
    147 Citations
    Highly Influencial Citations
    6
    Background Citations
    80
    Methods Citations
    32
    Results Citations
    17

    Supplemental Code

    Github Repo
    Via Papers with Code
    MaurizioFD/RecSys2019_DeepLearning_Evaluation
    This is the repository of our article published in RecSys 2019 "Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches" and of several follow-up studies.
    717 Stars • 198 Forks
    Explore Further
    Discover more papers related to the topics discussed in this paper
    147 Citations
    Citation Type

    Citation Type

    A Troubling Analysis of Reproducibility and Progress in Recommender Systems Research
    • 10
    • PDF
    Why Are Deep Learning Models Not Consistently Winning Recommender Systems Competitions Yet?: A Position Paper
    • 1
    Empirical Analysis of Session-Based Recommendation Algorithms
    • 4
    • PDF
    Performance comparison of neural and non-neural approaches to session-based recommendation
    • 32
    • PDF
    Methodological Issues in Recommender Systems Research (Extended Abstract)
    • 2
    • PDF
    Efficient Neural Matrix Factorization without Sampling for Recommendation
    • 12
    • PDF
    Deep Learning for Sequential Recommendation: Algorithms, Influential Factors, and Evaluations
    • 3
    • PDF
    Efficient Heterogeneous Collaborative Filtering without Negative Sampling for Recommendation
    • 7
    • PDF
    Mixed Dimension Embeddings with Application to Memory-Efficient Recommendation Systems
    • 15
    • PDF
    Session-aware Recommendation: A Surprising Quest for the State-of-the-art
    • PDF

    References

    SHOWING 1-5 OF 5 REFERENCES
    Neural Collaborative Filtering
    • 1,614
    • Highly Influential
    • PDF
    Variational Autoencoders for Collaborative Filtering
    • 270
    • Highly Influential
    • PDF
    Content-based Recommender Systems: State of the Art and Trends
    • 1,379
    • Highly Influential
    • PDF
    Item-based collaborative filtering recommendation algorithms
    • 7,241
    • Highly Influential
    • PDF
    Neural collaborative fltering
    • In Proceedings WWW
    • 2017