HybridSVD: When collaborative information is not enough

Evgeny Frolov, Ivan Oseledets

    Результат исследований: Глава в книге, отчете, сборнике статейМатериалы для конференциирецензирование

    13 Цитирования (Scopus)


    We propose a new hybrid algorithm that allows incorporating both user and item side information within the standard collaborative filtering technique. One of its key features is that it naturally extends a simple PureSVD approach and inherits its unique advantages, such as highly efficient Lanczos-based optimization procedure, simplified hyper-parameter tuning and a quick folding-in computation for generating recommendations instantly even in highly dynamic online environments. The algorithm utilizes a generalized formulation of the singular value decomposition, which adds flexibility to the solution and allows imposing the desired structure on its latent space. Conveniently, the resulting model also admits an efficient and straightforward solution for the cold start scenario. We evaluate our approach on a diverse set of datasets and show its superiority over similar classes of hybrid models.

    Язык оригиналаАнглийский
    Название основной публикацииRecSys 2019 - 13th ACM Conference on Recommender Systems
    ИздательAssociation for Computing Machinery, Inc
    Число страниц9
    ISBN (электронное издание)9781450362436
    СостояниеОпубликовано - 10 сент. 2019
    Событие13th ACM Conference on Recommender Systems, RecSys 2019 - Copenhagen, Дания
    Продолжительность: 16 сент. 201920 сент. 2019

    Серия публикаций

    НазваниеRecSys 2019 - 13th ACM Conference on Recommender Systems


    Конференция13th ACM Conference on Recommender Systems, RecSys 2019


    Подробные сведения о темах исследования «HybridSVD: When collaborative information is not enough». Вместе они формируют уникальный семантический отпечаток (fingerprint).