Overview of a neural network based sequence modeling approach to tackle a problem of personalized recommendations in streaming services
Анотація
Nowadays recommendation systems is a key component of any streaming service holding either UGC (user generated content) or VOD (video on demand) content, which allows to optimize search through the catalog and increase user engagement. Among different types of recommendation systems, personalized one is the most crucial in both helping overwhelmed users to choose the next item to watch and lifting the time spent on a platform. Choosing the right approach to tackle a problem of personalized recommendations is a tricky question that involves many factors. In this work, a sequence modeling approach to tackle a problem of personalized recommendations is reviewed. The general tips and advice about data preparation and algorithm’s modeling parts are provided.
Посилання
Kovenko V., Bogach I., Baraban M. ITEM-BASED COLLABORATIVE FILTERING BASED ON NLP TECHNIQUES, ІТКІ, vol. 51, no. 2, pp. 17–22, Oct. 2021.
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S. (2017). Neural collaborative filtering. Proceedings of the 26th international conference on world wide web. pp. 173–182.
Sedhain, S., Menon, A. K., Sanner, S., & Xie, L. (2015). Autorec: autoencoders meet collaborative filtering. Proceedings of the 24th International Conference on World Wide Web. pp. 111–112.
Hidasi B., Karatzoglou A., Baltrunas L., Tikk D. Session-based Recommendations with Recurrent Neural Networks , cite arxiv:1511.06939 Comment: Camera ready version (17th February, 2016) Affiliation update (29th March, 2016) .
