Method
ALS
Alternating least squares factorization for implicit or explicit feedback.
10 cases from 6 companies apply ALS. 9 were checked in an online A/B test; 1 report results in production.
Typical tasks: Retrieve and rank from millions of items, Recommend to new users and new items, Recommend from what the user just did, Balance relevance and diversity and 2 more. Companies: VK, Avito and 4 more. Often combined with: Gradient boosting, A/B test, Pairwise ranking, Word2Vec, k-means and 27 more.
| Ranking VK Stories | social | 2025-05 | |
| Developing Multimodal Content Models for Recommendations | social | 2025-05 | |
| How We Moved from a Markov Chain to a SASRec Model | classifieds | 2024-05 | |
| Experts Are Shocked: A Brilliant Strategy for Finding an Audience for Clickbait Content | social | 2024-05 | |
| From Small to Great: Finding and Growing Promising New Authors on VK | social | 2024-05 | |
| Transforming Recommendations | social | 2024-05 | |
| Warming up a Cold User | media-streaming | 2024-05 | |
| How we improved the 'Unfamiliar' setting recommendation algorithm in Yandex Music | media-streaming | 2024-05 | |
| How We Relaunched Recommendations at Auto.ru | classifieds | 2023-05 | |
| Creating the Conditions to Launch a Recommender System with Manual LabelingSE | SE | social | 2023-05 |