- als matrix factorization - Google Search
- alternating least squares - Google Search
- 🌿 FRESH 🌿Examples and best practices for building recommendation systems | Hacker News
- Яндекс изнутри: рекомендательные системы Музыки и Дзена - Запись трансляции - YouTube
- pyspark.mllib package — PySpark 2.1.1 documentation
- рекомендательные системы
- just google
- collaborative filtering
- Netflix papers + google
- Yandex Talks [прям good]
- habr.com
рекомендательные системы - YouTube
- GOOGLE: netflix prize - типа по факту древние методы 10 лет назад были разработаны
- https://en.wikipedia.org/wiki/Matrix_factorization_(recommender_systems)
- Matrix Factorization recommendations - YouTube
[прям good] - https://www.youtube.com/results?search_query=recommender+systems
- LightFM
- recommender systems - HN Search
- matrix factorization eli5
- ODS: https://opendatascience.slack.com/messages/C2GKW3KGU/convo/C047H3DP4-1536316132.000100/
pyspark.recommendation, Collaborative Filtering, exists for both RDD and DF API!- Implicit vs explicit
- benfred/implicit: Fast Python Collaborative Filtering for Implicit Feedback Datasets
- matrix factorization als recommender systems implicit feedback - Google Search
- PyData - YouTube
- Prototyping a Recommender System Step by Step Part 1: KNN Item-Based Collaborative Filtering
- Prototyping a Recommender System Step by Step Part 2: Alternating Least Square (ALS) Matrix…
- ALS Implicit Collaborative Filtering – Rn Engineering – Medium
contains many useful links - Singular Value Decomposition - Matrix Factorization (Part 1) | Coursera
[прям good]Recommendation Engines Using ALS in PySpark (MovieLens Dataset) - YouTube Также нужно делать crossvalidation, tune hyperparams with grid search (есть прям код для спарка с paramgridbuilder). Только я чет потерял ссылку, сейчас найду в истории.- Lecture 55 — Latent Factor Recommender System | Stanford University - YouTube
- Mining Massive Datasets - Stanford University [FULL COURSE] - YouTube
- 12 Matrix Models: Recommender systems and Matrix factorization (MLVU2018) - YouTube
- matrix factorization recommendation system missing values - Google Search
- matrix factorization ALS recommendation system missing values - Google Search
- alternating least squares - github search
Thre'are methods to use side features like description, genre, user age, user gender etc
ALS MF puts both users and items in one space. (latent features space). (f1, f2, ... fk are same in U and I matrices) So you can find nearest movies to a given user. Но это не обязательно. По началу пофиг на это.
- An update on Pixie, Pinterest’s recommendation system
- Twitter's Who to Follow algo
- Recommending items to more than a billion people - Facebook Code
goodBuilding an Implicit Recommendation Engine with Spark with Sophie Watson (Red Hat) - YouTube- Также в самом spark есть штуки для работы с implicit
- spark als implicit prefs - Google Search
- Explicit vs. implicit feedback
- также разобраться с cold-strategy
- coldStartStrategy = Param(parent='undefined', name='coldStartStrategy', doc="strategy for dealing with unknown or new users/items at prediction time. This may be useful in cross-validation or production scenarios, for handling user/item ids the model has not seen in the training data. Supported values: 'nan', 'drop'.")
als recommender system alpha parameter - Google Search- https://github.com/benhamner/Metrics/tree/master/Python/ml_metrics
- https://gist.github.com/bwhite/3726239
- Alternate explanation of Mean Average Precision | Kaggle
- Search · MeanAveragePrecision
- Search · Mean Average Precision
- https://gist.github.com/mblondel/7337391
- https://www.youtube.com/watch?v=pM6DJ0ZZee0