The YouTube video recommendation system

By James Davidson, Benjamin Liebald, Junning Liu, Palash Nandy, T. Vleet, Ullas Gargi, Sujoy Gupta, Yu He, Mike Lambert, Mike Lambert and 2 more. 2010

1,277 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Recommender Systems, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

In Recommender Systems

In Signal Processing

Citations in the corpus

Quantization based Fast Inner Product Search (2015, positive): “VideoRec This dataset consists of embeddings of user interests [8], trained via a deep neural network to predict a set of relevant videos for a user.”
Deep Neural Networks for YouTube Recommendations (2016, methodology): “Furthermore, this design enables blending candidates generated by other sources, such as those described in an earlier work [3].”
Session-based Recommendations with Recurrent Neural Networks (2015, neutral): “Despite of its simplicity it is usually a strong baseline (Linden et al., 2003; Davidson et al., 2010).”
Explainable Recommendation: A Survey and New Perspectives (2018, background): “Davidson et al. [2010] introduced the YouTube Video Recommendation System, and leveraged association rule mining to find the related videos as explanation of a recommended video.”
Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time (2017, background): “However, classical recommender systems that consider catalogs that only contain millions of items (movies [5, 18], videos [4, 7, 10], friends to follow [2, 14, 15]).”
Clipper: A Low-Latency Online Prediction Serving System (2016, background): “Much of this work has focused on content recommendation, including videorecommendation [20], ad-targeting [27, 39], and productrecommendations [37].”
Fast ALS-based tensor factorization for context-aware recommendation from implicit feedback (2012, background): “Association rules [24] are often used to determine which products are bought frequently together and it was reported that in certain cases association rule based recommendations yield the best performance [25].”
NAIS: Neural Attentive Item Similarity Model for Recommendation (2018, background): “mends items similar to the user’s profile—has been heavily used in industrial applications [2], [3], [4], [9].”
A Comprehensive Survey on Poisoning Attacks and Countermeasures in Machine Learning (2022, methodology): “Literature Adversarial Setting Target Model Poisoning Technique Application Evaluation Year Knowledge Capability [106] N-K to P-K IN PCA data inserting intrusion detection ROC 2009 [10] N-K to L-K MO SVM label manipulation binary classification accuracy 2011 [11] P-K IN SVM data manipulation image c”
LaMP: When Large Language Models Meet Personalization (2023, background): “It also represents an important element of large-scale industry recommender systems (Davidson et al., 2010; Das et al., 2007; Xu et al., 2022) and has also been extensively studied for search applications (Bennett et al., 2012; Dumais, 2016; Croft et al., 2001; Tabrizi et al., 2018; Zeng et al.,…”
Auditing radicalization pathways on YouTube (2019, background): “Their latest paper indicates they use embeddings for video searches and video histories as inputs for a dense neural network [12].”
Handling Information Loss of Graph Neural Networks for Session-based Recommendation (2020, methodology): “[4] proposed a method that computes item similarity from items’ co-occurrence patterns and recommends the items that are most likely to co-occur with any item in the active session.”
A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions (2021, background): “There are also other types of recommendation applications, such as video recommendation [11, 37, 110], music recommendation [87, 167, 186], job recommendation [139], food recommendation [131], etc.”
Adaptive Focus for Efficient Video Recognition (2021, background): “, on YouTube or TikTok) has fueled the demands for automatically recognizing human actions, events, or other contents within them, which benefits applications like recommendation [7, 8, 15], surveillance [6, 3] and content-based searching [22].”
Learning to rank recommendations with the k-order statistic loss (2013, methodology): “We next tried our method using the K-os loss (using WARP and k = 5, N = 5) in the live YouTube video recommendation system where we attempted to improve an already strong baseline machine learning system [1].”
Evaluation of session-based recommendation algorithms (2018, methodology): “Although item-to-item approaches are comparably simple, they are commonly used in practice and sometimes considered a strong baselines [8, 34].”
Deep Learning Based Recommender System (2017, background): “In industry, recommender systems are critical tools to enhance user experience and promote sales/services for many online websites and mobile applications [20, 27, 30, 43, 113].”