Item-to-Item Collaborative Filtering

By Greg Linden, Brent Smith, Jeremy York bullet. 2001

How researchers see this paper

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

Citations in the corpus

Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time (2017, background): “Several web-scale production systems have been described in the past [1, 7, 21].”
Temporal QoS-aware web service recommendation via non-negative tensor factorization (2014, methodology): “Among the previous algorithms, Collaborative Filtering (CF) is a family of popular methods, and has been widely used in commercial recommender [16, 20, 21].”
Auror: defending against poisoning attacks in collaborative deep learning systems (2016, background): “Several enterprises employ collaborative learning to understand their customer’s behavior towards their products [8,26].”
Content-based Recommender Systems: State of the Art and Trends (2011, methodology): “com, recommendation algorithms are used to personalize the online store for each customer, for example showing programming titles to a software engineer and baby toys to a new mother [50].”
Improving recommendation lists through topic diversification (2005, methodology): “com [16] and TiVo [1], are item-based, owing to the algorithm’s computational efficiency.”
Artificial intelligence in recommender systems (2020, methodology): “This technique has been widely researched in academia [14] and was quickly applied in the industry more than 20 years ago [15].”
Improving one-class collaborative filtering by incorporating rich user information (2010, methodology): “Collaborative Filtering (CF) approaches have been successfully applied to several real world problems, such as Amazon’s product recommendation [14], Netﬂix’s movie recommendation [13].”
GraphPIM: Enabling Instruction-Level PIM Offloading in Graph Computing Frameworks (2017, methodology): “Today, graph computing is being applied to a variety of domains, including social networks [1], e-commerce recommendations [2], and bio-informatics [3].”
Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach (2023, unknown): “As the early approaches, collaborative filtering algorithms [24, 30] were proposed by recommending items based on similar tastes (user side) or similar characteristics (item side).”
When large language models meet personalization: perspectives of challenges and opportunities (2023, unknown): “Item-based explainable methods relate recommendations to familiar items [131], explaining that the recommended item bears similarity to others the user prefers , which are prevalent on platforms like Amazon [132] and Netﬂix [133].”
AMER: Automatic Behavior Modeling and Interaction Exploration in Recommender System (2020, background): “Early works [30, 34, 35, 40, 49, 59] introduce collaborative filtering to model user’s preference.”
Coevolutionary Recommendation Model: Mutual Learning between Ratings and Reviews (2018, background): “Collaborative Filtering (CF) [24] is a common recommendation approach that has been adopted by many e-commerce sites, from Netflix and Amazon to Digg and Zalando.”
Learning Graph-based POI Embedding for Location-based Recommendation (2016, methodology): “Thus, just like the classic item-based collaborative ﬁltering method [11], our proposed dynamic user preference modeling method has the nice properties of making fast response to new check-in information and scaling to massive datasets without retraining model.”
Transfer Learning in Collaborative Filtering with Uncertain Ratings (2012, background): “Collaborative Filtering Collaborative filtering (Goldberg et al. 1992; Koren 2008; Rendle 2012) as an intelligent component in recommender systems (Linden, Smith, and York 2003; Zheng and Xie 2011) has gained extensive interest in both academia and industry, while most previous works can only make…”
Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions (2005, methodology): “…applications, including the ones for recommending books, CDs, and news articles [64], [88], and some of these methods are used in the “industrial-strength” recommender systems, such as the ones deployed at Amazon [61], MovieLens [67], and VERSIFI Technologies (formerly AdapiveInfo.com) [14].”
Relational Collaborative Topic Regression for Recommender Systems (2015, methodology): “The existing CF methods can be divided into two main categories [24]: memory-based methods [18], [28], [37] and model-based methods [19], [25], [35].”
Want to scale in centralized systems? Think P2P (2015, unknown): “However, accounting for dynamics requires periodic recomputations which turn out to be very costly [36, 38, 39].”
Recommender Systems (2012, methodology): “For example, Amazon’s patented algorithm [117] computes the cosine similarity between binary vectors representing users’ purchases and use it in item-based collaborative filtering.”
On bootstrapping recommender systems (2010, background): “The item-item approach [7, 10] evaluates the preference of a user to an item based on ratings of “neighboring” items by the same user.”