ITEM2VEC: Neural item embedding for collaborative filtering

By Oren Barkan, Noam Koenigstein. 2016

548 citations recorded by the source corpus.

Many Collaborative Filtering (CF) algorithms are item-based in the sense that they analyze item-item relations in order to produce item similarities. Recently, several works in the field of Natural Language Processing (NLP) suggested to learn a latent representation of words using neural embedding algorithms. Among them, the Skip-gram with Negative Sampling (SGNS), also known as word2vec, was shown to provide state-of-the-art results on various linguistics tasks. In this paper, we show that item-based CF can be cast in the same framework of neural word embedding. Inspired by SGNS, we describe a method we name item2vec for item-based CF that produces embedding for items in a latent space. The method is capable of inferring item-item relations even when user information is not available. We present experimental results that demonstrate the effectiveness of the item2vec method and show it is competitive with SVD.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

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

In Artificial Intelligence

In Information Systems

Citations in the corpus

SHOPPER: A Probabilistic Model of Consumer Choice with Substitutes and Complements (2017, background): “There has been some work on extrapolating embedding ideas to shopping items (Rudolph et al., 2016; Liang et al., 2016; Barkan and Koenigstein, 2016), but these newer methods are not readily appropriate to the analyses here.”
Bayesian Neural Word Embedding (2016, methodology): “Specifically, Skip-Gram (SG) with negative sampling, known also as word2vec [4], set new records in various linguistic tasks and its applications have been extended to other domains beyond NLP such as computer vision [5], [6], Collaborative Filtering (CF) and item similarity [7].”
Learning Text Similarity with Siamese Recurrent Networks (2016, background): “, 2013) have recently been applied in diverse settings, such as sentiment analysis (dos Santos and Gatti, 2014) and recommender systems (Barkan and Koenigstein, 2016).”
Milvus: A Purpose-Built Vector Data Management System (2021, background): “In particular, a recent popular approach in recommender systems is called vector embedding that converts an item to a feature vector (such as item2vec [11], word2vec [52], doc2vec [37], graph2vec [26]) and provides recommendations via finding similar vectors [13, 15, 25, 51].”
MeLU: Meta-Learned User Preference Estimator for Cold-Start Recommendation (2019, methodology): “g process to extract useful features to estimate user preferences from the contents, because the previous research showed that the embedding process improves the performance in the recommender system [1, 30]. Our model embeds user and item contents in a similar way as in previous work [3]. For categ”
Scalable Attentive Sentence-Pair Modeling via Distilled Sentence Embedding (2019, background): “text and target representations that appear in many neural embedding methods (Barkan 2017; Barkan and Koenigstein 2016; Mikolov et al. 2013; Mnih and Hinton 2009).”
Scalable Graph Embedding for Asymmetric Proximity (2017, background): “In recent years, Graph embedding has drawn a lot of attentions from the academic fields, due to its wide usage in many real world applications such as Recommendations (Barkan and Koenigstein 2016; Zhang et al. 2016), Social Network Analysis (Perozzi, Al-Rfou, and Skiena 2014), Natural Language Proce”
Modelling Session Activity with Neural Embedding (2016, background): “Specifically, [5] presented a neural model for embedding items in a latent manifold that encodes CF information.”
Large Language Models are Zero-Shot Rankers for Recommender Systems (2023, methodology): “5-turbo to (1) LLMs that have not been instruction-tuned, like LLaMA [28], and (2) LLMs that have been ﬁne-tuned on instructions, including Vicuna [3] and text-davinci-003 2 , and then instruct these LLMs to perform the ranking task.”
IRGAN: A Minimax Game for Unifying Generative and Discriminative Information Retrieval Models (2017, background): “In the application of recommender systems, we also see that a recommended target item (in the original document identier space) can be generated/selected from known context items [2].”
Real-time Personalization using Embeddings for Search Ranking at Airbnb (2018, background): “for various types of recommendations on the Web, including music recommendations [26], job search [13], app recommendations [21], movie recommendations [3, 7], etc.”