Factorizing personalized Markov chains for next-basket recommendation

By Steffen Rendle, Christoph Freudenthaler, Lars Schmidt-Thieme. The Web Conference, 2010

2,851 citations recorded by the source corpus.

Subfields: Information Systems, Management Science and Operations Research, Marketing

View Paper: https://www.semanticscholar.org/paper/50d85cb114f7c5e779a6772f2931e77dddd54a5e

How researchers see this paper

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

In Information Systems

In Management Science and Operations Research

In Marketing

Citations in the corpus

Iterative Splits of Quadratic Bounds for Scalable Binary Tensor Factorization (2014, background): “In order to deal with learning on various forms of structured data such as large-scale knowledge bases, time-varying networks or recommendation data, tensor factorizations have become increasingly popular [3, 21, 23].”
Translation-based Recommendation (2017, methodology): “Such systems focus on modeling user preferences, and typically rely on Collaborative Filtering (CF) techniques, especially Matrix Factorization (MF) [Ricci et al. , 2011].”
Text Is All You Need: Learning Language Representations for Sequential Recommendation (2023, methodology): “Sequential recommenders [11, 14, 25, 27] can capture both short-term and long-term preferences of users and hence are widely used in different recommendation scenarios.”
Neural Survival Recommender (2017, methodology): “We do not compare against FPMC [30] and NMF as they were already shown to perform worse than our competitors.”
Factorization Machines (2010, methodology): “Furthermore their model equations and optimization algorithms are derived individually for each task.”
Interpretable Policies for Dynamic Product Recommendations (2016, background): “Unlike these settings, we deal with cases in which the optimal policy may not be interpretable, which introduces an additional layer of computational difﬁculties.”
Long and Short-Term Recommendations with Recurrent Neural Networks (2017, methodology): “We also compare the RNN with a simple Markov chain that can be seen as a baseline for the sequence prediction approach, and with FPMC [17] and Fossil [8], the state-of-the-art for recommendations based on”
TEM: Tree-enhanced Embedding Model for Explainable Recommendation (2018, methodology): “In the literature, embedding-based methods such as matrix factorization [23, 26, 34] is the most popular CF approach, owing to the strong power of embeddings in generalizing from sparse user-item relations.”
Contrastive Learning for Sequential Recommendation (2022, methodology): “To cope with this problem, various methods have been proposed to make sequential recommendations by capturing users’ dynamic interests from their historical interactions [19, 23, 39–41].”
Factorization Machines with libFM (2012, methodology): “For speci.c tasks, specialized factorization models have been proposed that take noncategorical variables into account, for example, SVD++ [Koren 2008], STE [Ma et al. 2011], FPMC Author s address: S. Rendle, Social Network Analysis, University of Konstanz; email: steffen.rendle@uni-konstanz.de.”
Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation (2023, methodology): “Many early methods are frequently based on Markov Chains techniques [26], [27].”
A Dynamic Recurrent Model for Next Basket Recommendation (2016, methodology): “Factorizing Personalized Markov Chains (FPMC) can model sequential behaviors between every two adjacent baskets, and user general interests is shaped by items in baskets [7].”
A Survey on Accuracy-Oriented Neural Recommendation: From Collaborative Filtering to Information-Rich Recommendation (2021, background): “zation (MF) [3] later becomes the mainstream recommender model for a long time (from 2008 until 2016) [4], [5].”
Time Interval Aware Self-Attention for Sequential Recommendation (2020, methodology): “One typical solution is Markov Chain based approaches [7, 8, 21], where an L-order Markov chain makes recommendations based on the L previous actions.”
Personalized recommendation via cross-domain triadic factorization (2013, methodology): “Metrics: We use two frequently used metrics, recall and AUC [21; 26], to evaluate the quality of ranking for recommendation over TSu, i.”