Meta-User2Vec model for addressing the user and item cold-start problem in recommender systems

Misztal-Radecka, Joanna; Indurkhya, Bipin; Smywiński-Pohl, Aleksander

doi:10.1007/s11257-020-09282-4

Cited by 15 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As there may be a variety of reasons for generating explanations (Tintarev & Masthoff, 2011), we focus here on generating descriptions of particular user segments to better understand the underlying data characteristics and hidden biases. This is similar to Misztal-Radecka et al (2019), where segment descriptions were generated based on the content-based user profiles constructed from topic vectors, and Misztal-Radecka, Indurkhya, and Smywiński-Pohl (2020) where user and item metadata are represented in the same vector space. However, such model-specific approaches are limited to a certain class of models, and may suffer from the interpretability-performance tradeoff (Ribeiro, Singh, & Guestrin, 2016).…”

Section: Explaining Recommendationsmentioning

confidence: 97%

Bias-Aware Hierarchical Clustering for detecting the discriminated groups of users in recommendation systems

Misztal-Radecka

Indurkhya

2021

Information Processing & Management

Self Cite

View full text Add to dashboard Cite

One challenge for the modern recommendation systems is the Tyranny of Majority -the generated recommendations are often optimized for the mainstream trends so that the minority preference groups remain discriminated. Moreover, most modern recommendation techniques are characterized as black-box systems. Given a lack of understanding of the dataset characteristics and insufficient diversity of represented individuals, such approaches inevitably lead to amplifying hidden data biases and existing disparities. In this research, we address this problem by proposing a novel approach to detecting and describing potentially discriminated user groups for a given recommendation algorithm. We propose a Bias-Aware Hierarchical Clustering algorithm that identifies user clusters based on latent embeddings constructed by a black-box recommender to identify users whose needs are not met by the given recommendation method. Next, a post-hoc explainer model is applied to reveal the most important descriptive features that characterize these user segments. Our method is model-agnostic and does not require any a priori information about existing disparities and sensitive attributes. An experimental evaluation on a synthetic dataset and two real-world datasets from different domains shows that, compared with other clustering methods and arbitrarily selected user groups, our method is capable of identifying underperforming segments for different recommendation algorithms, and detect more severe disparities.

show abstract

Section: Explaining Recommendationsmentioning

confidence: 97%

Bias-Aware Hierarchical Clustering for detecting the discriminated groups of users in recommendation systems

Misztal-Radecka

Indurkhya

2021

Information Processing & Management

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recall, Precision, F1-Score, AUC [37], [38], [40], [42], [45], [49], [50], [53], [55], [61], [67], [73], [75] Top-K accuracy metric ndcg@K, Hit@K, recall@K, precision@K, AP@K [31], [32], [34]- [38], [46], [49], [52], [54], [55], [57], [58], [60], [63]- [66], [74] Some literature uses only one type of evaluation metric in the performance evaluation process. As in the literature [42] and [53] the AUC was used to evaluate the model.…”

Section: Classification Accuracy Metricmentioning

confidence: 99%

User Cold Start Problem in Recommendation Systems: A Systematic Review

Yuan,

Hernandez

2023

IEEE Access

View full text Add to dashboard Cite

The recommendation system makes recommendations based on the preferences of the users. These user preferences usually come from the user's basic information, item rating, historical data, and so on. The "user cold start problem" happens when a new user cannot be appropriately suggested due to a lack of more detailed preference information. In many instances, the user cold start problem hinders the use of the recommendation system. Many researchers are currently trying to discover a solution to the user cold start problem. Unfortunately, there are two drawbacks in the current systematic reviews of how to deal with the user cold start problem. First, systematic reviews on how to deal with the user cold start problem are scarce or outdated. Second, existing reviews lack the distinction between the user cold start problem and the item cold start problem. Nevertheless, the solutions to the two problems differ. To address these problems, our study thorough review of all literature published by researchers from January 2016 to April 2023 about 8 years. Firstly, this study analyzes the literatures on approaches that addressed the user cold start problem during the past eight years and divides them into two categories: data-driven technology and approach-driven technology, and then describes and classifies each type of technology in detail. Secondly, this study also analyzes the main evaluation criteria currently used in these methods to provide a reference for researchers in related fields. Finally, this paper also points out the future research direction of this field.

show abstract

“…The cold-start problem is a common problem in recommendation systems (RS) [1]- [3]. User coldstart problems [4], [5] and product cold-start problems [6], [7] are two categories of cold-start problems in the RS. Both types of problems arise because no information is available and cause the recommender system to underperform in handling sparse data [8].…”

Section: Introductionmentioning

confidence: 99%

Extending adamic adar for cold-start problem in link prediction based on network metrics

Yuliansyah

Othman

Bakar

2022

Int. J. Adv. Intell. Informatics

View full text Add to dashboard Cite

The cold-start problem is a condition for a new node to join a network with no available information or an isolated node. Most studies use topological network information with the Triadic Closure principles to predict links in future networks. However, the method based on the Triadic Closure principles cannot predict the future link due to no common neighbors between the predicted node pairs. Adamic Adar is one of the methods based on the Triadic Closure principles. This paper proposes three methods for extending Adamic Adar based on network metrics. The main objective is to utilize the network metrics to attract the isolated node or new node to make new relationships in the future network. The proposed method is called the extended Adamic Adar index based on Degree Centrality (DCAA), Closeness Centrality (CloCAA), and Clustering Coefficient (CluCAA). Experiments were conducted by sampling 10% of the dataset as testing data. The proposed method is examined using the four real-world networks by comparing the AUC score. Finally, the experiment results show that the DCAA and CloCAA can predict up to 99% of node pairs with a cold-start problem. DCAA and CloCAA outperform the benchmark, with an AUC score of up to 0,960. This finding shows that the extended Adamic Adar index can overcome prediction failures on node pairs with cold-start problems. In addition, prediction performance is also improved compared to the original Adamic Adar. The experiment results are promising for future research due to successfully improving the prediction performance and overcoming the cold-start problem.

show abstract

Meta-User2Vec model for addressing the user and item cold-start problem in recommender systems

Cited by 15 publications

References 19 publications

Bias-Aware Hierarchical Clustering for detecting the discriminated groups of users in recommendation systems

Bias-Aware Hierarchical Clustering for detecting the discriminated groups of users in recommendation systems

User Cold Start Problem in Recommendation Systems: A Systematic Review

Extending adamic adar for cold-start problem in link prediction based on network metrics

Contact Info

Product

Resources

About