Dataset Information

A social image recommendation system based on deep reinforcement learning.

ABSTRACT: Today, due to the expansion of the Internet and social networks, people are faced with a vast amount of dynamic information. To mitigate the issue of information overload, recommender systems have become pivotal by analyzing users' activity histories to discern their interests and preferences. However, most available social image recommender systems utilize a static strategy, meaning they do not adapt to changes in user preferences. To overcome this challenge, our paper introduces a dynamic image recommender system that leverages a deep reinforcement learning (DRL) framework, enriched with a novel set of features including emotion, style, and personality. These features, uncommon in existing systems, are instrumental in crafting a user's characteristic vector, offering a personalized recommendation experience. Additionally, we overcome the challenge of state representation definition in reinforcement learning by introducing a new state representation. The experimental results show that our proposed method, compared to some related works, significantly improves Recall@k and Precision@k by approximately 7%-10% (for the top 100 images recommended) for personalized image recommendation.

SUBMITTER: Ahmadkhani S

PROVIDER: S-EPMC10994284 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A social image recommendation system based on deep reinforcement learning.

Ahmadkhani Somaye S Moghaddam Mohsen Ebrahimi ME

PloS one 20240404 4

Today, due to the expansion of the Internet and social networks, people are faced with a vast amount of dynamic information. To mitigate the issue of information overload, recommender systems have become pivotal by analyzing users' activity histories to discern their interests and preferences. However, most available social image recommender systems utilize a static strategy, meaning they do not adapt to changes in user preferences. To overcome this challenge, our paper introduces a dynamic imag ...[more]

PMID: 38574062

Similar Datasets

Project description:BackgroundTraditional clinical assessments often lack individualization, relying on standardized procedures that may not accommodate the diverse needs of patients, especially in early stages where personalized diagnosis could offer significant benefits. We aim to provide a machine-learning framework that addresses the individualized feature addition problem and enhances diagnostic accuracy for clinical assessments.MethodsIndividualized Clinical Assessment Recommendation System (iCARE) employs locally weighted logistic regression and Shapley Additive Explanations (SHAP) value analysis to tailor feature selection to individual patient characteristics. Evaluations were conducted on synthetic and real-world datasets, including early-stage diabetes risk prediction and heart failure clinical records from the UCI Machine Learning Repository. We compared the performance of iCARE with a Global approach using statistical analysis on accuracy and area under the ROC curve (AUC) to select the best additional features.FindingsThe iCARE framework enhances predictive accuracy and AUC metrics when additional features exhibit distinct predictive capabilities, as evidenced by synthetic datasets 1-3 and the early diabetes dataset. Specifically, in synthetic dataset 1, iCARE achieved an accuracy of 0·999 and an AUC of 1·000, outperforming the Global approach with an accuracy of 0·689 and an AUC of 0·639. In the early diabetes dataset, iCARE shows improvements of 1·5-3·5% in accuracy and AUC across different numbers of initial features. Conversely, in synthetic datasets 4-5 and the heart failure dataset, where features lack discernible predictive distinctions, iCARE shows no significant advantage over global approaches on accuracy and AUC metrics.InterpretationiCARE provides personalized feature recommendations that enhance diagnostic accuracy in scenarios where individualized approaches are critical, improving the precision and effectiveness of medical diagnoses.FundingThis work was supported by startup funding from the Department of Psychology at the University of Kansas provided to A.A., and the R01MH125740 award from NIH partially supported J.M.G.'s work.

Project description:Recommender systems have become a core component of various online platforms, helping users get relevant information from the abundant digital data. Traditional RSs often generate static recommendations, which may not adapt well to changing user preferences. To address this problem, we propose a novel reinforcement learning (RL) recommendation algorithm that can give personalized recommendations by adapting to changing user preferences. However, a significant drawback of RL-based recommendation systems is that they are computationally expensive. Moreover, these systems often fail to extract local patterns residing within dataset which may result in generation of low quality recommendations. The proposed work utilizes biclustering technique to create an efficient environment for RL agents, thus, reducing computation cost and enabling the generation of dynamic recommendations. Additionally, biclustering is used to find locally associated patterns in the dataset, which further improves the efficiency of the RL agent's learning process. The proposed work experiments eight state-of-the-art biclustering algorithms to identify the appropriate biclustering algorithm for the given recommendation task. This innovative integration of biclustering and reinforcement learning addresses key gaps in existing literature. Moreover, we introduced a novel strategy to predict item ratings within the RL framework. The validity of the proposed algorithm is evaluated on three datasets of movies domain, namely, ML100K, ML-latest-small and FilmTrust. These diverse datasets were chosen to ensure reliable examination across various scenarios. As per the dynamic nature of RL, some specific evaluation metrics like personalization, diversity, intra-list similarity and novelty are used to measure the diversity of recommendations. This investigation is motivated by the need for recommender systems that can dynamically adjust to changes in customer preferences. Results show that our proposed algorithm showed promising results when compared with existing state-of-the-art recommendation techniques.

Dataset Information

A social image recommendation system based on deep reinforcement learning.

Publications

A social image recommendation system based on deep reinforcement learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets