Performance Of Recommender Systems Research Articles

Efficient and Flexible Long-Tail Recommendation Using Cosine Patterns

With the increasing use of recommender systems in various application domains, many algorithms have been proposed for improving the accuracy of recommendations. Among various dimensions of recommender systems performance, long-tail (niche) recommendation performance remains an important challenge in large part because of the popularity bias of many existing recommendation techniques. In this study, we propose CORE, a cosine pattern–based technique, for effective long-tail recommendation. Comprehensive experimental results compare the proposed approach with a wide variety of classic, widely used recommendation algorithms and demonstrate its practical benefits in accuracy, flexibility, and scalability in addition to the superior long-tail recommendation performance. 1 History: Accepted by Ramaswamy Ramesh, Area Editor for Data Science & Machine Learning. Funding: This work was supported by the National Natural Science Foundation of China [Grants 72031001, 72072091, 72242101]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0194 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2022.0194 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

Enhancing Fashion Recommendations with Facial Recognition: A Gender and Age Prediction Perspective

This research paper presents a real-time system for age and gender estimation integrated into a fashion recommendation context. Leveraging deep learning models for facial detection and classification, alongside a Random Forest-based recommender system for fashion items, the system provides personalized recommendations based on user age, gender, and preferences. The methodology involves detecting faces in video streams, extracting facial regions, and preprocessing them for age and gender estimation using pre-trained CNN models. Fashion recommendations are then generated by assigning weighted scores to items based on various attributes and training a Random Forest regressor. Implementation details, including the use of Python, OpenCV, and machine learning libraries, are discussed. Evaluation metrics for age and gender detection models, as well as recommendation system performance, are outlined. The paper concludes with future directions and challenges, emphasizing enhancements in age and gender estimation accuracy, integration with e-commerce platforms, and the importance of data privacy and computational resources. Keywords — Age and gender estimation, Facial detection, Deep learning, Fashion recommendation, Random Forest.

The Differentiation of Residents’ Cultural Consumption Tendency and Consumption Recommendation System Based on Network Inference Algorithm

Abstract To address the issue of insufficient accuracy in consumer recommendation systems, a new biased network inference algorithm is proposed based on traditional network inference algorithms. This new network inference algorithm can significantly improve the resource allocation ability of the original one, thereby improving recommendation performance. Then, the performance of this algorithm is verified through comparative experiments with network-based inference algorithms, network inference algorithms with initial resource optimization, and heterogeneous network inference algorithms. The results showed that the accuracy of the new network inference algorithm was 24.5%, which was superior to traditional one. In terms of system performance testing, the recommendation hit rate of the new network inference algorithm increased by 13.97%, which was superior to the other three comparative algorithms. The experimental results indicated that a novel network inference algorithm with bias can improve the performance of consumer recommendation systems, providing new ideas for improving the performance of consumer recommendation systems.

A Personalized Collaborative Filtering Recommendation System Based on Bi-Graph Embedding and Causal Reasoning.

The integration of graph embedding technology and collaborative filtering algorithms has shown promise in enhancing the performance of recommendation systems. However, existing integrated recommendation algorithms often suffer from feature bias and lack effectiveness in personalized user recommendation. For instance, users' historical interactions with a certain class of items may inaccurately lead to recommendations of all items within that class, resulting in feature bias. Moreover, accommodating changes in user interests over time poses a significant challenge. This study introduces a novel recommendation model, RCKFM, which addresses these shortcomings by leveraging the CoFM model, TransR graph embedding model, backdoor tuning of causal inference, KL divergence, and the factorization machine model. RCKFM focuses on improving graph embedding technology, adjusting feature bias in embedding models, and achieving personalized recommendations. Specifically, it employs the TransR graph embedding model to handle various relationship types effectively, mitigates feature bias using causal inference techniques, and predicts changes in user interests through KL divergence, thereby enhancing the accuracy of personalized recommendations. Experimental evaluations conducted on publicly available datasets, including "MovieLens-1M" and "Douban dataset" from Kaggle, demonstrate the superior performance of the RCKFM model. The results indicate a significant improvement of between 3.17% and 6.81% in key indicators such as precision, recall, normalized discount cumulative gain, and hit rate in the top-10 recommendation tasks. These findings underscore the efficacy and potential impact of the proposed RCKFM model in advancing recommendation systems.

A Dual Perspective Framework of Knowledge-correlation for Cross-domain Recommendation

Recommender System provides users with online services in a personalized way. The performance of traditional recommender systems may deteriorate because of problems such as cold-start and data sparsity. Cross-domain Recommendation System utilizes the richer information from auxiliary domains to guide the task in the target domain. However, direct knowledge transfer may lead to a negative impact due to data heterogeneity and feature mismatch between domains. In this article, we innovatively explore the cross-domain correlation from the perspectives of content semanticity and structural connectivity to fully exploit the information of Knowledge Graph. First, we adopt domain adaptation that automatically extracts transferable features to capture cross-domain semantic relations. Second, we devise a knowledge-aware graph neural network to explicitly model the high-order connectivity across domains. Third, we develop feature fusion strategies to combine the advantages of semantic and structural information. By simulating the cold-start scenario on two real-world datasets, the experimental results show that our proposed method has superior performance in accuracy and diversity compared with the SOTA methods. It demonstrates that our method can accurately predict users’ expressed preferences while exploring their potential diverse interests.

Weight Adjustment Framework for Self-Attention Sequential Recommendation

In recent years, sequential recommendation systems have become a hot topic in the field of recommendation system research. These systems predict future user actions or preferences by analyzing their historical interaction sequences, such as browsing history and purchase records, and then recommend items that users may be interested in. Among various sequential recommendation algorithms, those based on the Transformer model have become a focus of research due to their powerful self-attention mechanisms. However, one of the main challenges faced by sequential recommendation systems is the noise present in the input data, such as erroneous clicks and incidental browsing. This noise can disrupt the model’s accurate allocation of attention weights, thereby affecting the accuracy and personalization of the recommendation results. To address this issue, we propose a novel method named “weight adjustment framework for self-attention sequential recommendation” (WAF-SR). WAF-SR mitigates the negative impact of noise on the accuracy of the attention layer weight distribution by improving the quality of the input data. Furthermore, WAF-SR enhances the model’s understanding of user behavior by simulating the uncertainty of user preferences, allowing for a more precise distribution of attention weights during the training process. Finally, a series of experiments demonstrate the effectiveness of the WAF-SR in enhancing the performance of sequential recommendation systems.

Learning-to-rank debias with popularity-weighted negative sampling and popularity regularization

Most learning-to-rank recommendation models assume users prefer interacted items more than non-interacted ones. All non-interacted items have the same chance to be selected as negative samples. However, the observed interaction data usually shows long-tailed distributions, making traditional recommendation models trained on such data tend to rank popular items higher than users’ true preference and degrading the performance of recommender systems. Some existing debias methods focus on eliminating such popularity bias, while simply removing all the popularity bias may diminish recommendation accuracy because popularity also contains some useful information implying item attractiveness, making it pivotal to handling popularity bias in proper ways. In this paper, popularity is recognized to have both positive and negative effects on recommendation. The positive effect of popularity should be amplified while the negative effect should be mitigated. Therefore, this paper proposes two novel schemes to distinguish the positive and negative impact of popularity in learning-to-rank models. First, we propose a popular-weighted sampler to sample non-interacted popular items from users’ collaborative items as negative samples with higher probabilities. We believe that if a popular item is collaboratively similar to a user but not interacted with by him, it is more likely to be a true negative sample for the user. We sample such items with higher probabilities to mitigate the harmful negative popularity bias. Second, popularity could be a significant indicator of items’ intrinsic quality. Hence, we design a popularity regularization scheme to capture item attractiveness, prompting cold-start users to prefer popular items. Finally, we conduct extensive experiments to evaluate the performance of the proposed schemes. Experimental results demonstrate that our approach can improve existing learning-to-rank models’ performance and outperform state-of-the-art debias methods.

Methods of preference aggregation in group recommender systems

The rapid growth of data volumes has led to information overload, which impedes informed decision-making. To solve this problem, recommender systems have emerged that analyze user preferences and offer relevant products on their own. One type of recommender system is group recommender systems, which are designed to facilitate collaborative decision-making, increase user engagement, and promote diversity and inclusion. However, these systems face challenges such as accommodating diverse group preferences and maintaining transparency in recommendation processes. In this study, we propose a method for aggregating preferences in group recommendation systems to retain as much information as possible from group members and improve the accuracy of recommendations. The proposed method provides recommendations to groups of users by avoiding the aggregation process in the first steps of recommendation, which preserves information throughout the group recommendation process and delays the aggregation step to provide accurate and diverse recommendations. When the object of a collaborative filtering-based recommender system is not a single user but a group of users, the strategy for calculating similarity between individual users to find similarity should be adapted to avoid aggregating the preferences of group members in the first step. In the proposed model, the nearest neighbors of a group of users are searched, so the method of finding neighbors is adapted to compare individual users with the group profile. An experimental study has shown that the proposed method achieves a satisfactory balance between accuracy and diversity. This makes it well suited for providing recommendations to large groups in situations where accuracy is more or less important compared to diversity. These results support the assumption that retaining all information from group members without using aggregation techniques can improve the performance of group recommender systems, taking into account various features.

The Implementation of Recommender Systems for Mental Health Recovery Narratives: Evaluation of Use and Performance.

Recommender systems help narrow down a large range of items to a smaller, personalized set. NarraGive is a first-in-field hybrid recommender system for mental health recovery narratives, recommending narratives based on their content and narrator characteristics (using content-based filtering) and on narratives beneficially impacting other similar users (using collaborative filtering). NarraGive is integrated into the Narrative Experiences Online (NEON) intervention, a web application providing access to the NEON Collection of recovery narratives. This study aims to analyze the 3 recommender system algorithms used in NarraGive to inform future interventions using recommender systems for lived experience narratives. Using a recently published framework for evaluating recommender systems to structure the analysis, we compared the content-based filtering algorithm and collaborative filtering algorithms by evaluating the accuracy (how close the predicted ratings are to the true ratings), precision (the proportion of the recommended narratives that are relevant), diversity (how diverse the recommended narratives are), coverage (the proportion of all available narratives that can be recommended), and unfairness (whether the algorithms produce less accurate predictions for disadvantaged participants) across gender and ethnicity. We used data from all participants in 2 parallel-group, waitlist control clinical trials of the NEON intervention (NEON trial: N=739; NEON for other [eg, nonpsychosis] mental health problems [NEON-O] trial: N=1023). Both trials included people with self-reported mental health problems who had and had not used statutory mental health services. In addition, NEON trial participants had experienced self-reported psychosis in the previous 5 years. Our evaluation used a database of Likert-scale narrative ratings provided by trial participants in response to validated narrative feedback questions. Participants from the NEON and NEON-O trials provided 2288 and 1896 narrative ratings, respectively. Each rated narrative had a median of 3 ratings and 2 ratings, respectively. For the NEON trial, the content-based filtering algorithm performed better for coverage; the collaborative filtering algorithms performed better for accuracy, diversity, and unfairness across both gender and ethnicity; and neither algorithm performed better for precision. For the NEON-O trial, the content-based filtering algorithm did not perform better on any metric; the collaborative filtering algorithms performed better on accuracy and unfairness across both gender and ethnicity; and neither algorithm performed better for precision, diversity, or coverage. Clinical population may be associated with recommender system performance. Recommender systems are susceptible to a wide range of undesirable biases. Approaches to mitigating these include providing enough initial data for the recommender system (to prevent overfitting), ensuring that items can be accessed outside the recommender system (to prevent a feedback loop between accessed items and recommended items), and encouraging participants to provide feedback on every narrative they interact with (to prevent participants from only providing feedback when they have strong opinions).

An Improved Fusion-Based Semantic Similarity Measure for Effective Collaborative Filtering Recommendations

Semantic-enhanced recommendation systems are promising approaches to overcome the sparsity and cold-start problems, which are hard to handle using the conventional collaborative filtering (CF) approaches. Further research is needed to effectively integrate ontologies into collaborative filtering recommender systems. This paper proposes an ontology-based semantic similarity measure to evaluate similarities between items and eventually generate accurate recommendations. The proposed semantic similarity measure termed fusion-based semantic similarity takes into account the semantics of ontological instances (i.e. items) inferred from a specific domain ontology, which is determined by analyzing the hierarchical relationships among the instances, as well as the features of the instances and their relationships to other instances. The new measure comprehensively captures the semantic knowledge associated with instances by exploiting all possible shared semantics between instances in a given domain ontology. Furthermore, this paper proposes a new semantic-enhanced hybrid recommendation approach as a result of combining the new semantic similarity measure with the standard item-based CF to enhance the quality of generated recommendations. In order to assess the effectiveness of our semantic-enhanced hybrid collaborative filtering method, a series of experiments were conducted to compare the performance of the proposed approach against well-established benchmark techniques. The reported experimental results consistently emphasize its superiority, demonstrating enhanced predictive abilities and a notable improvement in the quality of recommendations. More specifically, the proposed approach achieved notable 6% reduction in Mean Absolute Error (MAE) in certain cases, outperforming other benchmark techniques. Additionally, this study highlights the potential of using semantic-based similarity to enhance the performance of recommendation systems. Such enhancements address challenges within collaborative filtering, potentially leading to advancements in recommendation system design and optimization.

Integrating textual reviews into neighbor-based recommender systems

Recommender systems are developed to personalize services for each user. The focus of recommender systems is to accurately discover the unknown preferences of users. To address this issue, neighbor-based recommender systems compute a weighted average of neighbors’ known preferences to predict the unknown preferences of an active user. Its weights represent the influence of neighbors’ preferences on the active user’s ones. Previous research estimates these weights solely based on users’ observed preferences. However, preferences are commonly expressed through numerical ratings, presenting a challenge for users to grasp the rating scale and assign the most precise numerical values. Therefore, the observed rating data is characterized by sparsity, a lack of precision, and insufficient detail. In modern recommender systems, besides collecting ratings, it’s entirely possible to gather various types of information. One valuable data source proven to enhance recommender systems significantly is textual reviews authored by users following their experiences with items. In this study, we utilize Bert models to derive user vectors from observed textual reviews. These vectors are then employed to estimate weights between the active user and each neighbor in a neighbor-based recommender system. In contrast to numerical ratings, textual reviews provide a more detailed and precise representation of users’ preferences. This contributes to improving the performance of neighbor-based recommender systems.

Distributionally Robust Optimization methods on robust medical diagnosis systems

In the medical field, modern recommendation systems face significant challenges due to distributional shifts in data. We propose utilizing Distributionally Robust Optimization (DRO) and Distributionally and Outlier Robust Optimization (DORO) methods to address this issue. This paper aims to develop suitable DRO and DORO frameworks for the medical domain and validate their effectiveness through extensive experiments. We employ the DDXPlus dataset for our investigations and cluster patients based on age, sex, and initial evidence to partition the data into distinct distributions. Using a simple three-layer neural network, we incorporate CVaR and CHISQ as DRO methods and their respective DORO forms. The experimental results show that the overall DRO approach demonstrates more significant enhancements while all four methods exhibit improvements over the original distributional scenarios. Our research contributes to optimizing deep learning models in the medical domain and enhancing their robustness. Furthermore, we intend to use these methods to estimate and provide best-fit patient therapies, addressing real-world medical challenges. The application of these approaches has the potential to enhance the performance and practicality of medical recommendation systems, offering improved medical services to patients.

HYBRID RECOMMENDER FOR VIRTUAL ART COMPOSITIONS WITH VIDEO SENTIMENTS ANALYSIS

Topicality. Recent studies confirm the growing trend to implement emotional feedback and sentiment analysis to improve the performance of recommender systems. In this way, a deeper personalization and current emotional relevance of the user experience is ensured. The subject of study in the article is a hybrid recommender system with a component of video sentiment analysis. The purpose of the article is to investigate the possibilities of improving the effectiveness of the results of the hybrid recommender system of virtual art compositions by implementing a component of video sentiment analysis. Used methods: matrix factorization methods, collaborative filtering method, content-based method, knowledge-based method, video sentiment analysis method. The following results were obtained. A new model has been created that combines a hybrid recommender system and a video sentiment analysis component. The average absolute error of the system has been significantly reduced. Added system reaction to emotional feedback in the context of user interaction with virtual art compositions. Conclusion. Thus, the system can not only select the most suitable virtual art compositions, but also create adaptive and dynamic content, which will increase user satisfaction and improve the immersive aspects of the system. A promising direction of further research may be the addition of a subsystem with a generative neural network, which will create new virtual art compositions based on the conclusions of the developed recommendation system.

Housing fuzzy recommender system: A systematic literature review

In recent years, significant attention has been paid to fuzzy recommender systems for housing, highlighting their ability to effectively handle the imprecision and uncertainty inherent in the real estate market. With the objective of improving the filtering of recommendations in the real estate sector, the PRISMA 2020 methodology was applied to perform new systematic reviews using its checklist on six academic databases from 1985 to 2024. RawGraph, Orange Data Minig, Jamovi and R software were used for document classification and data visualization. After classification, 1003 articles were obtained, of which 46.36% were in Scopus, and 57.82% were articles. At the end of the type, 50 articles were identified as primary, subjecting them to six research questions. It was found that 65% of the algorithms used fuzzy logic, 60% used spatial data, and 80% evaluated performance. The main difficulties were related to the integration of various sources of information. Although incorporating reclusive methods is anticipated in future systems, the need remains to address challenging areas to improve the overall performance of fuzzy recommender systems. The reviewed articles focus on enhancing fuzzy data-based recommendation systems by proposing flexible and less intrusive techniques. The significance of incorporating contextual information and exploring hybrid approaches is emphasized, along with the evaluation in real world environments, averaging artificial intelligence.

DCLGM: Fusion Recommendation Model Based on LightGBM and Deep Learning

The recommendation system can mine valuable information according to user preferences, so it is widely used in various industries. However, the performance of recommendation systems is generally affected by the problem of data sparsity, and LightGBM can alleviate the impact caused by data sparsity to a certain extent. To this end, this paper proposes a fusion recommendation model based on the LightGBM and deep learning—CLGM model. The model is composed of LighGBM, cross network and deep neural network. First, the features in the dataset are fused and extracted through LightGBM, and the feature with the highest classification accuracy is selected as the input of the neural network layer; Then, using the cross network and the deep neural network, the linear cross combination feature relationship and nonlinear correlation relationship between high-order features are respectively obtained; finally, the results obtained by the pre-order network are linearly weighted and combined to obtain the final recommendation result. In this paper, AUC and Logloss are used as evaluation indicators to verify the model on the public dataset Criteo and dataset Avazu. The simulation experiment results show that, compared with the four typical recommendation models, the recommendation effect of this model is better.

SMLP4Rec: An Efficient All-MLP Architecture for Sequential Recommendations

Self-attention models have achieved the state-of-the-art performance in sequential recommender systems by capturing the sequential dependencies among user–item interactions. However, they rely on adding positional embeddings to the item sequence to retain the sequential information, which may break the semantics of item embeddings due to the heterogeneity between these two types of embeddings. In addition, most existing works assume that such dependencies exist solely in the item embeddings, but neglect their existence among the item features. In our previous study, we proposed a novel sequential recommendation model, i.e., MLP4Rec, based on the recent advances of MLP-Mixer architectures, which is naturally sensitive to the order of items in a sequence because matrix elements related to different positions of a sequence will be given different weights in training. We developed a tri-directional fusion scheme to coherently capture sequential, cross-channel, and cross-feature correlations with linear computational complexity as well as much fewer model parameters than existing self-attention methods. However, the cascading mixer structure, the large number of normalization layers between different mixer layers, and the noise generated by these operations limit the efficiency of information extraction and the effectiveness of MLP4Rec. In this extended version, we propose a novel framework – SMLP4Rec for sequential recommendation to address the aforementioned issues. The new framework changes the flawed cascading structure to a parallel mode, and integrates normalization layers to minimize their impact on the model’s efficiency while maximizing their effectiveness. As a result, the training speed and prediction accuracy of SMLP4Rec are vastly improved in comparison to MLP4Rec. Extensive experimental results demonstrate that the proposed method is significantly superior to the state-of-the-art approaches. The implementation code is available online to ease reproducibility.

Enhancing Recommender System performance through the fusion of Fuzzy C-Means, Restricted Boltzmann Machine, and Extreme Learning Machine

Enhancing Recommender System performance through the fusion of Fuzzy C-Means, Restricted Boltzmann Machine, and Extreme Learning Machine

Application of AI technology in personalized recommendation system for financial services

Abstract How to quickly and accurately retrieve the products that users are interested in from a huge amount of financial products has become a business pain point that financial institutions must solve. In this study, an interpretable EPRSA model for personalized financial service recommendations based on self-attention mechanisms is proposed by combining an LSTM model and an LDA topic model with AI technology support. A customized recommendation system for financial services is constructed by introducing the Nginx server into the Flask framework, and the design of the database and personalized recommendation module is interpreted. For the financial service personalized recommendation system proposed in this paper, its recommendation performance and system performance are tested, and the stock financial products are selected as the recommendation objects to explore its recommendation effect. It is found that the DNCG index of personalized recommendation of financial products of the EPRSA model is improved by 40.18%, the average response time of the system when the number of concurrent users is 1,000 is 1.96 s. The average quality of the personalized recommendation of the collection of stocks reaches 0.153. The customized recommendation of financial services using AI technology can select financial products based on the investor’s preference, help investors better understand the product returns, and improve the service quality of the financial industry.

A Social-aware Gaussian Pre-trained model for effective cold-start recommendation

The use of pre-training is an emerging technique to enhance a neural model’s performance, which has been shown to be effective for many neural language models such as BERT. This technique has also been used to enhance the performance of recommender systems. In such recommender systems, pre-training models are used to learn a better initialisation for both users and items. However, recent existing pre-trained recommender systems tend to only incorporate the user interaction data at the pre-training stage, making it difficult to deliver good recommendations, especially when the interaction data is sparse. To alleviate this rcommon data sparsity issue, we propose to pre-train the recommendation model not only with the interaction data but also with other available information such as the social relations among users, thereby providing the recommender rsystem with a better initialisation compared with solely relying on the ruser interaction data. We propose a novel recommendation model, the Social-aware Gaussian Pre-trained model (SGP), which encodes the user social relations and interaction data at the pre-training stage in a Graph Neural Network (GNN). Afterwards, in the subsequent fine-tuning stage, our SGP model adopts a Gaussian Mixture Model (GMM) to factorise these pre-trained embeddings for further training, thereby benefiting the cold-start users from these pre-built social relations. Our extensive experiments on three public datasets show that, in comparison to 16 competitive baselines, our SGP model significantly outperforms the best baseline by upto 7.7% in terms of NDCG@10. In addition, we show that SGP permits to effectively alleviate the cold-start problem, especially when users newly register to the system through their friends’ suggestions.

A many objective commercial recommendation algorithm via Game-Based core node extraction

The development of the recommendation system (RS) has been focused on improving the accuracy of algorithms. As the number of new users grows, RS suffers from the cold start problem, data sparsity and long-tail problem. In this paper, we proposed a many objective commercial recommendation algorithm that extracts the core node through the pay-off function of game theory and makes it guide the evolutionary computation. Firstly, the core aim of commercial recommendations is to maximize total profit. Profits can vary wildly with the same accuracy. Thus, the profit value as one of the goals is considered in attention to accuracy, coverage, and novelty. Besides, the current user state value is updated according to user features and rating scores matrix with the attention mechanism at different time stamps. The nodes are divided into different communities, forming a dynamic community, and recommendations are made within each community. Meanwhile, to reduce the computational complexity and runtime, the pay-off function of game theory is used to extract the core user in each community. Each contribution value and user interaction as criteria to guide the current community of users. Experiments on real datasets show that selecting core nodes can improve the performance of business RS. Compared with other existing algorithms in many experiments, our algorithm performs better.