Conventional Recommender Systems Research Articles

An Effective Federated Recommendation Framework with Differential Privacy

This paper proposes a novel federated recommendation framework that incorporates differential privacy to safeguard user privacy without compromising on the accuracy of recommendations. Unlike conventional recommendation systems that centralize user data, leading to potential privacy breaches, our framework ensures that user data remain on local devices. It leverages a federated learning approach, where a global model is trained across multiple devices without exchanging raw data. To enhance privacy protection, we integrate a specially designed differential privacy algorithm that adds carefully calibrated noise to the aggregated data updates. This mechanism ensures that the global model cannot be exploited to infer individual user information. We evaluate our framework on two real-world datasets, one from the e-commerce sector and another from the multimedia content recommendation domain. The results exhibit that our framework achieves competitive recommendation accuracy compared to traditional centralized approaches, with minimal loss in precision and recall metrics, while significantly enhancing user privacy. Our work stands as a testament to the feasibility of creating recommendation systems that do not have to choose between privacy and performance, paving the way for more ethical AI applications in sensitive domains.

From Traditional Recommender Systems to GPT-Based Chatbots: A Survey of Recent Developments and Future Directions

Recommender systems are a key technology for many applications, such as e-commerce, streaming media, and social media. Traditional recommender systems rely on collaborative filtering or content-based filtering to make recommendations. However, these approaches have limitations, such as the cold start and the data sparsity problem. This survey paper presents an in-depth analysis of the paradigm shift from conventional recommender systems to generative pre-trained-transformers-(GPT)-based chatbots. We highlight recent developments that leverage the power of GPT to create interactive and personalized conversational agents. By exploring natural language processing (NLP) and deep learning techniques, we investigate how GPT models can better understand user preferences and provide context-aware recommendations. The paper further evaluates the advantages and limitations of GPT-based recommender systems, comparing their performance with traditional methods. Additionally, we discuss potential future directions, including the role of reinforcement learning in refining the personalization aspect of these systems.

LLMRG: Improving Recommendations through Large Language Model Reasoning Graphs

Recommendation systems aim to provide users with relevant suggestions, but often lack interpretability and fail to capture higher-level semantic relationships between user behaviors and profiles. In this paper, we propose a novel approach that leverages large language models (LLMs) to construct personalized reasoning graphs. These graphs link a user's profile and behavioral sequences through causal and logical inferences, representing the user's interests in an interpretable way. Our approach, LLM reasoning graphs (LLMRG), has four components: chained graph reasoning, divergent extension, self-verification and scoring, and knowledge base self-improvement. The resulting reasoning graph is encoded using graph neural networks, which serves as additional input to improve conventional recommender systems, without requiring extra user or item information. Our approach demonstrates how LLMs can enable more logical and interpretable recommender systems through personalized reasoning graphs. LLMRG allows recommendations to benefit from both engineered recommendation systems and LLM-derived reasoning graphs. We demonstrate the effectiveness of LLMRG on benchmarks and real-world scenarios in enhancing base recommendation models.

Do Trust based Social Recommendation Algorithms Work as Intended?

Recommender systems are powerful tools that filter and recommend content/information relevant to a given user. Collaborative filtering is the most popular technique used in building recommender systems and it has been successfully incorporated in many applications. These conventional recommendation systems require a minimum number of users, items, and ratings in order to provide effective recommendations. This results in the infamous cold-start problem where the system is not able to produce effective recommendations for new users. Recently, there has been an escalation in the popularity and usage of social networks, which persuades people to share their experiences in the form of reviews and ratings on social media. The components of social media such as the influence of friends, interests, and enjoyment create the opportunities to develop solutions for sparsity and cold start problems of recommendation systems. This paper aims to observe these patterns and analyze three of the existing social recommendation systems, SocialMF, SocialFD, and GraphRec. SocialMF and SocialFD algorithms are based on matrix factorization and distance metric learning respectively whereas GraphRec is an attention based deep learning model. Through extensive experimentation with the datasets that these algorithms were tested on and one new dataset, we compared the results based on evaluation metrics including Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). To investigate how trust impacts the performance of these models, we evaluated them by modifying the trust and social component. Experimental results show that there is no conclusive evidence that trust propagation plays a major part in these models. Moreover, these models show a slightly improved performance in the absence of trust statements.

Prompt-based and weak-modality enhanced multimodal recommendation

Beyond conventional recommendation systems that rely merely on user-item interaction data, multimodal recommendation systems additionally exploit the item multimodal data for boosting the recommendation performance. In this research line, late fusion-based approaches that first predict user ratings for each item modality independently and then merge these predictions for a final user rating have made significant advancements. Nevertheless, these methods still have the following two issues: (1) they utilize individual user embeddings to model user interest in different modalities, while overlooking the underlying relationship among modalities and significantly increasing the memory costs; and (2) they overlook the unreliable interest learned from certain modality, thus hindering the accurate final rating learning. To address these issues, we propose a prompt-based and weak-modality enhanced multimodal recommendation framework. It consists of two key components: (1) multimodal prompted user interest learning that adopts a single user embedding with different modality prompts to model different modality-specific user interests, and (2) weak-modality enhanced training that enhances the user interest learning in modalities where the predictions are less unreliable, ensuring well-balanced learning across all modalities. Extensive experiments on Amazon datasets have demonstrated the effectiveness of the proposed framework. The two components deployed onto existing methods help to make them more effective and efficient.

Hierarchical Reinforcement Learning for Conversational Recommendation With Knowledge Graph Reasoning and Heterogeneous Questions

User interaction history with items is used to infer user preferences in conventional recommendation systems. Among these, conversational recommendation systems (CRSs), which provide effective recommendations based on a framework combining recommendations and conversations with users, have been proposed. However, existing CRS model still have some shortcomings, such as the lack of using dialogue records for recommendations. In this research, a method was proposed to solve multiround conversational recommendations with heterogeneous questions. The model included a hierarchical reinforcement learning framework and introduced methods of effectively incorporating user feedback for online recommender updates. Experiments were conducted on several datasets to verify model effectiveness; the model outperformed the current state-of-the-art method. Finally, the architecture was more realistic than other conversational recommendation scenarios and provided richer explanations.

Mashup-Oriented Web API Recommendation Via Full-Text Semantic Mining of Developer Requirements

API recommendation differs from conventional recommendation systems in that it integrates the characteristics of development requirements, mashup, and API, where there are two primary challenges, i.e., how to effectively mine the personalized development requirements of developers and the sparseness of interaction data. This research proposes a novel model framework, called Bayesian Probabilistic Matrix Factorization Model with Text Similarity and Adversarial Training (SAMF), to address these two issues. Utilizing natural language processing technology to extract the full-text semantics of requirements documents, mashup descriptions, and API descriptions, fully mining personalized development requirements, and contextual information of mashups and APIs, and calculating text similarity, is our fundamental idea. Simultaneously, the collaborative filtering method is used to mine user needs, mashup, and API information from historical data, and the obtained text similarity is added to the training as auxiliary information. Furthermore, adversarial training is further incorporated to supplement data in order to minimize data sparsity and enhance model robustness and generalization. To assess the performance of SAMF, we run thorough experiments that illustrate the efficacy of each module of the model and explain the influence of hyperparameter settings. Specifically, compared to the baseline, the experimental findings demonstrate that SAMF can basically achieve better performance.

INTELLIGENT MOVIE RECOMMENDER FRAMEWORK BASED ON CONTENT-BASED & COLLABORATIVE FILTERING ASSISTED WITH SENTIMENT ANALYSIS

Recommendation systems for rating movies and forming opinion have grown exponentially in popularity, they make it very convenient for consumers to select films that suit their tastes. However, traditional recommendation systems often rely solely on user ratings or reviews, which may not accurately reflect the user's true feelings about a movie. To address this issue, sentiment analysis has been proposed as a more reliable method for capturing emotional information about movies. In this research paper, we propose a novel movie recommendation system that combines sentiment analysis with collaborative filtering and content-based methods. Our system is designed to provide accurate and timely recommendations to mobile users based on their preferences, reviews, and emotions. We evaluate the performance of our system using real-world data and demonstrate its effectiveness in improving the accuracy and timeliness of movie recommendations. In this research paper, we propose a comparative study of three popular movie recommendation techniques: content-based filtering, collaborative filtering, and sentiment analysis. We aim to evaluate the effectiveness of each approach in providing accurate and personalized movie recommendations. Work done has utilized MovieLens dataset for this study, which is a popular benchmark dataset for assessing movie recommendation systems. The dataset includes almost 100,000 ratings, ranging from 1 to 5, from 943 individuals on 1,682 films. Each rating has a timestamp, a special user ID, and a movie ID. The collection also contains details about each movie's genre. To prepare the dataset for our experiments, we first performed some data cleaning and pre-processing. We removed any duplicate ratings, and any movies or users with a low number of ratings were also removed. We also performed some feature engineering to extract relevant features from the raw data, such as movie genres, user demographics, and movie release dates. Our tests on the Movielens dataset show that the system we've suggested works well. Precision, recall, and F1-score were some of the measures we used to assess the system's performance. We discovered that it performed better than conventional recommendation systems that just use content-based or collaborative filtering techniques. A better user experience was produced with the use of sentiment analysis, which improved the accuracy and promptness of suggestions. Algorithm designed achieves a precision of 0.875, Recall of 0.0435, F1 Score of 0.0828, Accuracy of 0.9889, RMSE of 0.407, and F measure of 0.1813.

Group recommender systems for tourism: how does personality predict preferences for attractions, travel motivations, preferences and concerns?

To travel in leisure is an emotional experience, and therefore, the more the information about the tourist is known, the more the personalized recommendations of places and attractions can be made. But if to provide recommendations to a tourist is complex, to provide them to a group is even more. The emergence of personality computing and personality-aware recommender systems (RS) brought a new solution for the cold-start problem inherent to the conventional RS and can be the leverage needed to solve conflicting preferences in heterogenous groups and to make more precise and personalized recommendations to tourists, as it has been evidenced that personality is strongly related to preferences in many domains, including tourism. Although many studies on psychology of tourism can be found, not many predict the tourists’ preferences based on the Big Five personality dimensions. This work aims to find how personality relates to the choice of a wide range of tourist attractions, traveling motivations, and travel-related preferences and concerns, hoping to provide a solid base for researchers in the tourism RS area to automatically model tourists in the system without the need for tedious configurations, and solve the cold-start problem and conflicting preferences. By performing Exploratory and Confirmatory Factor Analysis on the data gathered from an online questionnaire, sent to Portuguese individuals from different areas of formation and age groups (n = 1035), we show all five personality dimensions can help predict the choice of tourist attractions and travel-related preferences and concerns, and that only neuroticism and openness predict traveling motivations.

Distributed Improved Deep Prediction for Recommender System using an Ensemble Learning

If online businesses possess valuable interest for suggesting their items by scoring them, then digital advertising gains their profits depending on their promotions or marketing task. Web users cannot be certain that the products handled via big-data recommendation are either advanced or interesting to their needs. In recent decades, recommender system models have been widely used to analyses large quantities of information. Amongst, a Distributed Improved Prediction with Matrix Factorization (MF) and Random Forest (RF) called DIPMF model exploits individual’s desires, choices and social context together for predicting the ratings of a particular item. But, the RF scheme needs high computation power and time for learning process. Also, its outcome was influenced by the training parameters. Hence this article proposes a Distributed Improved Deep Prediction with MF and ensemble learning (DIDPMF) model is proposed to decrease the computational difficulty of RF learning and increasing the efficiency of rating prediction. In this DIDPMF, a forest attribute extractor is ensemble with the Deep Neural Network (fDNN) for extracting the sparse attribute correlations from an extremely large attribute space. So, incorporating RF over DNN has the ability to provide prediction outcomes from all its base trainers instead of a single estimated possibility rate. This fDNN encompasses forest module and DNN module. The forest module is employed as an attribute extractor to extract the sparse representations from the given raw input data with the supervision of learning outcomes. First, independent decision trees are constructed and then ensemble those trees to obtain the forest. After, this forest is fed to the DNN module which acts as a learner to predict the individual’s ratings with the aid of novel attribute representations. Finally, the experimental results reveal that the DIDPMF outperforms than the other conventional recommender systems.

An approach for ontology-based research and recommendation on systems engineering projects

Recommendation systems (RS) have dramatically evolved these past few years, generating more and more accurate results and developing innovative filtering methods. A huge breakthrough was the integration of ontologies into their recommendation process. It integrated the domain's knowledge into its reasoning process, thus overcoming the limitations of the conventional RS. The main purpose of this paper is to explore the application of ontology-based RS within critical infrastructure projects implementing the systems engineering or model-based engineering approach.

A Multi-Objective Optimization Framework for Multi-Stakeholder Fairness-Aware Recommendation

Nowadays, most online services are hosted on multi-stakeholder marketplaces, where consumers and producers may have different objectives. Conventional recommendation systems, however, mainly focus on maximizing consumers’ satisfaction by recommending the most relevant items to each individual. This may result in unfair exposure of items, thus jeopardizing producer benefits. Additionally, they do not care whether consumers from diverse demographic groups are equally satisfied. To address these limitations, we propose a multi-objective optimization framework for fairness-aware recommendation, Multi-FR , that adaptively balances accuracy and fairness for various stakeholders with Pareto optimality guarantee. We first propose four fairness constraints on consumers and producers. In order to train the whole framework in an end-to-end way, we utilize the smooth rank and stochastic ranking policy to make these fairness criteria differentiable and friendly to back-propagation. Then, we adopt the multiple gradient descent algorithm to generate a Pareto set of solutions, from which the most appropriate one is selected by the Least Misery Strategy. The experimental results demonstrate that Multi-FR largely improves recommendation fairness on multiple stakeholders over the state-of-the-art approaches while maintaining almost the same recommendation accuracy. The training efficiency study confirms our model’s ability to simultaneously optimize different fairness constraints for many stakeholders efficiently.

Double Attention Convolutional Neural Network for Sequential Recommendation

The explosive growth of e-commerce and online service has led to the development of recommender system. Aiming to provide a list of items to meet a user’s personalized need by analyzing his/her interaction 1 history, recommender system has been widely studied in academic and industrial communities. Different from conventional recommender systems, sequential recommender systems attempt to capture the pattern of users’ sequential behaviors and the evolution of users’ preferences. Most of the existing sequential recommendation models only focus on user interaction sequence, but neglect item interaction sequence. An item interaction sequence also contains rich contextual information for capturing the item’s dynamic characteristic, since an item’s dynamic characteristic can be reflected by the users who interact with it in a period. Furthermore, existing dual sequential models use the same method to handle the user interaction sequence and item interaction sequence, and do not consider their different characteristics. Hence, we propose a novel D ouble A ttention C onvolution N eural N etwork (DACNN) , which incorporates user interaction sequence and item interaction sequence into an integrated neural network framework. DACNN leverages the strength of attention mechanism to capture the temporary suitability and adopts CNN to extract local sequential features. Experimental evaluations on the real datasets show that DACNN outperforms the baseline approaches.

Ethical and legal considerations for nutrition virtual coaches

Choices and preferences of individuals are nowadays increasingly influenced by countless inputs and recommendations provided by artificial intelligence-based systems. The accuracy of recommender systems (RS) has achieved remarkable results in several domains, from infotainment to marketing and lifestyle. However, in sensitive use-cases, such as nutrition, there is a need for more complex dynamics and responsibilities beyond conventional RS frameworks. On one hand, virtual coaching systems (VCS) are intended to support and educate the users about food, integrating additional dimensions w.r.t. the conventional RS (i.e., leveraging persuasion techniques, argumentation, informative systems, and recommendation paradigms) and show promising results. On the other hand, as of today, VCS raise unexplored ethical and legal concerns. This paper discusses the need for a clear understanding of the ethical/legal-technological entanglements, formalizing 21 ethical and ten legal challenges and the related mitigation strategies. Moreover, it elaborates on nutrition sustainability as a further nutrition virtual coaches dimension for a better society.

An automated system recommending background music to listen to while working

Many people listen to music while working nowadays. However, conventional recommendation systems that are designed for playing songs matching user preferences cannot be applied for such a situation. This is because previous research showed that listeners’ concentration can be negatively affected not only by music that listeners strongly dislike but also by music that the listeners strongly like. Therefore, when we consider a recommendation system to be used while working, it is desirable to avoid both songs the user likes very much and songs the user dislikes very much. Given this background, we propose FocusMusicRecommender, a system designed specifically for recommending music to listen to while working. It summarizes songs automatically and plays them successively in order to enable users to give not only “dislike (very much)” feedback via a “skip” button but also “like (very much)” feedback via a “keep listening” button. The feedback is then combined with the users’ concentration level that is estimated from their behavioral history during the playback of the corresponding song, which allows the system to obtain preference information that distinguishes between “like” and “like very much” without burdening the user who is working. Based on the preference information, the system estimates the preference levels of unplayed songs and prioritizes the songs for subsequent playback by also considering the user’s current concentration level. Our experiments showed the validity and effectiveness of the proposed method, including the accuracy of the concentration level estimation. Moreover, our user study verified the suitability of the recommendation results from both the observed behavior and obtained comments of the participants.

Content-Based Collaborative Filtering with Hierarchical Agglomerative Clustering Using User/ Item based Ratings

The recommender system (RS) plays the major role in online networks, online shopping, and online services etc. The conventional RSs are suffering with the inaccurate quality of experience to the users, so the improper content is recommending to customers. The content based collaborative filtering (CBCF) method is introduced to solve the issues presented in the RSs. However, the CBCF method is suffering with the cold start problem for new users and suffering with data accuracy, data sparsity, and scalable data in clustering process. Thus, to solve these problems, this article proposes hierarchical agglomerative clustering (HAC) based collaborative filtering (HAC-CF) for RSs. The proposed HAC-CF based RS functions by utilizing the incentivized/ penalized user (IPU) model with user-based and item-based ratings. To this end, users are divided into several clusters through single link graph partitioning through minimum distance criteria. Then, the final item ranking is computed using Pearson correlation coefficient (PCC) similarity of users. Hence, recommendation efficiency and accuracy are increased at the end user by combining user, item models. The simulation results show the performance enhancement of proposed method with respect to F1-score, recall, and precision as compared to the conventional approaches.

Session-based social and dependency-aware software recommendation

With the increase of complexity of modern software, social collaborative coding and reuse of open source software packages become more and more popular, which thus greatly enhances the development efficiency and software quality. However, the explosive growth of open source software packages exposes developers to the challenge of information overload. While this can be addressed by conventional recommender systems, they usually do not consider particular constraints of social coding such as social influence among developers and dependency relations among software packages. In this paper, we aim to model the dynamic interests of developers with both social influence and dependency constraints, and propose the Session-based Social and Dependency-aware software Recommendation (SSDRec) model. This model integrates recurrent neural network (RNN) and graph attention network (GAT) into a unified framework. An RNN is employed to model the short-term dynamic interests of developers in each session and two GATs are utilized to capture social influence from friends and dependency constraints from dependent software packages, respectively. Extensive experiments are conducted on real-world datasets and the results demonstrate that our model significantly outperforms the competitive baselines.

A survey on personality-aware recommendation systems

With the emergence of personality computing as a new research field related to artificial intelligence and personality psychology, we have witnessed an unprecedented proliferation of personality-aware recommendation systems. Unlike conventional recommendation systems, these new systems solve traditional problems such as the cold start and data sparsity problems. This survey aims to study and systematically classify personality-aware recommendation systems. To the best of our knowledge, this survey is the first that focuses on personality-aware recommendation systems. We explore the different design choices of personality-aware recommendation systems, by comparing their personality modeling methods, as well as their recommendation techniques. Furthermore, we present the commonly used datasets and point out some of the challenges of personality-aware recommendation systems.

Personalizing Diversity Versus Accuracy in Session-Based Recommender Systems

One of the most important concerns about recommender systems is the filter bubble phenomenon. While recommender systems try to personalize information, they tighten the filter bubble around the users and deprive them of a wide range of content. To overcome this problem, one can diversify the personalized recommendation list. A diversified list usually presents a broader content to the user. Session-based recommender systems are types of recommenders in which only the current session of the user is available, and therefore, they should recommend the next item given the items in the current session. While diversifying conventional recommender systems has been well assessed in the literature, it has gained less attention in session-based recommenders. Diversity and accuracy usually have a negative correlation, i.e., by improving one the other one will be declined. In this study, we propose diversity and accuracy enhancing approaches based on sequential rule mining and session-based k-nearest neighbor methods. Finally, we propose a performance balancing approach that improves both the diversity and accuracy of these session-based recommender systems. We demonstrate the performance of the proposed methods on four music recommender datasets.

Recommenders in improving students’ engagement in large scale open learning

Technology Enhanced Learning recommender systems (TEL RS) have become an attractive research area in the recent decade as they promote users’ selection process within limited time in educational domain. Actually, they have the ability to support e-learning through personalization of the learning process whether for teachers’ requirements or for individual students’ needs. Conventional recommender systems have proposed various methods focusing on recommendations to individual learners. Recently, due to significant increase in students’ number, especially in the field of massive open online courses (MOOCs) and regarding their diversity, the need of offering adaptation is becoming more important to help improve learning quality. For that reason, several recommender systems have been developed to adapt learning to personal students’ needs in MOOC context. This paper reports and discuss recommendations in large scale open learning for improving students’ engagement. We present an analysis and a comparison between TEL RS in the context of MOOCs for different purposes. Then, we present our proposed recommender system of learning activities in accordance with a set of comparison criteria.