Recommender System Model Research Articles

Transforming Personalized Travel Recommendations: Integrating Generative AI with Personality Models

Over the past few years, the incorporation of generative Artificial Intelligence (AI) techniques, particularly the Retrieval-Augmented Generator (RAG) framework, has opened up revolutionary opportunities for improving personalized travel recommendation systems. The RAG framework seamlessly combines the capabilities of large-scale language models with retriever models, facilitating the generation of diverse and contextually relevant recommendations tailored to individual preferences and interests, all of which are based on natural language queries. These systems iteratively learn and adapt to user feedback, thereby continuously refining and improving recommendation quality over time. This dynamic learning process enables the system to dynamically adjust to changes in user preferences, emerging travel trends, and contextual factors, ensuring that the recommendations remain pertinent and personalized. Furthermore, we explore the incorporation of personality models like the Myers–Briggs Type Indicator (MBTI) and the Big Five (BF) personality traits into personalized travel recommendation systems. By incorporating these personality models, our research aims to enrich the understanding of user preferences and behavior, allowing for even more precise and tailored recommendations. We explore the potential synergies between personality psychology and advanced AI techniques, specifically the RAG framework with a personality model, in revolutionizing personalized travel recommendations. Additionally, we conduct an in-depth examination of the underlying principles, methodologies, and technical intricacies of these advanced AI techniques, emphasizing their ability to understand natural language queries, retrieve relevant information from vast knowledge bases, and generate contextually rich recommendations tailored to individual personalities. In our personalized travel recommendation system model, results are achieved such as user satisfaction (78%), system accuracy (82%), and the performance rate based on user personality traits (85% for extraversion and 75% for introversion).

CONTENT-BASED FILTERING CULINARY RECOMMENDATION SYSTEM USING DEEP CONVOLUTIONAL NEURAL NETWORK ON TWITTER (X)

Along with the development of technology, social media has become integral to everyday life, especially for sharing content like culinary reviews. Social media platform X (formerly Twitter) is often used for sharing culinary recommendations, but the abundance of information makes it difficult for users to find relevant suggestions. In order to improve rating prediction performance, this study suggests a recommendation system model that is more thoroughly created utilizing Content-Based Filtering (CBF) combined with Deep Convolutional Neural Network (CNN) and optimised with Particle Swarm Optimization (PSO). Data was collected from PergiKuliner and Twitter, totaling 2644 reviews and 200 cuisines. The preprocessing involved text processing, translation, and polarity assessment. Post-labeling, 7438 data were labeled with 0 and 1562 with 1. Label 0 means not recommended while label 1 means recommended. The imbalance is handled by applying the SMOTE method after observing that the fraction of data labeled 0 and 1 is 65.2%. CBF employed TF-IDF feature extraction and FastText word embedding, while Deep CNN handled classification. PSO optimisation was applied to enhance the accuracy of culinary rating predictions. The results showed an initial accuracy of 76.32% with the baseline Deep CNN model, which increased to 86.06% after Nadam optimisation with the best learning rate, and further reached 86.18% after PSO optimisation on dense units. The 9.86% accuracy improvement from the baseline model demonstrates the effectiveness of the combined methods.

A Study of Recommendation Methods Based on Graph Hybrid Neural Networks and Deep Crossing

In the face of complex user behavior patterns and massive data, improving the performance of recommender system models is an urgent challenge. Traditional methods often struggle to effectively handle feature interactions and complex user-item relationships. Combining the advantages of graph neural networks and the Deep Crossing network, this paper proposes a recommendation method based on hybrid neural networks with Deep Crossing (Deep Crossing with Graph Convolution and GRU, DCGCN-GRU). First, by constructing the graph structure of users and items, higher-order feature representations are extracted, and node features are updated using a multilayer graph convolution operation. Then, the higher-order features learned by the graph convolution network are spliced and weighted with the original features to form new feature inputs. Next, a Gated Recurrent Unit (GRU) is introduced to capture the inter-feature temporal dynamic relationships and sequence information. Finally, the Deep Crossing model is utilized to learn the interactions between the fused features at multiple levels and enhance the interactions between the features. Comparative experiments on three public datasets, MovieLens-ml-25m, Book-Crossings, and Amazon Reviews’23, show that the model achieves significant improvements in accuracy, mean square error (MSE), and mean absolute error (MAE).

Our Model Achieves Excellent Performance on MovieLens: What Does It Mean?

A typical benchmark dataset for recommender system (RecSys) evaluation consists of user-item interactions generated on a platform within a time period. The interaction generation mechanism partially explains why a user interacts with (e.g., like, purchase, rate) an item, and the context of when a particular interaction happened. In this study, we conduct a meticulous analysis of the MovieLens dataset and explain the potential impact of using the dataset for evaluating recommendation algorithms. We make a few main findings from our analysis. First, there are significant differences in user interactions at the different stages when a user interacts with the MovieLens platform. The early interactions largely define the user portrait which affect the subsequent interactions. Second, user interactions are highly affected by the candidate movies that are recommended by the platform's internal recommendation algorithm(s). Third, changing the order of user interactions makes it more difficult for sequential algorithms to capture the progressive interaction process. We further discuss the discrepancy between the interaction generation mechanism that is employed by the MovieLens system and that of typical real-world recommendation scenarios. That is, the MovieLens dataset records \(\langle user-MovieLens\rangle\) interactions, but not \(\langle user-movie\rangle\) interactions. All research articles using the MovieLens dataset model the \(\langle user-MovieLens\rangle\) rather than the \(\langle user-movie\rangle\) interactions, making their results less generalizable to many practical recommendation scenarios in real-world settings. In summary, the MovieLens platform demonstrates an efficient and effective way of collecting user preferences to address cold-starts. However, models that achieve excellent recommendation accuracy on the MovieLens dataset may not demonstrate superior performance in practice , for at least two kinds of differences: (1) the differences in the contexts of user-item interaction generation, and (2) the differences in user knowledge about the item collections. While results on MovieLens can be useful as a reference, they should not be solely relied upon as the primary justification for the effectiveness of a recommendation system model.

Enhancing Recommendation Diversity and Novelty with Bi-LSTM and Mean Shift Clustering

In the digital age, personalized recommendation systems have become crucial for information dissemination and user experience. While traditional systems focus on accuracy, they often overlook diversity, novelty, and serendipity. This study introduces an innovative recommendation system model, Time-based Outlier Aware Recommender (TOAR), designed to address the challenges of content homogenization and information bubbles in personalized recommendations. TOAR integrates Neural Matrix Factorization (NeuMF), Bidirectional Long Short-Term Memory Networks (Bi-LSTM), and Mean Shift clustering to enhance recommendation accuracy, novelty, and diversity. The model analyzes temporal dynamics of user behavior and facilitates cross-domain knowledge exchange through feature sharing and transfer learning mechanisms. By incorporating an attention mechanism and unsupervised clustering, TOAR effectively captures important time-series information and ensures recommendation diversity. Experimental results on a news recommendation dataset demonstrate TOAR’s superior performance across multiple metrics, including AUC, precision, NDCG, and novelty, compared to traditional and deep learning-based recommendation models. This research provides a foundation for developing more intelligent and personalized recommendation services that balance accuracy with content diversity.

Artificial Intelligence Recommendation System Model for Military Cabin Space Environment Design

This paper proposes an artificial intelligence-based recommendation model for military cabin space environment design schemes to satisfy military cabin operators with a comfortable and efficient working environment and improve the safety and health of the operators. Each key design feature is used to extract the interactive relationship between the operator and the cabin design. Through the neural network, the operator's psychological comfort is interactively matched with the cabin space environment factors. As result, the interactive artificial intelligence cabin space environment design scheme recommendation system is recommended and established. The results showed that the method can optimize the environmental design of military cabins, alleviate some of the design complexities associated with military cabins, and improve the physical and mental health of operators. Additionally, by using an AI-based recommendation system for cabin space environment design, engineers can reduce the time and resources required to design military cabins, resulting in faster turnaround times and lower costs.

A neural network-based collaborative filtering model for social recommendation systems

A neural network-based collaborative filtering model for social recommendation systems

Recommendation system model based on graph neural network combined with support vector machine and comparative learning for filtering

Recommendation system model based on graph neural network combined with support vector machine and comparative learning for filtering

MODEL FOR DATA OBJECTS SELECTION BY SEARCH IMAGE FOR INTELLIGENT RECOMMENDER SYSTEMS

The research is conducted to develop and analyze an object filtering model for intelligent recommender systems. The main objective is to solve the problem of orientation in the vast amounts of information accumulated by mankind. The aim of the research is to create an effective tool for systematization and knowledge management, which in turn contributes to the optimization of decision-making processes and interaction with data. The paper focuses on the research and development of an object filtering model for intelligent recommender systems. Within the methodology and research area, the developed model is described in detail and the theoretical and practical aspects of the methodology are analyzed. In this paper, a variant of the problem statement of the research and development of an object filtering model for intelligent recommender systems is presented. In addition, the paper deconstructs the second stage of this problem, emphasizing its importance in the context of achieving system performance. The results of the study are analyzed in detail, highlighting the key points and features of the proposed model. The scope of the study is reviewed, detailing the prospects of applying the results in scientific and practical applications, providing the reader with a deeper understanding of the potential of the proposed model. The object filtering model has a high potential of usefulness for business and manufacturing. This work will be useful for developers and researchers of recommender systems in which users rarely or never interact with the same object.

Toward Bias-Agnostic Recommender Systems: A Universal Generative Framework

User behavior data, such as ratings and clicks, has been widely used to build personalizing models for recommender systems. However, many unflattering factors (e.g., popularity, ranking position, users’ selection) significantly affect the performance of the learned recommendation model. Most existing work on unbiased recommendation addressed these biases from sample granularity (e.g., sample reweighting, data augmentation) or from the perspective of representation learning (e.g., bias-modeling). However, these methods are usually designed for a specific bias, lacking the universal capability to handle complex situations where multiple biases co-exist. Besides, rare work frees itself from laborious and sophisticated debiasing configurations (e.g., propensity scores, imputed values, or user behavior-generating process). Towards this research gap, in this article, we propose a universal G enerative framework for B ias D isentanglement termed as GBD , constantly generating calibration perturbations for the intermediate representations during training to keep them from being affected by the bias. Specifically, a bias-identifier that tries to retrieve the bias-related information from the representations is first introduced. Subsequently, the calibration perturbations are generated to significantly deteriorate the bias-identifier’s performance, making the bias gradually disentangled from the calibrated representations. Therefore, without relying on notorious debiasing configurations, a bias-agnostic model is obtained under the guidance of the bias identifier. We further present its universality by subsuming the representative biases and their mixture under the proposed framework. Finally, extensive experiments on the real-world, synthetic, and semi-synthetic datasets have demonstrated the superiority of the proposed approach against a wide range of recommendation debiasing methods. The code is available at https://github.com/Zhidan-Wang/GBD .

Fashion Fusion: Exploring Apparel Recommendation Systems across companies using Machine Learning Approach

Recommendation technology is an advanced technology which gives users an improved service to get more information about a product. Fashion recommendation system is same as having a personal stylist who can recommend products based on user's preference. It uses data from users shopping history to recommend the trending outfits for making shopping easier. This paper aims to analyse a model for fashion recommendation system using K-Nearest Neighbor (KNN). Analysis of the services provided to the consumer like category, gender, and brand across companies should be relevant which helps in improving consumer overall shopping experience. Using Content-based Filtering the products are filtered based on user’s preference and out of all recommended products the one with highest rating product is predicted using Random Forest Regressor. The comparison between algorithms is made to take best of all, to enhance the performances.

Deep learning model for recommendation system using web of things based knowledge graph mining

Deep learning model for recommendation system using web of things based knowledge graph mining

Dual-blend insight recommendation system for e-commerce recommendations and enhance personalization

E-commerce, short for electronic commerce, refers to the buying and selling of goods and services over the internet. This digital transaction model has revolutionized the way businesses operate and consumers shop. In response to the burgeoning complexity of e-commerce datasets, this work addresses the need for advanced recommendation systems. This work introduces the dual-blend insight recommendation system (DIRS) model for personalized e-commerce recommendation system. The DIRS model involves dataset loading, preprocessing, and feature extraction, enabling training with recurrent neural network (RNN) and Bayesian personalized ranking (BPR) models. Recommendations are generated based on user-defined functions, i.e., location and session, and evaluation metrics such as hit rate (HR) and mean reciprocal rate (MRR) highlight DIRS’s superior performance. The model is evaluated using the Tmall dataset. Results reveal DIRS consistently outperforms alternative algorithms, showcasing its effectiveness in 10k and 20k recommendation sets. This study provides valuable insights into optimizing e-commerce recommendations, emphasizing DIRS as a powerful model for enhancing user experience and engagement.

Optimal hybrid classification model for event recommendation system

There is a growing need for recommender systems and other ML-based systems as an abundance of data is now available across all industries. Various industries are currently using recommender systems in slightly different ways. These programs utilize algorithms to propose appropriate products to consumers based on their prior choices and interactions. Moreover, Systems for recommending events to users suggest pertinent happenings that they might find interesting. As opposed to an object recommender that suggests books or movies; event-based recommender systems typically require distinct algorithms. A developed event recommendation method is introduced which includes two stages: feature extraction and recommendation. In stage, I, a Set of features like personal willingness, community willingness, informative content, edge weight, and node interest degree are extracted. Stage II of the event recommendation system performs a hybrid classification by combining LSTM and CNN. In the LSTM classifier, optimal tuning is done by Improvised Cat and Mouse optimization (ICMO) algorithm. The results of the ICMO technique at an 80% training percentage have the maximum sensitivity value of 95.19%, whereas those of the existing approaches SSA, DINGO, BOA, and CMBO have values of 93.89%, 93.35%, 92.36%, and 92.24%. Finally, the best result is then determined by evaluating the whole performance.

SSGCL: Simple Social Recommendation with Graph Contrastive Learning

As user–item interaction information is typically limited, collaborative filtering (CF)-based recommender systems often suffer from the data sparsity issue. To address this issue, recent recommender systems have turned to graph neural networks (GNNs) due to their superior performance in capturing high-order relationships. Furthermore, some of these GNN-based recommendation models also attempt to incorporate other information. They either extract self-supervised signals to mitigate the data sparsity problem or employ social information to assist with learning better representations under a social recommendation setting. However, only a few methods can take full advantage of these different aspects of information. Based on some testing, we believe most of these methods are complex and redundantly designed, which may lead to sub-optimal results. In this paper, we propose SSGCL, which is a recommendation system model that utilizes both social information and self-supervised information. We design a GNN-based propagation strategy that integrates social information with interest information in a simple yet effective way to learn user–item representations for recommendations. In addition, a specially designed contrastive learning module is employed to take advantage of the self-supervised signals for a better user–item representation distribution. The contrastive learning module is jointly optimized with the recommendation module to benefit the final recommendation result. Experiments on several benchmark data sets demonstrate the significant improvement in performance achieved by our model when compared with baseline models.

Hybrid recommender system model for digital library from multiple online publishers

Background The demand for online education promotion platforms has increased. In addition, the digital library system is one of the many systems that support teaching and learning. However, most digital library systems store books in the form of libraries that were developed or purchased exclusively by the library, without connecting data with different agencies in the same system. Methods A hybrid recommender system model for digital libraries, developed from multiple online publishers, has created a prototype digital library system that connects various important knowledge sources from multiple digital libraries and online publishers to create an index and recommend e-books. The developed system utilizes an API-based linking process to connect various important sources of knowledge from multiple data sources such as e-books on education from educational institutions, e-books from government agencies, and e-books from religious organizations are stored separately. Then, a hybrid recommender system suitable for users was developed using Collaborative Filtering (CF) model together with Content-Based Filtering. This research purposed the hybrid recommender system model, which took into account the factors of book category, reading habits of users, and sources of information. The evaluation of the experiments involved soliciting feedback from system users and comparing the results with conventional recommendation methods. Results A comparison of NDCG scores was conducted for Hybrid Score 50:50, Hybrid Score 20:80, Hybrid Score 80:20, CF-score and CB-score. The experimental result was found that the Hybrid Score 80:20 method had the highest average NDCG score. Conclusions Using a hybrid recommender system model that combines 80% Collaborative Filtering and 20% Content-Based Filtering can improve the recommender method, leading to better referral efficiency and greater overall efficiency compared to traditional approaches.

Dynamic matrix factorization-based collaborative filtering in movie recommendation services

Movies are a primary source of entertainment, but finding specific content can be challenging given the exponentially increasing number of movies produced each year. Recommendation systems are extremely useful for solving this problem. While various approaches exist, Collaborative Filtering (CF) is the most straightforward. CF leverages user input and historical preferences to determine user similarity and suggest movies. Matrix Factorization (MF) is one of the most popular Collaborative Filtering (CF) techniques. It maps users and items into a joint latent space, using a vector of latent features to represent each user or item. However, traditional MF techniques are static, while user cognition and product variety are constantly evolving. As a result, traditional MF approaches struggle to accommodate the dynamic nature of user-item interactions. To address this challenge, we propose a Dynamic Matrix Factorization CF model for movie recommendation systems (DMF-CF) that considers the dynamic changes in user interactions. To validate our approach, we conducted evaluations using the standard MovieLens dataset and compared it to state-of-the-art models. Our preliminary findings highlight the substantial benefits of DMF-CF, which outperforms recent models on the MovieLens-100K and MovieLens-1M datasets in terms of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) metrics.

MAInt: A multi-task learning model with automatic feature interaction learning for personalized recommendations

Nowadays, more and more recommender systems apply multi-task learning to provide users with more accurate personalized recommendations. However, most of the existing multi-task learning models ignore the importance of explicit feature interactions in different tasks, which are an important part of recommender system models. In addition, the knowledge sharing mechanism of the existing model is difficult to meet the needs of all tasks, resulting in the performance degradation of some tasks. To address these issues, we propose a Multi-task learning model based on Automatic feature Interaction learning (MAInt). The Tasks Interaction Layer in MAInt automatically learns explicit feature interactions in different tasks through self-attention neural networks. At the same time, this process allows each task to interact with each other, so that each task can adaptively extract helpful information from other task features. We conducted multiple experiments on two popular public datasets: Census-income Dataset and Ali-CCP Dataset, and the experimental results verified the effectiveness of our proposed model.

Unbiased, Effective, and Efficient Distillation from Heterogeneous Models for Recommender Systems

In recent years, recommender systems have achieved remarkable performance by using ensembles of heterogeneous models. However, this approach is costly due to the resources and inference latency proportional to the number of models, creating a bottleneck for production. Our work aims to transfer the ensemble knowledge of heterogeneous teachers to a lightweight student model using knowledge distillation (KD), reducing inference costs while maintaining high accuracy. We find that the efficacy of distillation decreases when transferring knowledge from heterogeneous teachers. To address this, we propose a new KD framework, named HetComp, that guides the student model by transferring easy-to-hard sequences of knowledge generated from teachers’ trajectories. HetComp uses dynamic knowledge construction to provide progressively difficult ranking knowledge and adaptive knowledge transfer to gradually transfer finer-grained ranking information. Although HetComp improves accuracy, it exacerbates popularity bias, resulting in a high popularity lift. To mitigate this issue, we introduce two strategies that leverage models’ disagreement knowledge (i.e., dissensus) for heterogeneous comparison. Our experiments demonstrate that HetComp significantly enhances distillation quality and the student model’s generalization capabilities. Furthermore, we provide extensive experimental results supporting the effectiveness of our dissensus-based debiasing techniques in mitigating the popularity lift caused by HetComp.

Investigation of frontier Multi-Armed Bandit algorithms and applications

Since the amount of content online is growing exponentially and people's time is limited, there is an urgent need for a high-performance algorithm that can make effective recommendations. This paper will introduce a recommendation system model, a sequential decision model, which is called Multi-Armed Bandit. The main idea of the Multi-Armed Bandit model is that at the beginning of the algorithm, all the recommended items are set to the same weight. In the subsequent recommendation process, the model explores the distribution of each item while changing the weight of each item according to the average revenue of each item, and selects more items with larger weight. This paper will introduce three cutting-edge Multi-Armed Bandit algorithms, their algorithmic ideas and their respective characteristics. The idea of Explore-Then-Commit (ETC) algorithm is to explore each item a certain number of times, and then select the best item for subsequent recommendation. The idea of the Upper Confidence Bound (UCB) algorithm is to represent the "exploration" and "exploitation" of each item by numerical values and add them to the UCB value, and select the item with the largest UCB value each time. The idea of TS is to first assume the distribution of each item, and then change the parameters of the distribution of each item based on the reward. At the end, this paper will introduce several scenarios where Multi-Armed Bandit algorithms can be used to give the reader an idea of how to use Multi-Armed Bandit algorithms.