Data Sparsity Problem Research Articles

A multi-criteria trust-enhanced collaborative filtering algorithm for personalized tourism recommendations

The exponential growth of online information has LED to significant challenges in navigating data overload, particularly in the tourism industry. Travelers are overwhelmed with choices regarding destinations, accommodations, dining, and attractions, making it difficult to select options that best meet their needs. Recommender systems have emerged as a promising solution to this problem, aiding users in decision-making by providing personalized suggestions based on their preferences. Traditional collaborative filtering (CF) methods, however, face limitations, such as data sparsity and reliance on single rating scores, which do not fully capture the complexity of user preferences. This study proposes a hybrid multi-criteria trust-enhanced CF (HMCTeCF) algorithm to improve the accuracy and robustness of tourism recommendations. HMCTeCF improves the quality of recommendations by integrating multi-criteria user preferences with trust relationships among users and between items. Experimental results using real-world datasets, including Restaurants-TripAdvisor and Hotels-TripAdvisor, demonstrate that HMCTeCF outperforms benchmark CF-based recommendation methods. It achieves higher prediction accuracy and coverage rate, effectively addressing the data sparsity problem. This innovative algorithm facilitates a more personalized and enriching travel experience, particularly in scenarios with limited user data. The findings highlight the importance of considering multiple criteria and trust relationships in developing robust recommendation systems for the tourism industry.

Question Embedding on Weighted Heterogeneous Information Network for Knowledge Tracing

Knowledge Tracing (KT) aims to predict students’ future performance on answering questions based on their historical exercise sequences. To alleviate the problem of data sparsity in KT, recent works have introduced auxiliary information to mine question similarity, resulting in the enhancement of question embeddings. Nonetheless, there remains a gap in developing an approach that effectively incorporates various forms of auxiliary information, including relational information (e.g., question-student , question-skill relation), relationship attributes (e.g., correctness indicating a student's performance on a question), and node attributes (e.g., student ability ). To tackle this challenge, the Similarity-enhanced Question Embedding (SimQE) method for KT is proposed, with its central feature being the utilization of weighted and attributed meta-paths for extracting question similarity. To capture multi-dimensional question similarity semantics by integrating multiple relations, various meta-paths are constructed for learning question embeddings separately. These embeddings, each encoding different similarity semantics, are then fused to serve the task of KT. To capture finer-grained similarity by leveraging the relationship attributes and node attributes on the meta-paths, the biased random walk algorithm is designed. In addition, the auxiliary node generation method is proposed to capture high-order question similarity. Finally, extensive experiments conducted on 6 datasets demonstrate that SimQE performs the best among 10 representative question embedding methods. Furthermore, SimQE proves to be more effective in alleviating the problem of data sparsity.

DKVMN&MRI: A new deep knowledge tracing model based on DKVMN incorporating multi-relational information.

Knowledge tracing is a technology that models students' changing knowledge state over learning time based on their historical answer records, thus predicting their learning ability. It is the core module that supports the intelligent education system. To address the problems of sparse input data, lack of interpretability and weak capacity to capture the relationship between exercises in the existing models, this paper build a deep knowledge tracing model DKVMN&MRI based on the Dynamic Key-Value Memory Network (DKVMN) that incorporates multiple relationship information including exercise-knowledge point relations, exercise-exercise relations, and learning-forgetting relations. In the model, firstly, the Q-matrix is utilized to map the link between knowledge points and exercises to the input layer; secondly, improved DKVMN and LSTM are used to model the learning process of learners, then the Ebbinghaus forgetting curve function is introduced to simulate the process of memory forgetting in learners, and finally, the prediction strategies of Item Response Theory (IRT) and attention mechanism are used to combine the similarity relationship between learners' knowledge state and exercises to calculate the probability that learners would correctly respond during the subsequent time step. Through extensive experiments on three real-world datasets, we demonstrate that DKVMN&MRI has significant improvements in both AUC and ACC metrics contrast with the latest models. Furthermore, the study provides explanations at both the exercise level and learner knowledge state level, demonstrating the interpretability and efficacy of the proposed model.

Synthesizing Landsat images using time series model-fitting methods for China’s coastal areas against sparse and irregular observations

ABSTRACT Long historical records and free accessibility have made Landsat data valuable for time-series analysis. However, Landsat time-series analysis is restricted for coastal areas due to the lack of sufficient numbers of clear images. The generation of synthetic Landsat images using model-fitting methods is accepted as an effective means of overcoming this problem. But for coastal areas, available observations are typically sparse and irregular, leading to ill-fitted models that distort synthetic Landsat images. In this study, we propose a linear harmonic model with different orders and implement it on an annual basis to generate synthetic Landsat images based on the Google Earth Engine (GEE) platform. First, we incorporated the available observations from adjacent years and filtered them by season to address the problems of data sparsity and irregularity. Then, we combined the threshold segmentation and ridge regression approaches to address the fake cloud problem, which sometimes contaminates synthetic Landsat images in summer. All cloud-free real Landsat images from five typical sites along China’s coasts during 1986–2020 were used to evaluate the accuracy and robustness of the synthetic Landsat images generated using our proposed methods. The principal results are as follows: (1) the R2 value of the linear harmonic model averaged across different Landsat bands and land cover types was 0.640, and the model was especially successful in simulating the surface reflectance in near NIR bands in forest and grassland areas; (2) 81.1% of the synthetic Landsat images covered by fake clouds were effectively restored, and there was the particular need to remove fake clouds for the synthetic Landsat images at low latitudes; (3) The mean absolute error for our synthetic Landsat images was 0.015, with the rate of 0.124; this was achieved under the clear-sky probability of only 36% and the average number of annual observations below 9, indicating a good performance. Compared with the continuous change detection and classification (CCDC) and seasonal median composite (SMC) methods, our proposed method offers advantages in both the accuracy and integrity of synthetic Landsat images. Our proposed method for synthesizing images also has potential for application to global coastal areas and other satellite datasets.

Cross-feature fusion speech emotion recognition based on attention mask residual network and Wav2vec 2.0

Speech Emotion Recognition (SER) has received widespread attention as a crucial way for understanding human emotional states. However, the impact of irrelevant information on speech signals and data sparsity limit the development of SER system. To address these issues, this paper proposes a framework that incorporates the Attentive Mask Residual Network (AM-ResNet) and the self-supervised learning model Wav2vec 2.0 to obtain AM-ResNet features and Wav2vec 2.0 features respectively, together with a cross-attention module to interact and fuse these two features. The AM-ResNet branch mainly consists of maximum amplitude difference detection, mask residual block, and an attention mechanism. Among them, the maximum amplitude difference detection and the mask residual block act on the pre-processing and the network, respectively, to reduce the impact of silent frames, and the attention mechanism assigns different weights to unvoiced and voiced speech to reduce redundant emotional information caused by unvoiced speech. In the Wav2vec 2.0 branch, this model is introduced as a feature extractor to obtain general speech features (Wav2vec 2.0 features) through pre-training with a large amount of unlabeled speech data, which can assist the SER task and cope with data sparsity problems. In the cross-attention module, AM-ResNet features and Wav2vec 2.0 features are interacted with and fused to obtain the cross-fused features, which are used to predict the final emotion. Furthermore, multi-label learning is also used to add ambiguous emotion utterances to deal with data limitations. Finally, experimental results illustrate the usefulness and superiority of our proposed framework over existing state-of-the-art approaches.

The deep learning-based physical education course recommendation system under the internet of things

This study aims to propose a deep learning (DL)-based physical education course recommendation system by combining the Internet of Things (IoT) technology and DL, to improve the accuracy and personalization of recommendation. Firstly, IoT devices such as smart bracelets and smart clothing are used to monitor students' physiological data in real-time, and IoT sensors are utilized to sense the environment around students. Secondly, IoT devices capture students' social interactions with their peers, recommending socially oriented courses. Meanwhile, by integrating IoT data with students' academic data, course recommendations are optimized to match students' learning progress and schedule. Finally, Generative Adversarial Network (GAN) models, especially the improved Regularization Penalty Conditional Feature Generative Adversarial Network (RP-CFGAN) model, deal with data sparsity and cold start problems. The experimental results show that this model performs well in TopN evaluation and is markedly enhanced compared with traditional models. This study denotes that integrating IoT technology and GAN models can more accurately understand student needs and provide personalized recommendations. Although the model performs well, there is still room for improvement, such as exploring more regularization techniques, protecting user privacy, and extending the system to diverse platforms and scenarios.

Quantum computing enhanced knowledge tracing: Personalized KT research for mitigating data sparsity

With the development of artificial intelligence in education, knowledge tracing (KT) has become a current research hotspot and is the key to the success of personalized instruction. However, data sparsity remains a significant challenge in the KT domain. To address this challenge, this paper applies quantum computing (QC) technology to KT for the first time. It proposes two personalized KT models incorporating quantum mechanics (QM): quantum convolutional enhanced knowledge tracing (QCE-KT) and quantum variational enhanced knowledge tracing (QVE-KT). Through quantum superposition and entanglement properties, QCE-KT and QVE-KT effectively alleviate the data sparsity problem in the KT domain through quantum convolutional layers and variational quantum circuits, respectively, and significantly improve the quality of the representation and prediction accuracy of students’ knowledge states. Experiments on three datasets show that our models outperform ten benchmark models. On the most sparse dataset, QCE-KT and QVE-KT improve their performance by 16.44% and 14.78%, respectively, compared to DKT. Although QC is still in the developmental stage, this study reveals the great potential of QM in personalized KT, which provides new perspectives for solving personalized instruction problems and opens up new directions for applying QC in education.

Contrastive Learning-Based Personalized Tag Recommendation.

Personalized tag recommendation algorithms generate personalized tag lists for users by learning the tagging preferences of users. Traditional personalized tag recommendation systems are limited by the problem of data sparsity, making the personalized tag recommendation models unable to accurately learn the embeddings of users, items, and tags. To address this issue, we propose a contrastive learning-based personalized tag recommendation algorithm, namely CLPTR. Specifically, CLPTR generates augmented views of user-tag and item-tag interaction graphs by injecting noises into implicit feature representations rather than dropping nodes and edges. Hence, CLPTR is able to greatly preserve the underlying semantics of the original user-tag or the item-tag interaction graphs and avoid destroying their structural information. In addition, we integrate the contrastive learning module into a graph neural network-based personalized tag recommendation model, which enables the model to extract self-supervised signals from user-tag and item-tag interaction graphs. We conduct extensive experiments on real-world datasets, and the experimental results demonstrate the state-of-the-art performance of our proposed CLPTR compared with traditional personalized tag recommendation models.

A privacy-preserving framework with multi-modal data for cross-domain recommendation

Cross-domain recommendation (CDR) aims to enhance the recommendation accuracy in a target domain with sparse data by leveraging rich information in a source domain, thereby addressing the data-sparsity problem. Some existing CDR methods highlight the advantages of extracting domain-common and domain-specific features to learn comprehensive user and item representations. However, these methods cannot effectively disentangle these components, as they often rely on simple user-item historical interaction information (such as ratings, clicks, and browsing), neglecting the rich multi-modal features. In addition, they do not protect user-sensitive data from potential leakage during knowledge transfer between domains. To address these challenges, we propose a Privacy-Preserving Framework with Multi-Modal Data for Cross-Domain Recommendation, called P2M2-CDR. Specifically, we first design a multi-modal disentangled encoder that utilizes multi-modal information to disentangle more informative domain-common and domain-specific embeddings. Furthermore, we introduce a privacy-preserving decoder to mitigate user privacy leakage during knowledge transfer. Local differential privacy (LDP) is used to obfuscate disentangled embeddings before the inter-domain exchange, thereby enhancing privacy protection. To ensure both consistency and differentiation among these obfuscated disentangled embeddings, we incorporate contrastive learning-based domain-inter and domain-intra losses. Extensive experiments conducted on six CDR tasks from two real-world datasets demonstrate that P2M2-CDR outperforms other state-of-the-art single- and cross-domain baselines. The code is available at https://github.com/Lili1013/P2M2-CDR.

Cross-biased contrastive learning for answer selection with dual-tower structure

A large number of unanswered products-related questions appear in the E-commerce platforms, necessitating the deployment of question-answering models to automatically provide precise responses for the user. However, the substantial absence of ground truth presents challenges for implementing supervised learning, and existing unsupervised contrastive learning methods have not addressed this issue fundamentally, with limitations in integrating multi-level features. This paper presents a dual-tower cross-biased contrastive learning model for answer selection, named CCLM-AS. By taking questions and augmented product reviews as positive pairs for unsupervised contrastive learning and employing a dual-tower structure encoder, CCLM-AS learns sentence representations from unlabeled data effectively. Furthermore, the designed training objective realizes multi-level feature integration, including character, sentence and dialogue levels, which enhances the model's inference ability. The experimental results on AmazonQA dataset indicate that our proposed model outperforms existing contrastive learning-based sentence representation models and also achieves comparable performance to supervised answer selection models. This demonstrates that CCLM-AS can not only alleviate the data sparsity problem effectively but also retain the excellent performance.

Multimodal Recommender Systems: A Survey

The recommender system (RS) has been an integral toolkit of online services. They are equipped with various deep learning techniques to model user preference based on identifier and attribute information. With the emergence of multimedia services, such as short videos, news and etc. , understanding these contents while recommending becomes critical. Besides, multimodal features are also helpful in alleviating the problem of data sparsity in RS. Thus, M ultimodal R ecommender S ystem (MRS) has attracted much attention from both academia and industry recently. In this paper, we will give a comprehensive survey of the MRS models, mainly from technical views. First, we conclude the general procedures and major challenges for MRS. Then, we introduce the existing MRS models according to four categories, i.e., Modality Encoder , Feature Interaction , Feature Enhancement and Model Optimization . Besides, to make it convenient for those who want to research this field, we also summarize the dataset and code resources. Finally, we discuss some promising future directions of MRS and conclude this paper. To access more details of the surveyed papers, such as implementation code, we open source a repository.

Utilizing deep transfer learning to discover changes in landscape patterns in urban wetland parks based on multispectral remote sensing

Urban wetland parks are essential for protecting ecosystems and alleviating urban heat island effects. Owing to the impact of urban sprawl and human activities, habitats in wetland parks have become increasingly fragmented, evoking an urgent need to accurately monitor and analyze such changes. In this study, a transfer learning-based ResNet-18 method was proposed to classify the landscape patterns of urban wetland parks by integrating the advantages of remote sensing technologies, i.e., long-time series of Gaofen-2 and high-accuracy data of unmanned aerial vehicle remote sensing. The proposed method solves the dual problems of low precision and sparse sample data in landscape pattern classification. By employing the proposed method, we realized long-time-series, high-accuracy analysis of landscape pattern changes in a national wetland park. Our results showed that the overall accuracy was 90.69–97.96 % and Kappa coefficient was stable between 0.865 and 0.968, fully verifying the effectiveness and reliability of our method. We revealed a shrinking trend in the area of water bodies along with an expanding trend in the area of other land types. Thus, our findings reflect the significant impact of urban sprawl on the landscape patterns of wetland parks.

Graphical contrastive learning for multi-interest sequential recommendation

Multi-interest sequential recommendation aims to deliver accurate recommendations by modeling users’ sequential behaviors and multi-faceted interests. Contrastive learning (CL) possesses an inborn advantage in distinguishing users’ multiple interests. However, existing CL-based multi-interest sequential recommenders tend to overlook the issue of data sparsity which deteriorates model performance. Graph neural network (GNN), on the other hand, has been proven as an effective measure of tackling the data sparsity problem. However, incorporating GNN into CL requires a delicate design to simultaneously achieve distinctive interest representation learning and relieve data sparsity. To address the challenges, we propose a Graphical Contrastive Learning for Multi-Interest Sequential Recommendation (GCL4MI), which first extracts user interest and constructs interest-specific graph based on item clusters and then perform graphical contrastive learning to enhance the independence between multiple interests. Furthermore, we explore different extensible CL sampling strategies with the facilitation of GNN to learn distinct interests in a high-cohesion and low-coupling manner. Extensive experiments on three real-world datasets demonstrate that our proposed GCL4MI significantly outperforms state-of-the-arts, with average lifts of 7.0%, 7.1%, and 5.8% on recall, NDCG, and HR, respectively.

The Right Triangle Model: Overcoming the Sparse Data Problem in Thermal/Optical Remote Sensing of Soil Moisture

The simplicity of the so-called triangle method allows estimates of evapotranspiration and soil water content to be made without ancillary data external to the image and with just a few simple algebraic calculations. Drawing on many examples in the literature showing that the pixel distribution in temperature/fractional vegetation cover (NDVI) space closely resembles a right triangle, this paper shows that adoption of a right triangle shape further simplifies the triangle model. Moreover, it allows one to mostly avoid the problem of sparse or low-resolution data. A time dimension can be included showing that trajectories inside the triangle can provide additional information on root zone soil water content. After discussing some of the ambiguities in the triangle method, and the advantageous properties of the right triangle, a proposal is made to illuminate the relationship between thermal/optical measurements and root zone water content within the right triangle framework.

Spatiotemporal context transition model based on graph convolutional network and its implementation

PurposeThe purpose of this paper is to present a method that addresses the data sparsity problem in points of interest (POI) recommendation by introducing spatiotemporal context features based on location-based social network (LBSN) data. The objective is to improve the accuracy and effectiveness of POI recommendations by considering both spatial and temporal aspects.Design/methodology/approachTo achieve this, the paper introduces a model that integrates the spatiotemporal context of POI records and spatiotemporal transition learning. The model uses graph convolutional embedding to embed spatiotemporal context information into feature vectors. Additionally, a recurrent neural network is used to represent the transitions of spatiotemporal context, effectively capturing the user’s spatiotemporal context and its changing trends. The proposed method combines long-term user preferences modeling with spatiotemporal context modeling to achieve POI recommendations based on a joint representation and transition of spatiotemporal context.FindingsExperimental results demonstrate that the proposed method outperforms existing methods. By incorporating spatiotemporal context features, the approach addresses the issue of incomplete modeling of spatiotemporal context features in POI recommendations. This leads to improved recommendation accuracy and alleviation of the data sparsity problem.Practical implicationsThe research has practical implications for enhancing the recommendation systems used in various location-based applications. By incorporating spatiotemporal context, the proposed method can provide more relevant and personalized recommendations, improving the user experience and satisfaction.Originality/valueThe paper’s contribution lies in the incorporation of spatiotemporal context features into POI records, considering the joint representation and transition of spatiotemporal context. This novel approach fills the gap left by existing methods that typically separate spatial and temporal modeling. The research provides valuable insights into improving the effectiveness of POI recommendation systems by leveraging spatiotemporal information.

Efficient and adaptive secure cross-domain recommendations

In recommendation systems, cross-domain recommendation (CDR) has emerged as an effective method for solving the problem of data sparsity and cold start. However, challenges such as data heterogeneity, data security, and recommendation accuracy are faced by existing CDR methods. An efficient and adaptive secure cross-domain recommendation (EAAS-CDR) model is proposed. First, the Automatic Encoder–Decoder (AutoED) model is proposed to address variations in user–item interaction quantity and item characteristics across various domains. The AutoED model enables adaptive and secure cross-domain recommendations by modeling user preferences across various domains, unifying vector dimensions, and avoiding the release of the rating matrix. An efficient cross-domain privacy protection method is proposed to resist data leakage. To ensure feature invariance for knowledge transfer, the inherent low perturbation characteristics of the Johnson–Lindenstrauss Transform mechanism are leveraged and combined with the output of our encoder. Finally, domain relevance measurement metrics and a domain fusion algorithm are proposed to use the source domain knowledge in the target domain and improve the recommendation accuracy of the model. Experimental results demonstrate that our approach is not only more efficient but also more secure than existing CDR approaches.

Multichannel Adaptive Data Mixture Augmentation for Graph Neural Networks

Graph neural networks (GNNs) have demonstrated significant potential in analyzing complex graph-structured data. However, conventional GNNs encounter challenges in effectively incorporating global and local features. Therefore, this paper introduces a novel approach for GNN called multichannel adaptive data mixture augmentation (MAME-GNN). It enhances a GNN by adopting a multi-channel architecture and interactive learning to effectively capture and coordinate the interrelationships between local and global graph structures. Additionally, this paper introduces the polynomial–Gaussian mixture graph interpolation method to address the problem of single and sparse graph data, which generates diverse and nonlinear transformed samples, improving the model's generalization ability. The proposed MAME-GNN is validated through extensive experiments on publicly available datasets, showcasing its effectiveness. Compared to existing GNN models, the MAME-GNN exhibits superior performance, significantly enhancing the model's robustness and generalization ability.

TB-BGAT With TinyBERT and BiGRU in Personalized Course Recommendations

Aiming at the problems of inadequate user feature extraction, cold start and sparse data, a personalized course recommendation algorithm that utilizes TB-BGAT is suggested. First, the Tiny Bidirectional Encoder Representation from Transformers (TinyBERT) model is utilized to output character-level word vectors; then, Bidirectional Recurrent Neural Network (BiGRU) model is utilized to obtain the embedded contextual semantic features. Finally, the attention mechanism is utilized to allocate weights to various course features by assigning their importance and to obtain the output results. The results of experiment on the publicly available dataset MOOCs-Course prove that the proposed method improves at least 3.62%, 3.04%, and 3.33% in precision, recall, and F1-score, correspondingly, in contrast to several other state-of-the-art course resource recommendation algorithms. The proposed method can enhance the effectiveness of the course recommendation model, enhance the quality of learners' online learning, and provide good technical support for online education learning platforms.

KGCF: Social relationship-aware graph collaborative filtering for recommendation

With the increasing popularity of recommendation techniques and social networks, social network recommendation has become a significant research field, i.e., predicting a user's preferences based on her or his historical interaction data, because the social relationships of users can not only enrich their interaction information, but also imply the accurate characterization of their preferences. Although there are some existing studies that predict users' preferences by using the characteristics of social relationships, their time complexity and hardware resource consumption are often very expensive, which limits the feasibility in large-scale real-world scenarios. To address these issues, in this paper, we propose a simple and effective K-core Graph Collaborative Filtering (KGCF) model to incorporate the user's social features into a recommendation framework. The user's interaction information is auto-encoded linearly by incorporating the social relationship graph into the user's interaction information and constructing a multi-layer graph filter. The linear autoencoder graph filter alleviates the highly sparse data problem and dramatically reduces the training time and hardware resource consumption. Experimental results on several real-world datasets show the superior effectiveness and spatiotemporal efficiency compared with the existing baselines, especially the training speed is significantly accelerated for large-scale datasets.

Research on Hybrid Recommendation Algorithm based on Collaborative Filtering and Spearman Rank Correlation Coefficient

A recommendation system is an information filtering tool that helps users find the products or services they need from a large amount of information. However, collaborative filtering is quite sensitive to data sparsity and cold start problems, and it may encounter certain difficulties when handling outliers. To address the issue of handling outliers, it is necessary to study and improve the existing collaborative filtering techniques. This paper proposes a personalized recommendation algorithm that integrates collaborative filtering with the Spearman rank correlation coefficient. By combining collaborative filtering and the Spearman rank correlation coefficient, the algorithm uses the latter to handle outliers, making it more suitable for nonlinear relationships. This hybrid recommendation algorithm can better handle outliers while maintaining personalized recommendations, providing a basis and reference for personalized recommendations.