In this study, a novel top-K ranking recommendation method called collaborative social metric learning (CSML) is proposed, which implements a trust network that provides both user-item and user-user... Show More

In this study, a novel top-K ranking recommendation method called collaborative social metric learning (CSML) is proposed, which implements a trust network that provides both user-item and user-user interactions in simple structure. Most existing recommender systems adopting trust networks focus on item ratings, but this does not always guarantee optimal top-K ranking prediction. Conventional direct ranking systems in trust networks are based on sub-optimal correlation approaches that do not consider item-item relations. The proposed CSML algorithm utilizes the metric learning method to directly predict the top-K items in a trust network. A new triplet loss is further proposed, called socio-centric loss, which represents user-user interactions to fully exploit the information contained in a trust network, as an addition to the two commonly used triplet losses in metric learning for recommender systems, which consider user-item and item-item relations. Experimental results demonstrate that the proposed CSML outperformed existing recommender systems for real-world trust network data.

The photo response non-uniformity (PRNU) is used to connect an image to its source sensor. In this paper, researchers propose a PRNU anonymity method based on image segmentation to cut the... Show More

The photo response non-uniformity (PRNU) is used to connect an image to its source sensor. In this paper, researchers propose a PRNU anonymity method based on image segmentation to cut the relationship between the image and its source camera. According to the distribution rule of PRNU in the high and low frequency band of the image, the high and low frequency information of the part is also processed differently, which ensures the quality of the output image to a large extent. Experiments on the datasets show that the proposed method can preserve the biometric characteristics of the device while maintaining the anonymity of the device. Comparing with prior art, peak signal to noise ratio (PSNR) and cosine similarity are improved by 1.9 dB and 0.02 points, respectively.

Entity alignment aims to identify equivalent entity pairs from different knowledge graphs (KGs). Recently, aligning temporal knowledge graphs (TKGs) that contain time information has aroused... Show More

Entity alignment aims to identify equivalent entity pairs from different knowledge graphs (KGs). Recently, aligning temporal knowledge graphs (TKGs) that contain time information has aroused increasingly more interest, as the time dimension is widely used in real-life applications. The matching between TKGs requires seed entity pairs, which are lacking in practice. Hence, it is of great significance to study TKG alignment under scarce supervision. In this work, the authors formally formulate the problem of TKG alignment with limited labeled data and propose to solve it under the active learning framework. As the core of active learning is to devise query strategies to select the most informative instances to label, the authors propose to make full use of time information and put forward novel time-aware strategies to meet the requirement of weakly supervised temporal entity alignment. Extensive experimental results on multiple real-world datasets show that it is important to study TKG alignment with scarce supervision, and the proposed time-aware strategy is effective.

New models of entrepreneurship are emerging because of increasing digitalization and the development of artificial intelligence (AI). There is a lack of existing research on the intersection between... Show More

New models of entrepreneurship are emerging because of increasing digitalization and the development of artificial intelligence (AI). There is a lack of existing research on the intersection between digitalization and entrepreneurship. Therefore, this systematic literature analysis aims to expand knowledge in this area and provide a semantic analysis of existing contributions. Following the SPAR-4-SLR protocol, it analyzes 520 scientific articles from the Dimensions.ai database up to July 2022. The methodology uses natural language processing (NLP) and tools such as bibliometrix and VosViewer, which reveal the main characteristics of the titles and texts of the abstracts and their links with the numbers of citations and with scientific impact. This study provides guidelines and clear recommendations for scientists to focus their scientific research on AI and entrepreneurship and entrepreneurs by including the link between AI and entrepreneurship in their strategies. As future lines of research, the authors highlight the potential of using NLP in bibliometric analysis.

Cloud gaming (CG) has gradually gained popularity. By leveling shared computing resources on the cloud, CG technology allows those without expensive hardware to enjoy AAA games using a low-end... Show More

Cloud gaming (CG) has gradually gained popularity. By leveling shared computing resources on the cloud, CG technology allows those without expensive hardware to enjoy AAA games using a low-end device. However, the bandwidth requirement for streaming game video is high, which can cause backbone network congestion for large-scale deployment and expensive bandwidth bills. To address this challenge, the authors proposed an innovative edge-assisted computing architecture that collaboratively uses AI-powered foveated rendering (FR) and super-resolution (SR). Using FR, the cloud server can stream gaming video in lower resolution, significantly reducing the transmitted data volume. The edge server will then upscale the video using a game-specific SR model, recovering the quality of the video, especially for the areas players pay the most attention. The authors built a prototype system called FRSR and did thorough, objective comparative experiments to demonstrate that this architecture can reduce bandwidth usage by 39.47% compared with classic CG implementation for similar perceived quality.

Many existing works compute the binary vulnerability similarity based on binary procedure, which has coarse detection granularity and cannot locate the vulnerability trigger position accurately, and... Show More

Many existing works compute the binary vulnerability similarity based on binary procedure, which has coarse detection granularity and cannot locate the vulnerability trigger position accurately, and have a higher false positive rate, so a new binary vulnerability similarity detection method based on function parameter dependency in hazard API is proposed. First, convert the instructions of different architectures into an intermediate language, and use the compiler with a back-end optimizer to optimize and normalize the binary procedure. Then, locate the hazard API that appears in the binary procedure, and perform the function parameters dependency analysis to generate a set of parameter slices on the hazard API. Experiments show that the method has a higher recall rate (up to 14.3% better than the baseline model) in real-world scenarios, and not only locates the triggering position of the vulnerability but also identifies the fixed vulnerability.

As one of the most widely used federated chains, hyperledger fabric uses many cryptographic algorithms to ensure the security of information on the chain, but the ECDSA cryptographic algorithm used... Show More

As one of the most widely used federated chains, hyperledger fabric uses many cryptographic algorithms to ensure the security of information on the chain, but the ECDSA cryptographic algorithm used in the fabric system has backdoor security risks. In this paper, the authors adopt SM2 algorithm to replace the corresponding ECDSA algorithm for blockchain design based on fabric platform. Firstly, they optimize the part of SM2 signature algorithm process with inverse operation and effectively reduce the time complexity by reducing the inverse operation in the whole process, and the experimental results show that the improved SM2 algorithm improves the signature and verification efficiency by about 5.7%. Secondly, by adding SM2 algorithm template and interface to the BCCSP module of fabric platform to realize the shift value of SM2 algorithm and compare the performance with the native fabric system, the network startup time is reduced by about 29%. The experimental results show the effectiveness of the improved SM2 algorithm, and also the performance of the optimized fabric system is improved.

With the emergence of a large-scale pre-training model based on the transformer model, the effect of all-natural language processing tasks has been pushed to a new level. However, due to the high... Show More

With the emergence of a large-scale pre-training model based on the transformer model, the effect of all-natural language processing tasks has been pushed to a new level. However, due to the high complexity of the transformer's self-attention mechanism, these models have poor processing ability for long text. Aiming at solving this problem, a long text processing method named HBert based on Bert and hierarchical attention neural network is proposed. Firstly, the long text is divided into multiple sentences whose vectors are obtained through the word encoder composed of Bert and the word attention layer. And the article vector is obtained through the sentence encoder that is composed of transformer and sentence attention. Then the article vector is used to complete the subsequent tasks. The experimental results show that the proposed HBert method achieves good results in text classification and QA tasks. The F1 value is 95.7% in longer text classification tasks and 75.2% in QA tasks, which are better than the state-of-the-art model longformer.

Discover the deep semantics from the massively structured data in knowledge graph and provide reasonable explanations are a series of important foundational research issues of artificial... Show More

Discover the deep semantics from the massively structured data in knowledge graph and provide reasonable explanations are a series of important foundational research issues of artificial intelligence. However, the deep semantics hidden between entities in knowledge graph cannot be well expressed. Moreover, considering many predicates express fuzzy relationships, the existing reasoning methods cannot effectively deal with these fuzzy semantics and interpret the corresponding reasoning process. To counter the above problems, in this article, a new interpretable reasoning schema is proposed by introducing fuzzy theory. The presented method focuses on analyzing the fuzzy semantic between related entities in a knowledge graph. By annotating the fuzzy semantic features of adjacency predicates, a novel semantic reasoning model is designed to realize the fuzzy semantic extension over knowledge graph. The evaluation, based on both visualization and query experiments, shows that this proposal has advantages over the initial knowledge graph and can discover more valid semantic information.

A citation is a reference to the source of information used in an article. Citations are very useful for students and researchers to locate relevant information on a topic. Proper citation is also... Show More

A citation is a reference to the source of information used in an article. Citations are very useful for students and researchers to locate relevant information on a topic. Proper citation is also important in the academic ethics of article writing. Due to the rapid growth of scientific works published each year, how to automatically recommend citations to students and researchers has become an interesting but challenging research problem. In particular, a citation recommendation system can assist students to identify relevant papers and literature for academic writing. Citation recommendation can be classified into local and global citation recommendation depending on whether a specific local citation context is given; e.g., the text surrounding a citation placeholder. This article provides a systematic review on global citation recommendation models and compares the reviewed methods from the traditional topic- based models to the recent models embedded with deep neural networks, aiming to summarize this field to facilitate researchers working on citation recommendation.

Machine learning and deep learning are one of the most sought-after areas in computer science which are finding tremendous applications ranging from elementary education to genetic and space... Show More

Machine learning and deep learning are one of the most sought-after areas in computer science which are finding tremendous applications ranging from elementary education to genetic and space engineering. The applications of machine learning techniques for the development of smart cities have already been started; however, still in their infancy stage. A major challenge for Smart City developments is effective waste management by following proper planning and implementation for linking different regions such as residential buildings, hotels, industrial and commercial establishments, the transport sector, healthcare institutes, tourism spots, public places, and several others. Smart City experts perform an important role for evaluation and formulation of an efficient waste management scheme which can be easily integrated with the overall development plan for the complete city. In this work, we have offered an automated classification model for urban waste into multiple categories using Convolutional Neural Networks. We have represented the model which is being implemented using Fine Tuning of Pretrained Neural Network Model with new datasets for litter classification. With the help of this model, software, and hardware both can be developed using low-cost resources and can be deployed at a large scale as it is the issue associated with healthy living provisions across cities. The main significant aspects for the development of such models are to use pre-trained models and to utilize transfer learning for fine-tuning a pre-trained model for a specific task.

Traditional landscape design methods rely entirely on the experience of designers and are difficult to adapt to the needs of modern society. This article proposes a landscape design method based on... Show More

Traditional landscape design methods rely entirely on the experience of designers and are difficult to adapt to the needs of modern society. This article proposes a landscape design method based on a distributed integrated model. Based on landscape design scheme data, the intelligent landscape design function is achieved by constructing a distributed geographic model, extracting features through data analysis and key point analysis, and using virtual environments in computer-aided design to display and restore the actual effects of landscape design. The results indicate that the landscape design method based on distributed integration mode is more in line with the needs of modern society and has significant advantages over traditional landscape design in terms of public interest and evaluation coefficient. The intelligent landscape design method based on distributed integrated models has important significance in modern urbanization construction, which can effectively improve the accuracy and speed of landscape design and create better living spaces for people.

Emotion-cause pair extraction is an emergent natural language processing task; the target is to extract all pairs of emotion clauses and corresponding cause clauses from unannotated emotion text.... Show More

Emotion-cause pair extraction is an emergent natural language processing task; the target is to extract all pairs of emotion clauses and corresponding cause clauses from unannotated emotion text. Previous studies have employed two-step approaches. However, this research may lead to error propagation across stages. In addition, previous studies did not correctly handle the situation where emotion clauses and cause clauses are the same clauses. To overcome these issues, the authors first use a multitask learning model that is based on graph from the perspective of sorting, which can simultaneously extract emotion clauses, cause clauses and emotion-cause pairs via an end-to-end strategy. Then the authors propose to convert text into graph structured data, and process this scenario through a unique graph convolutional neural network. Finally, the authors design a semantic decision mechanism to address the scenario in which there are multiple emotion-cause pairs in a text.

With the rapid development and popularization of intelligent terminals, app software has also developed rapidly. The research and practical value of mining user experience (UX) of app software form... Show More

With the rapid development and popularization of intelligent terminals, app software has also developed rapidly. The research and practical value of mining user experience (UX) of app software form interaction information are becoming increasingly prominent. The interactive information of app software is multisource homogeneous and heterogeneous. In order to obtain more accurate and more comprehensive app software UX results, the fused multisource information should be analyzed. In this paper, the app software UX analysis method based on multisource information fusion is proposed. First, feature engineering is carried out to extract the features. Then, the feature combination tree is constructed after feature correlation mining. Finally, the multisource app software interactive data are fused, and the result is further analyzed to obtain the information of app software UX. The experiments clearly show that the method can effectively fuse multisource app software interaction data and help to comprehensively mine the app software UX embodied in the data.

After years of development, Hainan Province has established a significant sports tourism industry. This paper explores how digitalization is driving the transformation of traditional sports tourism... Show More

After years of development, Hainan Province has established a significant sports tourism industry. This paper explores how digitalization is driving the transformation of traditional sports tourism through the utilization of the DRAT model to reconstruct the value of both consumers and businesses. The analysis shows that the digitalization of sports tourism has deconstructed the traditional industry using innovative technologies such as technological virtualization and data platform construction. The effective development and utilization of highly developed information technology in modern society are essential for ensuring the smooth and healthy growth of the sports tourism industry. The traditional sports tourism industry has been digitally deconstructed, leading to the formation of new production, management, and business models. The status of consumers continues to improve during this process, and companies are seeking to collaborate with consumers to create new value positioning.

The emergence of the Internet and the growing development of online platforms (like Facebook and Instagram) opened the way for disseminating information that hasn't been experienced in the history... Show More

The emergence of the Internet and the growing development of online platforms (like Facebook and Instagram) opened the way for disseminating information that hasn't been experienced in the history of mankind earlier. Consumers generate and share more information and a massive amount of data than ever with the growing utilization of social media sites, many of which are deceptive with little relevance to reality. A daunting task is the automated classification of a text article as misleading or misinformation. To see the latest news alerts, individuals often utilize e-newspapers, Twitter, Instagram, Youtube, and many more. Fake news created on social media can lead to uncertainty amongst individuals and psychiatric illness. We may detect that news obtained based on machine learning techniques is either true or false. This study proposes a machine learning technique to detect fake news by carrying out filtration on social media data, classifying the preprocessed data using a machine learning algorithm, evaluating the developed system, and evaluating the results.

The danger of distributed denial of service (DDoS) attacks has grown in tandem with the proliferation of intelligent information systems. Because of the sheer volume of connected devices, constantly... Show More

The danger of distributed denial of service (DDoS) attacks has grown in tandem with the proliferation of intelligent information systems. Because of the sheer volume of connected devices, constantly shifting network circumstances, and the need for instantaneous reaction, conventional DDoS detection methods are inadequate for the IoT. In this context, this study aims to survey the current state of the art in the topic by reading relevant articles found in the Scopus database, with a brief overview of the IoT and DDoS as this study examines neural networks and their applicability to DDoS detection. Finally, a decision tree-based model is developed for the detection of DDoS attacks. The analysis sheds light on the present trends and issues in this field and suggests avenues for further study.

There are problems of knowledge deficiency and effective unified expression of knowledge in the process of relevant knowledge data acquired by workers in the ceramic domain. In this study, the... Show More

There are problems of knowledge deficiency and effective unified expression of knowledge in the process of relevant knowledge data acquired by workers in the ceramic domain. In this study, the authors designed relevant experiments to construct ceramic field knowledge graphs to solve these problems. In the experiments of named entity recognition and relationship recognition, the authors compared the performance of several models in OwnThink and ceramics field datasets. The experimental results showed that the BiLSTM-CRF model is the best for named entity recognition and the TextCNN model is the best for relationship recognition in ceramics field datasets. Therefore, the first used the BiLSTM-CRF model to complete the naming entity recognition and then combined with the TextCNN model to complete the relationship recognition to construct the ceramic field knowledge graph. Then, they applied the constructed graph to the ceramic knowledge Q&A service to provide accurate data retrieval service for ceramic domain workers.

In recent years, knowledge-aware recommendation systems have gained popularity as a solution to address the challenges of data sparsity and cold start in collaborative filtering. However... Show More

In recent years, knowledge-aware recommendation systems have gained popularity as a solution to address the challenges of data sparsity and cold start in collaborative filtering. However, traditional knowledge graph convolutional networks impose significant computational burdens during training, demanding substantial resources and increasing the cost of recommendations. To address this issue, this article proposes a lightweight knowledge graph convolutional network for collaborative filtering (LKGCF). LKGCF eliminates the feature transformation and nonlinear activation components, by focusing on essential elements such as neighborhood aggregation and layer combination. LKGCF captures the user's long-distance personalized interests on the knowledge graph by sampling from neighborhood information and constructing a weighted sum of item embeddings. Experimental results demonstrate that the proposed model is easy to train and implement due to its coherence and simplicity. Furthermore, notable improvements in recommendation performance are observed compared to strong baselines.

Effective extraction of patent technology points in new energy fields is profitable, which motivates technological innovation and facilitates patent transformation and application. However, since... Show More

Effective extraction of patent technology points in new energy fields is profitable, which motivates technological innovation and facilitates patent transformation and application. However, since patent data exists the ununiform distribution of technology points information, long length of term, and long sentences, technology point extraction faces the dilemmas of poor readability and logic confusion. To mitigate these problems, the article proposes a method to generate patent technology points called IGPTP—a two-stage strategy, which fuses the advantage of extractive and generative ways. IGPTP utilizes the RoBERTa+CNN model to obtain the key sentences of text and takes the output as input of UNILM (unified pre-trained language model). Simultaneously, it takes a multi-strategies integration technique to enhance the quality of patent technology points by combining the copy mechanism and external knowledge guidance model. Substantial experimental results manifest that IGPTP outperforms the current mainstream models, which can generate more coherent and richer text.

Knowledge graphs are a valuable tool for intelligent tutoring systems and are typically constructed with a focus on objectivity and accuracy. However, they may not effectively capture the... Show More

Knowledge graphs are a valuable tool for intelligent tutoring systems and are typically constructed with a focus on objectivity and accuracy. However, they may not effectively capture the subjectivity and complex relationships often present in the humanities. To address this issue, a dynamic visualization of subject matter knowledge graph was developed using a collective intelligence approach that integrates the individual intelligence of learners and considers cognitive diversity to construct and evolve the knowledge graph. The approach resulted in the construction of 722 knowledge associations and the evolution of 584 triples. A survey assessed the effectiveness and user-friendliness, revealing that this approach is effective, easy to use, and can improve subject matter knowledge ontology. In conclusion, combining individual and collective intelligence is a promising approach for building effective knowledge graphs in subject areas with subjectivity and complexity.

Relying on features such as high-speed, low latency, support for cutting-edge technology, internet of things, and multimodality, 5G networks will greatly contribute to the transformation of Web 3.0.... Show More

Relying on features such as high-speed, low latency, support for cutting-edge technology, internet of things, and multimodality, 5G networks will greatly contribute to the transformation of Web 3.0. In order to realize low-latency and high-speed information exchange in 5G communication networks, a method based on the allocation of network computing resource in view of edge computing model is proposed. The method first considers three computing modes: local device computing, local mobile edge computing (MEC) server computing, and adjacent MEC server computing. Then, a multi-scenario edge computing model is further constructed for optimizing energy consumption and delay. At the same time, the encoding-decoding mode is used to optimize PSO algorithm and combined with the improvement of fitness function, which can effectively support the communication network to achieve reasonable allocation of resources, ensuring efficiency of information exchange in the network. In the end, the results show that when the number of users is 500, the method can complete the task assignment within 44s.

This paper explores the potential of utilizing semantic web technologies to improve class management in Chinese schools. By analyzing a comprehensive dataset obtained from the Scopus database, the... Show More

This paper explores the potential of utilizing semantic web technologies to improve class management in Chinese schools. By analyzing a comprehensive dataset obtained from the Scopus database, the study investigates publication trends, document types, keyword distributions, and author contributions in the field of semantic web technologies for class management. The findings reveal a growing interest in this research area and highlight the benefits of semantic web technologies in personalized learning, information retrieval, collaboration, and assessment. The paper discusses the practical implications, challenges, and considerations for implementing semantic web technologies in Chinese schools. It aims to provide valuable insights for educators, researchers, policymakers, and educational technology practitioners interested in enhancing class management practices through the innovative use of semantic web technologies.

With the rapid development of computer and information technology, people can obtain and store data in a faster and cheaper way, which makes the amount of data and information grow exponentially.... Show More

With the rapid development of computer and information technology, people can obtain and store data in a faster and cheaper way, which makes the amount of data and information grow exponentially. Based on data mining technology, this paper systematically analyzes and compares the basic situation and evolution of international trade networks and international human relations networks using the bilateral trade data and human relations data of countries and regions in the world. The research shows that the strength external entropy, strength internal entropy, and network weight entropy of the whole international trade network from 2008 to 2014 are generally lower than 0.75. The whole network still showed a downward trend from 2010 to 2015, but it began to rise steadily after 2016. The monopolistic behavior in international technology trade has a significant impact on the fundamental freedoms and rights of citizens.

The topic about consensus target track seeking for high-order nonlinear multi-agent systems (MASs) with unmodeled dynamics, dynamic disturbances, and dead-zone input is considered in the paper.... Show More

The topic about consensus target track seeking for high-order nonlinear multi-agent systems (MASs) with unmodeled dynamics, dynamic disturbances, and dead-zone input is considered in the paper. Using the strong nonlinear map characteristic of radial basis function neural networks (RBFNNs), the complex functions arising from recursive procedure are simplified. Also, inspired by input-to-state practical stability (ISpS), the authors construct a dynamical signal in order to counteract the impact of unmodeled dynamics and dynamic disturbances. The bounded inequality expression has been applied to tackle the unknown input of dead zone. Based on this, consensus control protocol suitable for nonlinear constraints has been constructed by using the recursive backstepping technique and adaptive neural network method. Theoretical analysis indicates not only the uniform boundary of all signals in the closed-loop under the neuro-based consensus controller, but uniform ultimate convergence of consensus tracking errors. The final simulations also confirmed the correctness of the theoretical analysis.

This paper delves into the dynamic intersection of athletic psychology and emerging technologies, aiming to understand their interplay and implications for sports performance. The study examines the... Show More

This paper delves into the dynamic intersection of athletic psychology and emerging technologies, aiming to understand their interplay and implications for sports performance. The study examines the latest research and literature in this field, encompassing the use of social media, digital devices, and virtual reality as technological advancements. It explores the impact of these technologies on athlete psychology, mental resilience, motivation, and goal setting. By analyzing country-specific scientific production, author contributions, and keyword trends, the paper provides insights into the global landscape of research in athletic psychology and emerging technologies. The findings contribute to a better understanding of the evolving relationship between technology and athlete psychology, offering potential avenues for optimizing performance, mental well-being, and training strategies in the realm of sports.

In order to enhance the utility of online educational digital resources, the authors propose a practical and efficient multi-strategy relation extraction (RE) model in online education scenarios.... Show More

In order to enhance the utility of online educational digital resources, the authors propose a practical and efficient multi-strategy relation extraction (RE) model in online education scenarios. First, the effective relation discrimination model is used to make relation predictions for non-structured teaching resources and eliminate the noise data. Then, they extract relations from different path strategies using multiple low-computational resources and efficient relation extraction strategies and use their proposed multi-strategy weighting calculator to weigh the relation extraction strategies to derive the final target relations. To cope with the low-resource relation extraction scenario, the relation extraction results are complemented by using prompt learning with a big model paradigm. They also consider the model to serve the commercial scenario of online education, and they propose a global rate controller to adjust and adapt the rate and throughput requirements in different scenarios, so as to achieve the best balance of system stability, computation speed, and extraction performance.

In the domain of target detection in mobile and embedded devices, neural network model inference speed is a crucial metric. This paper introduces YOLO-FLNet, a lightweight algorithm for detecting... Show More

In the domain of target detection in mobile and embedded devices, neural network model inference speed is a crucial metric. This paper introduces YOLO-FLNet, a lightweight algorithm for detecting people in open scenes. The model utilizes the DFEM structure to capture and process high-frequency and low-frequency information in the feature map. Additionally, the VoV-DFEM structure, based on the concept of one-shot aggregation, enhances feature aggregation from different scales and frequencies in the backbone network. To validate its performance, experiments were conducted using publicly available datasets on a computer with dedicated GPUs. As a result, compared to YOLOv7-tiny, YOLO-FLNet achieved a 0.3% mAP@0.5 improvement, reduced parameter size by 52.9%, and increased inference speed by 30.2%. These characteristics make it valuable for person detection in engineering domains, providing theoretical guidance for lightweight models in edge computing.

Unmanned motion platforms are being used in a wide range of industries. Unmanned motion platforms must have an autonomous and intelligent navigation procedure in order to carry out their system... Show More

Unmanned motion platforms are being used in a wide range of industries. Unmanned motion platforms must have an autonomous and intelligent navigation procedure in order to carry out their system functions. Traditional inertial navigation and radio navigation have poor accuracy and autonomy when not dependent on satellite circumstances. The accuracy of image recognition algorithms must meet strict standards. This study and exploration of the high-precision scene image recognition system is based on convolutional neural network structure optimization. To demonstrate the viability of the approach, simulation experiments are carried out on the NUC dataset using the recognition technique based on a convolutional neural network that is proposed. The fundamental network architecture of a convolutional neural network is optimized using the L2 regularization technique. The experimental findings demonstrate that the NUC dataset now has better recognition accuracy. In terms of recognition accuracy, the suggested method satisfies the predetermined requirements.

In location-based social networks (LBSN), users can check-in at points of interest (POI) to record their trips. POI recommendation is an important service provided by LBSN; it can help users quickly... Show More

In location-based social networks (LBSN), users can check-in at points of interest (POI) to record their trips. POI recommendation is an important service provided by LBSN; it can help users quickly find POI of interest, and also help POI providers more comprehensively understand user preferences and improve service quality. This paper proposes a POI recommendation algorithm that is based on attention mechanism. The sequence characteristics and short-term preferences of historical data are captured through the attention mechanism module and long short-term memory network (LSTM), and the POI location prediction is carried out in combination with the user embedding characteristics, and a better prediction accuracy is obtained. These results simulated show that the proposed method can realize the reliable analysis of complex data sets, and its precision index remains above 0.1 and recall index remains above 0.08, and it can also alleviate the cold start problem and better meet the personalized needs of users.

The facility allocation of the supply chain is critical since it directly influences cost efficiency, customer service, supply chain responsiveness, risk reduction, network optimization, and overall... Show More

The facility allocation of the supply chain is critical since it directly influences cost efficiency, customer service, supply chain responsiveness, risk reduction, network optimization, and overall competitiveness. When enterprises deploy their facilities wisely, they may achieve operational excellence, exceed customer expectations, and obtain a competitive advantage in today's volatile business climate. Due to this reason, a multi-objective facility allocation problem is introduced in this research with cooperative-based multi-level backup coverage considering distance-based facility attractiveness. The facility of the coverage is further described as two different layers of the coverage process, where demand can be covered as full, partial, and no coverage by their respective facilities. The main objectives of this facility allocation problem are to maximize the coverage of the facility to maximize overall facility coverage in the supply chain network and simultaneously minimize the overall cost.

Semantic web-based video surveillance systems can provide strong decision-making support for managers, and they have high requirements for real-time and precision of vehicle detection models in... Show More

Semantic web-based video surveillance systems can provide strong decision-making support for managers, and they have high requirements for real-time and precision of vehicle detection models in complex night scenes. To address this issue, a lightweight nighttime vehicle detection method (MC-YOLO) integrating MobileNetV2 and YOLOV3 is proposed. Firstly, in the preprocessing stage, image enhancement is performed on nighttime images to facilitate model feature extraction. Then, the lightweight network MobileNetV2 is used to extract feature by replacing the backbone network DarkNet53 in YOLOv3, thus accelerating the speed of target detection. Finally, after the convolution operation of the backbone network, a convolution block attention module is added to enhance the important feature information and suppress the secondary features, thereby improving the detection precision. The experimental results on the BDD100K dataset show that the proposed MC-YOLO model has a precision of up to 92.75%, which is superior to several other advanced comparative models.

In order to address the drawbacks of semantic ambiguity, inaccurate quantifiers, and low translation accuracy in traditional grammar-based translation methods, this paper proposes an artificial... Show More

In order to address the drawbacks of semantic ambiguity, inaccurate quantifiers, and low translation accuracy in traditional grammar-based translation methods, this paper proposes an artificial intelligence translation method based on semantic analysis for English fuzzy semantics. Firstly, a comprehensive analysis of English language semantics was carried out from different semantic levels such as language, knowledge, and pragmatics, and the key points of fuzzy semantics were identified. Then, key feature quantities for accurate translation of fuzzy semantics in English vocabulary and literature were constructed, and artificial intelligence methods were used to optimize fuzzy semantics. The experimental results show that the proposed method can avoid semantic understanding ambiguity and improve the accuracy of English language translation.

In recent studies, graph convolutional neural networks (GCNs) have been used to solve different natural language processing (NLP) tasks. However, few researches apply graph convolutional networks to... Show More

In recent studies, graph convolutional neural networks (GCNs) have been used to solve different natural language processing (NLP) tasks. However, few researches apply graph convolutional networks to short text classification. Emoji prediction, as a complex sentiment analysis task, has received even less attention. In this work, the authors propose TGCN-Bert which combines pre-trained BERT temporal convolutional networks (TCNs) and graph convolutional networks for short text classification and emoji prediction. They initialize the nodes with the help of BERT and define the edges in text graph based on the term frequency-inverse document frequency (TF-IDF) and positive point-wise mutual information (PPMI). They employ the model for emoji prediction task, and a metric based on emoji clustering is developed to better measure the validity of emoji prediction results. To validate the performance of TGCN-Bert, they compare it with other GCN variants on short text classification datasets and emoji prediction datasets; experiments show that TGCN-Bert achieves better performance.

Aerial image target detection is a challenging task due to the complex backgrounds, dense target distribution, and large-scale differences often present in aerial images. Existing methods often... Show More

Aerial image target detection is a challenging task due to the complex backgrounds, dense target distribution, and large-scale differences often present in aerial images. Existing methods often struggle to effectively extract detailed features and address the issue of imbalanced positive and negative samples. To tackle these challenges, an aerial image target detection method (dense RFB-FE-CGAM) based on dense RFB-FE and channel-global attention mechanism (CGAM) was proposed. First, the authors design a shallow feature enhancement module using dense RFB feature multiplexing and expand convolution within an SSD network, improving detailed feature extraction. Second, they introduce CGAM, a global attention module, to enhance semantic feature extraction in backbone networks. Finally, they incorporate a focal loss function for joint training, addressing sample imbalance. In experiments, the method achieved an mAP of 0.755 on the DOTA dataset and recall/AP values of 0.889/0.906 on HRSC2016, confirming the effectiveness of dense RFB-FE-CGAM for aerial image target detection.

Every year brings numerous security breaches that lead to highly destructive ransomware attacks, data leaks, and reputational damage to governments, companies, and other organizations around the... Show More

Every year brings numerous security breaches that lead to highly destructive ransomware attacks, data leaks, and reputational damage to governments, companies, and other organizations around the world. As a result, there is a growing need to ensure that workers comply with critical policies put in place to avoid such incidents. This study investigated how factors from social bond theory and involvement theory affected compliance with information security policies and procedures. All of the factors examined were found to have a significant influence on attitudes about compliance, and attitude had a significant impact on intention to comply. The findings of this study revealed that it is vital to raise employees' awareness about compliance with security policies by improving their information security behavior. Moreover, all the factors were found to have a significant influence on the attitude of employees towards compliance with their organizational information security policies and procedures.

A risk assessment method for urban railway investment and financing based on an improved SVM model under big data is proposed. First, the inner product in the traditional SVM is replaced by a kernel... Show More

A risk assessment method for urban railway investment and financing based on an improved SVM model under big data is proposed. First, the inner product in the traditional SVM is replaced by a kernel function to obtain a more accurate non-linear SVM, and a classifier with high classification accuracy is achieved by finding the optimal separating hyperplane. Then, a risk index system is constructed based on the grounded theory combining with intuitionistic fuzzy sets, interval intuitionistic fuzzy sets, weighted averaging operators and the distance measure, and the selection method of assessment indexes is analyzed based on the statistical methods. Finally, the SVM model with fuzzy membership is obtained by fuzzifying the input samples of the SVM based on the given rules of fuzzy membership design. The results show that the maximum relative error between the final test results and the actual value is 0.316%, and the minimum relative error is 0.133% with three different test sets being tested in the proposed method, which can accurately assess the investment.

In the coastal areas of China, scientists have collected nearly 500 species of coastal plants and seaweeds. The collected information includes species description, morphological characteristics... Show More

In the coastal areas of China, scientists have collected nearly 500 species of coastal plants and seaweeds. The collected information includes species description, morphological characteristics, habitat distribution and resource value of plants in China. By effectively extracting Chinese text information, this article establishes a Chinese text information extraction model based on DL. This article is based on short-term and short-term memory artificial neural networks for short text classification. In addition, this article also integrates the L-MFCNN models of MFCNN for short text classification. Comparing the two methods with traditional text recognition algorithms, information extraction based on syntax analysis and deep learning, the results show that, compared with the comparison method, the recognition accuracy of Chinese text information of this neural network model can reach 96.69%. Through model training and parameter adjustment, Chinese text information of coastal biodiversity can be quickly extracted, and species categories or names can be identified.

With the surge in online video content, finding highlights and key video segments have garnered widespread attention. Given a textual query, video highlight detection (HD) and temporal grounding... Show More

With the surge in online video content, finding highlights and key video segments have garnered widespread attention. Given a textual query, video highlight detection (HD) and temporal grounding (TG) aim to predict frame-wise saliency scores from a video while concurrently locating all relevant spans. Despite recent progress in DETR-based works, these methods crudely fuse different inputs in the encoder, which limits effective cross-modal interaction. To solve this challenge, the authors design QD-Net (query-guided refinement and dynamic spans network) tailored for HD&TG. Specifically, they propose a query-guided refinement module to decouple the feature encoding from the interaction process. Furthermore, they present a dynamic span decoder that leverages learnable 2D spans as decoder queries, which accelerates training convergence for TG. On QVHighlights dataset, the proposed QD-Net achieves 61.87 HD-HIT@1 and 61.88 TG-mAP@0.5, yielding a significant improvement of +1.88 and +8.05, respectively, compared to the state-of-the-art method.

An improved SegNet semantic segmentation model is proposed to address the issue of traditional classification algorithms and shallow learning algorithms not being suitable for extracting information... Show More

An improved SegNet semantic segmentation model is proposed to address the issue of traditional classification algorithms and shallow learning algorithms not being suitable for extracting information from high-resolution remote sensing images. During the research process, space remote sensing images obtained from the GF-1 satellite were used as the data source. In order to improve the operational efficiency of the encoding network, the pooling layer in the encoding network is removed and the ordinary convolutional layer is replaced with a depth-wise separable convolution. By decoding the last layer of the network to obtain the reshaped output results, and then calculating the probability of each classification using a Softmax classifier, the classification of pixels can be achieved. The output result of the classifier is the final result of the remote sensing image semantic segmentation model. The results showed that the proposed algorithm had the highest Kappa coefficient of 0.9531, indicating good classification performance.

Simultaneous localization and mapping (SLAM) serves as a cornerstone in autonomous systems and has seen exponential growth in its roles, particularly in facilitating advanced path planning... Show More

Simultaneous localization and mapping (SLAM) serves as a cornerstone in autonomous systems and has seen exponential growth in its roles, particularly in facilitating advanced path planning solutions. One emerging avenue of research that is rapidly evolving is the incorporation of multi-sensor fusion techniques to enhance SLAM-based path planning. The paper initiates with a thorough review of various sensor types and their attributes before covering a broad spectrum of both traditional and contemporary algorithms for multi-sensor fusion within SLAM. Performance evaluation metrics pertinent to SLAM and sensor fusion are explored. A special focus is laid on the interconnected roles and applications of multi-sensor fusion in SLAM-based path planning, discussing its significance in navigation scenarios as well as addressing challenges such as computational burden and real-time implementation. This paper sets the stage for future developments in creating more robust, resilient, and efficient SLAM-based path planning systems enabled by multi-sensor fusion.