Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or...Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or indirect slurs.To address this gap,we propose a hybrid framework combining Term Frequency-Inverse Document Frequency(TF-IDF),word-to-vector(Word2Vec),and Bidirectional Encoder Representations from Transformers(BERT)based models for multi-class cyberbullying detection.Our approach integrates TF-IDF for lexical specificity and Word2Vec for semantic relationships,fused with BERT’s contextual embeddings to capture syntactic and semantic complexities.We evaluate the framework on a publicly available dataset of 47,000 annotated social media posts across five cyberbullying categories:age,ethnicity,gender,religion,and indirect aggression.Among BERT variants tested,BERT Base Un-Cased achieved the highest performance with 93%accuracy(standard deviation across±1%5-fold cross-validation)and an average AUC of 0.96,outperforming standalone TF-IDF(78%)and Word2Vec(82%)models.Notably,it achieved near-perfect AUC scores(0.99)for age and ethnicity-based bullying.A comparative analysis with state-of-the-art benchmarks,including Generative Pre-trained Transformer 2(GPT-2)and Text-to-Text Transfer Transformer(T5)models highlights BERT’s superiority in handling ambiguous language.This work advances cyberbullying detection by demonstrating how hybrid feature extraction and transformer models improve multi-class classification,offering a scalable solution for moderating nuanced harmful content.展开更多
We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-expl...We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-explicit Runge-Kutta method for time integration. The resulting method retains the simplicity of the relaxation schemes. There is no need to involve Riemann solvers and characteristic decomposition. Even the computation of the eigenvalues is not required. This makes the scheme particularly well suited for the MCLWR model in which the analytical expressions of the eigenvalues are difficult to obtain for more than four classes of road users. The numerical results illustrate the effectiveness of the presented method.展开更多
A high-resolution relaxed scheme which requires little information of the eigenstructure is presented for the multiclass Lighthill-Whitham-Richards （LWR） model on an inhomogeneous highway. The scheme needs only an e...A high-resolution relaxed scheme which requires little information of the eigenstructure is presented for the multiclass Lighthill-Whitham-Richards （LWR） model on an inhomogeneous highway. The scheme needs only an estimate of the upper boundary of the maximum of absolute eigenvalues. It is based on incorporating an improved fifth-order weighted essentially non-oscillatory （WENO） reconstruction with relaxation approximation. The scheme benefits from the simplicity of relaxed schemes in that it requires no exact or approximate Riemann solvers and no projection along characteristic directions. The effectiveness of our method is demonstrated in several numerical examples.展开更多
Based on field research data on farmers in 30 counties(districts)of six cities in Shandong Province,this study systematically examines the methods of disposal of livestock and poultry waste and the factors that influe...Based on field research data on farmers in 30 counties(districts)of six cities in Shandong Province,this study systematically examines the methods of disposal of livestock and poultry waste and the factors that influence them,that is,the determinants of farmers’waste disposal behaviors,using the UTAUT theoretical framework and a disordered multi-class logit model.The results show that,first,economic performance expectancy,subjective norms,farming population,and poultry and livestock breeding numbers have significant effects on the four waste recycling methods considered:direct return,compost fermentation,biogas fermentation,and fresh-packed sale.Second,annual family income has a positive effect on the direct return method.Third,compost fermentation is positively affected by farmers'knowledge,distance from sources of water,and farming scale,and negatively affected by gender and marital status.Fourth,biogas fermentation is negatively affected by the age of farmers and positively affected by the farming income ratio and environmental convenience.Finally,the most important factors affecting the four methods are subjective norms,farming scale,economic performance expectancy,and farming population.Therefore,this study proposes that policies to promote resource utilization of livestock and poultry waste must prioritize raising awareness among the farmers of the advantages of resource disposal of waste;provide reasonable subsidies for waste recycling and enhance policy applicability,pertinence,and motivation;strengthen the promotion of waste recycling technology and improve the practicality of lectures or training content;and improve relevant laws and regulations and enhance authority and usability.展开更多
In today’s information technology(IT)world,the multi-hop wireless sensor networks(MHWSNs)are considered the building block for the Internet of Things(IoT)enabled communication systems for controlling everyday tasks o...In today’s information technology(IT)world,the multi-hop wireless sensor networks(MHWSNs)are considered the building block for the Internet of Things(IoT)enabled communication systems for controlling everyday tasks of organizations and industry to provide quality of service(QoS)in a stipulated time slot to end-user over the Internet.Smart city(SC)is an example of one such application which can automate a group of civil services like automatic control of traffic lights,weather prediction,surveillance,etc.,in our daily life.These IoT-based networks with multi-hop communication and multiple sink nodes provide efficient communication in terms of performance parameters such as throughput,energy efficiency,and end-to-end delay,wherein low latency is considered a challenging issue in next-generation networks(NGN).This paper introduces a single and parallels stable server queuing model with amulti-class of packets and native and coded packet flowto illustrate the simple chain topology and complexmultiway relay(MWR)node with specific neighbor topology.Further,for improving data transmission capacity inMHWSNs,an analytical framework for packet transmission using network coding at the MWR node in the network layer with opportunistic listening is performed by considering bi-directional network flow at the MWR node.Finally,the accuracy of the proposed multi-server multi-class queuing model is evaluated with and without network coding at the network layer by transmitting data packets.The results of the proposed analytical framework are validated and proved effective by comparing these analytical results to simulation results.展开更多
This paper proposes an improved multi-class dynamic network traffic flow propagation model with a consideration of physical queues. Each link is divided into two areas: Free flow area and queue area. The vehicles of t...This paper proposes an improved multi-class dynamic network traffic flow propagation model with a consideration of physical queues. Each link is divided into two areas: Free flow area and queue area. The vehicles of the same class are assumed to satisfy the first-in-first-out(FIFO) principle on the whole link, and the vehicles of the different classes also follow FIFO in the queue area but not in the free flow area. To characterize this phenomenon by numerical methods, the improved model is directly formulated in discrete time space. Numerical examples are developed to illustrate the unrealistic flows of the existing model and the performance of the improved model. This analysis can more realistically capture the traffic flow propagation, such as interactions between multi-class traffic flows, and the dynamic traffic interactions across multiple links.展开更多
Despite the widespread use of machine learning(ML)models for geospatial applications,adaptations to imbalanced multitemporal land cover(LC)datasets remain underexplored.For over two dec-ades,studies have predominantly...Despite the widespread use of machine learning(ML)models for geospatial applications,adaptations to imbalanced multitemporal land cover(LC)datasets remain underexplored.For over two dec-ades,studies have predominantly trained ML models on a single interval of LC data to model changes,with detriments of imbal-anced training datasets managed through manual manipulations.Therefore,this study proposes and implements an ML-spatial sam-ple weighting(ML-SSW)approach to leverage available multitem-poral LC data while adjusting sample influence to reflect recency of change occurrence and class-level spatial pattern measures to enable data-driven LC change modeling.Random Forest(RF),Neural Network(NN),and Extreme Gradient Boosting Machine(XGB)models are trained under the ML-SSW strategy on three study areas located in British Columbia,Canada.The RF-SSW,NN-SSW,and XGB-SSW models forecasted more realistic changes across multiple timesteps with fewer errors than baseline configurations.The presented methodology provides a step toward establishing spatialized cost-sensitive learning strategies and extending classical ML models to multitemporal LC datasets.展开更多
基金funded by Scientific Research Deanship at University of Hail-Saudi Arabia through Project Number RG-23092.
文摘Cyberbullying on social media poses significant psychological risks,yet most detection systems over-simplify the task by focusing on binary classification,ignoring nuanced categories like passive-aggressive remarks or indirect slurs.To address this gap,we propose a hybrid framework combining Term Frequency-Inverse Document Frequency(TF-IDF),word-to-vector(Word2Vec),and Bidirectional Encoder Representations from Transformers(BERT)based models for multi-class cyberbullying detection.Our approach integrates TF-IDF for lexical specificity and Word2Vec for semantic relationships,fused with BERT’s contextual embeddings to capture syntactic and semantic complexities.We evaluate the framework on a publicly available dataset of 47,000 annotated social media posts across five cyberbullying categories:age,ethnicity,gender,religion,and indirect aggression.Among BERT variants tested,BERT Base Un-Cased achieved the highest performance with 93%accuracy(standard deviation across±1%5-fold cross-validation)and an average AUC of 0.96,outperforming standalone TF-IDF(78%)and Word2Vec(82%)models.Notably,it achieved near-perfect AUC scores(0.99)for age and ethnicity-based bullying.A comparative analysis with state-of-the-art benchmarks,including Generative Pre-trained Transformer 2(GPT-2)and Text-to-Text Transfer Transformer(T5)models highlights BERT’s superiority in handling ambiguous language.This work advances cyberbullying detection by demonstrating how hybrid feature extraction and transformer models improve multi-class classification,offering a scalable solution for moderating nuanced harmful content.
基金Project supported by the Aoxiang Project and the Scientific and Technological Innovation Foundation of Northwestern Polytechnical University, China (No 2007KJ01011)
文摘We present a high-resolution relaxation scheme for a multi-class Lighthill-Whitham-Richards (MCLWR) traffic flow model. This scheme is based on high-order reconstruction for spatial discretization and an implicit-explicit Runge-Kutta method for time integration. The resulting method retains the simplicity of the relaxation schemes. There is no need to involve Riemann solvers and characteristic decomposition. Even the computation of the eigenvalues is not required. This makes the scheme particularly well suited for the MCLWR model in which the analytical expressions of the eigenvalues are difficult to obtain for more than four classes of road users. The numerical results illustrate the effectiveness of the presented method.
基金Project supported by the National Natural Science Foundation of China （No. 11102165） and the Special Fund for Basic Scientific Research of Central Colleges, Chang＇an University, China （No. CHD 2011 JC039）
文摘A high-resolution relaxed scheme which requires little information of the eigenstructure is presented for the multiclass Lighthill-Whitham-Richards （LWR） model on an inhomogeneous highway. The scheme needs only an estimate of the upper boundary of the maximum of absolute eigenvalues. It is based on incorporating an improved fifth-order weighted essentially non-oscillatory （WENO） reconstruction with relaxation approximation. The scheme benefits from the simplicity of relaxed schemes in that it requires no exact or approximate Riemann solvers and no projection along characteristic directions. The effectiveness of our method is demonstrated in several numerical examples.
基金supported by the National Natural Science Foundation of China“Behavioral Experiment and Policy Research of Pig Farmers when Diseased Dead Pigs Entering the Market”[Gant number 71540008]the Key Projects of National Natural Science Foundation of China“Experimental Assessment of Agricultural Producer Safety Production Policy and Its Combined Design:Taking the Treatment of Diseased Dead Pigs as an Example”[Grant number 71673115]+3 种基金the Fundamental Research Funds for the Central Universities“Food Safety Risk Management Logic and Realistic Path Based on Big Data”[Grant number.JUSRP1808ZD]the National College Student Innovation Training Program[grant number.201810295025]the National Key R&D Program Funding Project[Grant number.2018YFC1603303]the National Key R&D Program Funding Project[Grant number.2018YFC1604000].
文摘Based on field research data on farmers in 30 counties(districts)of six cities in Shandong Province,this study systematically examines the methods of disposal of livestock and poultry waste and the factors that influence them,that is,the determinants of farmers’waste disposal behaviors,using the UTAUT theoretical framework and a disordered multi-class logit model.The results show that,first,economic performance expectancy,subjective norms,farming population,and poultry and livestock breeding numbers have significant effects on the four waste recycling methods considered:direct return,compost fermentation,biogas fermentation,and fresh-packed sale.Second,annual family income has a positive effect on the direct return method.Third,compost fermentation is positively affected by farmers'knowledge,distance from sources of water,and farming scale,and negatively affected by gender and marital status.Fourth,biogas fermentation is negatively affected by the age of farmers and positively affected by the farming income ratio and environmental convenience.Finally,the most important factors affecting the four methods are subjective norms,farming scale,economic performance expectancy,and farming population.Therefore,this study proposes that policies to promote resource utilization of livestock and poultry waste must prioritize raising awareness among the farmers of the advantages of resource disposal of waste;provide reasonable subsidies for waste recycling and enhance policy applicability,pertinence,and motivation;strengthen the promotion of waste recycling technology and improve the practicality of lectures or training content;and improve relevant laws and regulations and enhance authority and usability.
文摘In today’s information technology(IT)world,the multi-hop wireless sensor networks(MHWSNs)are considered the building block for the Internet of Things(IoT)enabled communication systems for controlling everyday tasks of organizations and industry to provide quality of service(QoS)in a stipulated time slot to end-user over the Internet.Smart city(SC)is an example of one such application which can automate a group of civil services like automatic control of traffic lights,weather prediction,surveillance,etc.,in our daily life.These IoT-based networks with multi-hop communication and multiple sink nodes provide efficient communication in terms of performance parameters such as throughput,energy efficiency,and end-to-end delay,wherein low latency is considered a challenging issue in next-generation networks(NGN).This paper introduces a single and parallels stable server queuing model with amulti-class of packets and native and coded packet flowto illustrate the simple chain topology and complexmultiway relay(MWR)node with specific neighbor topology.Further,for improving data transmission capacity inMHWSNs,an analytical framework for packet transmission using network coding at the MWR node in the network layer with opportunistic listening is performed by considering bi-directional network flow at the MWR node.Finally,the accuracy of the proposed multi-server multi-class queuing model is evaluated with and without network coding at the network layer by transmitting data packets.The results of the proposed analytical framework are validated and proved effective by comparing these analytical results to simulation results.
基金jointly supported by the National Natural Science Foundation of China (Grant Nos. 71571150 and 71361006)the Humanities and Social Science Foundation of The Ministry of Education (Grant No. 14YJA630026)the Fundamental Research Funds for the Central Universities (Grant No. 26815WCX03)
文摘This paper proposes an improved multi-class dynamic network traffic flow propagation model with a consideration of physical queues. Each link is divided into two areas: Free flow area and queue area. The vehicles of the same class are assumed to satisfy the first-in-first-out(FIFO) principle on the whole link, and the vehicles of the different classes also follow FIFO in the queue area but not in the free flow area. To characterize this phenomenon by numerical methods, the improved model is directly formulated in discrete time space. Numerical examples are developed to illustrate the unrealistic flows of the existing model and the performance of the improved model. This analysis can more realistically capture the traffic flow propagation, such as interactions between multi-class traffic flows, and the dynamic traffic interactions across multiple links.
基金supported by the Natural Sciences and Engineering Research Council of Canada[RGPIN-2023-04052].
文摘Despite the widespread use of machine learning(ML)models for geospatial applications,adaptations to imbalanced multitemporal land cover(LC)datasets remain underexplored.For over two dec-ades,studies have predominantly trained ML models on a single interval of LC data to model changes,with detriments of imbal-anced training datasets managed through manual manipulations.Therefore,this study proposes and implements an ML-spatial sam-ple weighting(ML-SSW)approach to leverage available multitem-poral LC data while adjusting sample influence to reflect recency of change occurrence and class-level spatial pattern measures to enable data-driven LC change modeling.Random Forest(RF),Neural Network(NN),and Extreme Gradient Boosting Machine(XGB)models are trained under the ML-SSW strategy on three study areas located in British Columbia,Canada.The RF-SSW,NN-SSW,and XGB-SSW models forecasted more realistic changes across multiple timesteps with fewer errors than baseline configurations.The presented methodology provides a step toward establishing spatialized cost-sensitive learning strategies and extending classical ML models to multitemporal LC datasets.
