In response to the rapid growth of future business and data traffic, the widespread deployment of small base stations (BSs) in 5G networks has emerged as a promising solution, albeit intensifying the network's energy consumption. Additionally, traditional BSs lack adaptive adjustment of parameter information, posing challenges in delivering satisfactory quality of service (QoS), particularly in the context of highly uneven business distribution. Reducing energy consumption while ensuring that QoS represents a critical challenge. To address this issue, this paper first comprehensively considers the interests of both supply and demand, proposing a service utility measurement method in communication networks to achieve a balance between energy consumption and QoS. Furthermore, this paper integrates cell zooming and sleeping strategies for small BSs, designing a dynamic game algorithm aimed at optimizing service utility in a two-tier heterogeneous network. Through ten distinct scenario simulations, our proposed algorithm demonstrates significant enhancements in service utility while achieving near-optimal optimization results more expeditiously compared to the genetic algorithm (GA).
{"title":"Optimization of Service Utility in 5G Heterogeneous Networks Using Dynamic Game","authors":"Linhao Zhang;Xudong Lu;Lizhen Cui;Deyu Zhou;Wei Guo","doi":"10.26599/IJCS.2024.9100023","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100023","url":null,"abstract":"In response to the rapid growth of future business and data traffic, the widespread deployment of small base stations (BSs) in 5G networks has emerged as a promising solution, albeit intensifying the network's energy consumption. Additionally, traditional BSs lack adaptive adjustment of parameter information, posing challenges in delivering satisfactory quality of service (QoS), particularly in the context of highly uneven business distribution. Reducing energy consumption while ensuring that QoS represents a critical challenge. To address this issue, this paper first comprehensively considers the interests of both supply and demand, proposing a service utility measurement method in communication networks to achieve a balance between energy consumption and QoS. Furthermore, this paper integrates cell zooming and sleeping strategies for small BSs, designing a dynamic game algorithm aimed at optimizing service utility in a two-tier heterogeneous network. Through ten distinct scenario simulations, our proposed algorithm demonstrates significant enhancements in service utility while achieving near-optimal optimization results more expeditiously compared to the genetic algorithm (GA).","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 4","pages":"159-167"},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10681260","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142235845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.26599/IJCS.2024.9100024
Guowen Li;Wenbo Hu;Yang Zhao;Xudong Lu
The dense deployment of Femto Base Stations (FBS) assisting Macro Base Stations (MBS) in a Heterogeneous Network (HetNet) resolves the coverage issue of 5G signal transmission. However, the imprudent layout of FBSs results in extensive energy consumption and increased signal interference among base stations. Regulating the transmission power of each base station in the HetNets through the main controller or MBS is essential to maximize the power efficiency of the entire HetNets while adhering to the constraints of basic signal throughput and fairness. To address this challenge, this paper proposes an Adaptive Acceleration Particle Swarm Optimization (AA-PSO) algorithm. This algorithm dynamically determines the inertia weight based on each particle's optimal position and the global optimal position, and introduces the concept of time-varying parameters to control the learning rate, thus managing the search range and convergence speed of the particle swarm. The results demonstrate that the AA-PSO algorithm can efficiently determine the optimal transmission power of each base station in the HetNets, reduce interference between MBS and FBSs, as well as among FBSs, and ultimately improve the service efficacy of the entire network.
{"title":"Optimizing Service Efficacy in 5G HetNets: An Adaptive Acceleration PSO Approach","authors":"Guowen Li;Wenbo Hu;Yang Zhao;Xudong Lu","doi":"10.26599/IJCS.2024.9100024","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100024","url":null,"abstract":"The dense deployment of Femto Base Stations (FBS) assisting Macro Base Stations (MBS) in a Heterogeneous Network (HetNet) resolves the coverage issue of 5G signal transmission. However, the imprudent layout of FBSs results in extensive energy consumption and increased signal interference among base stations. Regulating the transmission power of each base station in the HetNets through the main controller or MBS is essential to maximize the power efficiency of the entire HetNets while adhering to the constraints of basic signal throughput and fairness. To address this challenge, this paper proposes an Adaptive Acceleration Particle Swarm Optimization (AA-PSO) algorithm. This algorithm dynamically determines the inertia weight based on each particle's optimal position and the global optimal position, and introduces the concept of time-varying parameters to control the learning rate, thus managing the search range and convergence speed of the particle swarm. The results demonstrate that the AA-PSO algorithm can efficiently determine the optimal transmission power of each base station in the HetNets, reduce interference between MBS and FBSs, as well as among FBSs, and ultimately improve the service efficacy of the entire network.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 4","pages":"168-175"},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10681262","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142235873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.26599/IJCS.2024.9100013
Bo Xu;Xinpu Su;Yijun He
Predicting financial market trends poses significant challenges due to the complex, dynamic, and often chaotic nature of the market, especially when dealing with data featuring a multitude of characteristics. In this research, we propose an effective data mining approach that combines Least Absolute Shrinkage and Selection Operator (LASSO) and Principal Component Analysis (PCA) for two-stage feature dimensionality reduction, resulting in a refined dataset. To enhance the model's capacity to capture medium- and long-term index trends, we implement a sliding time window approach, utilizing data from the preceding 60 trading days. Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) models are employed to predict the return rate of Securities Bank of China (399986) over a 30-day trading period. We conducted a comprehensive comparative analysis involving our proposed model and established methods, namely, LASSO, PCA, and Hybrid Recurrent Neural Networks (RNN). Our empirical findings unequivocally demonstrate the superior performance of our model in terms of both prediction accuracy and stability. Specifically, our model exhibits significantly higher predictive accuracy when forecasting the return rate of Securities Bank of China (399986) over a 30-day trading period, all while maintaining enhanced stability. These results underscore the exceptional efficacy of our approach within the realm of financial market time series forecasting, thus providing robust support for further research and practical applications within this domain.
{"title":"Index Prediction Model Based on LASSO-PCA and Deep Learning","authors":"Bo Xu;Xinpu Su;Yijun He","doi":"10.26599/IJCS.2024.9100013","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100013","url":null,"abstract":"Predicting financial market trends poses significant challenges due to the complex, dynamic, and often chaotic nature of the market, especially when dealing with data featuring a multitude of characteristics. In this research, we propose an effective data mining approach that combines Least Absolute Shrinkage and Selection Operator (LASSO) and Principal Component Analysis (PCA) for two-stage feature dimensionality reduction, resulting in a refined dataset. To enhance the model's capacity to capture medium- and long-term index trends, we implement a sliding time window approach, utilizing data from the preceding 60 trading days. Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) models are employed to predict the return rate of Securities Bank of China (399986) over a 30-day trading period. We conducted a comprehensive comparative analysis involving our proposed model and established methods, namely, LASSO, PCA, and Hybrid Recurrent Neural Networks (RNN). Our empirical findings unequivocally demonstrate the superior performance of our model in terms of both prediction accuracy and stability. Specifically, our model exhibits significantly higher predictive accuracy when forecasting the return rate of Securities Bank of China (399986) over a 30-day trading period, all while maintaining enhanced stability. These results underscore the exceptional efficacy of our approach within the realm of financial market time series forecasting, thus providing robust support for further research and practical applications within this domain.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 4","pages":"176-183"},"PeriodicalIF":0.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10681264","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142235874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-19DOI: 10.26599/IJCS.2023.9100026
Xiaonuo Yang;Yueting Chai
With heart health issues attracting much attention, wearable electrocardiogram (ECG) monitoring devices show a broad market prospect. This paper develops a generic ECG pre-processing algorithm and proposes a method for the single-lead ECG classification problem based on model stacking. Features such as RR-intervals, power spectrum, and higher-order statistics are computed and grouped into three classes. The support vector machine (SVM) classifier is trained separately based on each class of features, and subsequently, a fourth SVM classifier is trained on the prediction results of the three SVM classifiers at the first layer. To obtain more realistic results and better comparisons with previous studies, the algorithm follows the ANSI/AAMI EC57:2012 standard and is tested on a real ECG database. The experimental results show that the algorithm in this paper better overcomes the impact of the data imbalance problem and obtains good results.
{"title":"ECG Signal Processing and Automatic Classification Algorithms","authors":"Xiaonuo Yang;Yueting Chai","doi":"10.26599/IJCS.2023.9100026","DOIUrl":"https://doi.org/10.26599/IJCS.2023.9100026","url":null,"abstract":"With heart health issues attracting much attention, wearable electrocardiogram (ECG) monitoring devices show a broad market prospect. This paper develops a generic ECG pre-processing algorithm and proposes a method for the single-lead ECG classification problem based on model stacking. Features such as RR-intervals, power spectrum, and higher-order statistics are computed and grouped into three classes. The support vector machine (SVM) classifier is trained separately based on each class of features, and subsequently, a fourth SVM classifier is trained on the prediction results of the three SVM classifiers at the first layer. To obtain more realistic results and better comparisons with previous studies, the algorithm follows the ANSI/AAMI EC57:2012 standard and is tested on a real ECG database. The experimental results show that the algorithm in this paper better overcomes the impact of the data imbalance problem and obtains good results.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"122-129"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638681","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Heterogeneous Electric Vehicle Routing Problem with Multiple Compartments and Multiple Trips for the Collection of Classified Waste","authors":"","doi":"","DOIUrl":"https://doi.org/","url":null,"abstract":"","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"1-3"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638668","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-19DOI: 10.26599/IJCS.2024.9100001
Toh Hsiang Benny Tan;Sufang Lim;Chan Hua Nicholas Vun
Mental health challenges, accentuated by stress, are escalating in high-income countries, especially among adolescents and university students. Traditional mental health approaches face issues such as scalability and accessibility, making the emergence of digital tools crucial. However, adherence remains a challenge. This study examines the role of technology self-efficacy and digital alliance in influencing the acceptance of digital mental health tools among Singaporean university students. The results provide strong support for the role of digital alliance as a key factor impacting a student's intention to use mental health tools, as well as technology self-efficacy and digital alliance as serial mediators of task-technology fit and intention to use, highlighting our ever-evolving relationship with technology.
{"title":"Role of Technology Self-Efficacy and Digital Alliance in Digital Mental Health Tool Acceptance Among University Students in Singapore","authors":"Toh Hsiang Benny Tan;Sufang Lim;Chan Hua Nicholas Vun","doi":"10.26599/IJCS.2024.9100001","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100001","url":null,"abstract":"Mental health challenges, accentuated by stress, are escalating in high-income countries, especially among adolescents and university students. Traditional mental health approaches face issues such as scalability and accessibility, making the emergence of digital tools crucial. However, adherence remains a challenge. This study examines the role of technology self-efficacy and digital alliance in influencing the acceptance of digital mental health tools among Singaporean university students. The results provide strong support for the role of digital alliance as a key factor impacting a student's intention to use mental health tools, as well as technology self-efficacy and digital alliance as serial mediators of task-technology fit and intention to use, highlighting our ever-evolving relationship with technology.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"101-109"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638680","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-19DOI: 10.26599/IJCS.2024.9100020
Qian Liu;Yang Lyu;Jian Tang;Weiguo Fan
Curbing the dissemination of fake news in social media has been a major issue in recent years. Previous studies have suggested that the general public can recognize fake news, showing the feasibility of applying crowd ratings to identify fake news. However, the effectiveness of crowd ratings for curbing the dissemination of fake news is uncertain. This study constructed an online experimental platform to simulate Sina Microblog and designed a crowd rating strategy to compare and validate the difference between the absence vs. the presence of crowd ratings, and crowd ratings vs. expert ratings, in curbing the dissemination of fake news. We found that the presence of crowd ratings inhibited users' dissemination of fake news compared to the absence of crowd ratings. Moreover, there was no significant difference between the suppression effects of crowd ratings and expert ratings, both of which were effective in curbing the dissemination of fake news. Crowd rating uses collective intelligence to intervene in users' perceptions and behaviors at the onset of fake news dissemination, providing a cost-effective and efficient solution to combat the spread of fake news on social media.
{"title":"Optimizing the Service Efficacy of Crowd Ratings in Curbing Fake News Dissemination on Social Media","authors":"Qian Liu;Yang Lyu;Jian Tang;Weiguo Fan","doi":"10.26599/IJCS.2024.9100020","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100020","url":null,"abstract":"Curbing the dissemination of fake news in social media has been a major issue in recent years. Previous studies have suggested that the general public can recognize fake news, showing the feasibility of applying crowd ratings to identify fake news. However, the effectiveness of crowd ratings for curbing the dissemination of fake news is uncertain. This study constructed an online experimental platform to simulate Sina Microblog and designed a crowd rating strategy to compare and validate the difference between the absence vs. the presence of crowd ratings, and crowd ratings vs. expert ratings, in curbing the dissemination of fake news. We found that the presence of crowd ratings inhibited users' dissemination of fake news compared to the absence of crowd ratings. Moreover, there was no significant difference between the suppression effects of crowd ratings and expert ratings, both of which were effective in curbing the dissemination of fake news. Crowd rating uses collective intelligence to intervene in users' perceptions and behaviors at the onset of fake news dissemination, providing a cost-effective and efficient solution to combat the spread of fake news on social media.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"110-121"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638683","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-19DOI: 10.26599/IJCS.2024.9100015
Leiju Qiu;Xiao Sun;Yong Tu;Yang Zhao
The high efficacy of metro network services not only enhances residents' travel quality but also brings significant socio-economic benefits, thus is of great importance to urban land use and city development. Existing methods for measuring metro service efficacy often overlook metro network connectivity and rely heavily on subjective questionnaire data analysis from the user experience perspective. This paper proposes a method to measure metro network service efficacy from the user's perspective. The approach first calculates the connectivity index of metro network and estimates the housing premium brought by metro network connectivity, which reveals users' willingness to pay for metro network connectivity. This method objectively measures metro network service efficacy from the user's perspective. Based on this, efficacy optimization methods are proposed, providing quantitative simulation methods for metro expansion, site selection, operation quality adjustments, etc., which are of great reference value to metro management departments and even urban sustainable development.
{"title":"Measurement and Optimization of Metro Network Service Efficacy","authors":"Leiju Qiu;Xiao Sun;Yong Tu;Yang Zhao","doi":"10.26599/IJCS.2024.9100015","DOIUrl":"https://doi.org/10.26599/IJCS.2024.9100015","url":null,"abstract":"The high efficacy of metro network services not only enhances residents' travel quality but also brings significant socio-economic benefits, thus is of great importance to urban land use and city development. Existing methods for measuring metro service efficacy often overlook metro network connectivity and rely heavily on subjective questionnaire data analysis from the user experience perspective. This paper proposes a method to measure metro network service efficacy from the user's perspective. The approach first calculates the connectivity index of metro network and estimates the housing premium brought by metro network connectivity, which reveals users' willingness to pay for metro network connectivity. This method objectively measures metro network service efficacy from the user's perspective. Based on this, efficacy optimization methods are proposed, providing quantitative simulation methods for metro expansion, site selection, operation quality adjustments, etc., which are of great reference value to metro management departments and even urban sustainable development.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"149-158"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638682","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, great success has been achieved in many tasks of natural language processing (NLP), e.g., named entity recognition (NER), especially in the high-resource language, i.e., English, thanks in part to the considerable amount of labeled resources. More labeled resources, better word representations. However, most low-resource languages do not have such an abundance of labeled data as high-resource English, leading to poor performance of NER in these low-resource languages due to poor word representations. In the paper, we propose converse attention network (CAN) to augment word representations in low-resource languages from the high-resource language, improving the performance of NER in low-resource languages by transferring knowledge learned in the high-resource language. CAN first translates sentences in low-resource languages into high-resource English using an attention-based translation module. In the process of translation, CAN obtains the attention matrices that align word representations of high-resource language space and low-resource language space. Furthermore, CAN augments word representations learned in low-resource language space with word representations learned in high-resource language space using the attention matrices. Experiments on four low-resource NER datasets show that CAN achieves consistent and significant performance improvements, which indicates the effectiveness of CAN.
近年来，自然语言处理（NLP）的许多任务都取得了巨大成功，例如命名实体识别（NER），特别是在高资源语言（即英语）中，这在一定程度上要归功于大量的标注资源。更多的标注资源，更好的词语表征。然而，大多数低资源语言并不像高资源英语那样拥有如此丰富的标注数据，从而导致这些低资源语言的 NER 因单词表征不佳而表现不佳。在本文中，我们提出了反向注意力网络（CAN）来增强高资源语言在低资源语言中的单词表征，通过转移在高资源语言中学到的知识来提高低资源语言的 NER 性能。CAN 首先使用基于注意力的翻译模块将低资源语言的句子翻译成高资源英语。在翻译过程中，CAN 获得了将高资源语言空间和低资源语言空间的单词表征对齐的注意力矩阵。此外，CAN 还利用注意力矩阵将在低资源语言空间学习到的单词表征与在高资源语言空间学习到的单词表征进行增强。在四个低资源 NER 数据集上进行的实验表明，CAN 实现了持续而显著的性能提升，这表明了 CAN 的有效性。
{"title":"Converse Attention Knowledge Transfer for Low-Resource Named Entity Recognition","authors":"Shengfei Lyu;Linghao Sun;Huixiong Yi;Yong Liu;Huanhuan Chen;Chunyan Miao","doi":"10.26599/IJCS.2023.9100014","DOIUrl":"https://doi.org/10.26599/IJCS.2023.9100014","url":null,"abstract":"In recent years, great success has been achieved in many tasks of natural language processing (NLP), e.g., named entity recognition (NER), especially in the high-resource language, i.e., English, thanks in part to the considerable amount of labeled resources. More labeled resources, better word representations. However, most low-resource languages do not have such an abundance of labeled data as high-resource English, leading to poor performance of NER in these low-resource languages due to poor word representations. In the paper, we propose converse attention network (CAN) to augment word representations in low-resource languages from the high-resource language, improving the performance of NER in low-resource languages by transferring knowledge learned in the high-resource language. CAN first translates sentences in low-resource languages into high-resource English using an attention-based translation module. In the process of translation, CAN obtains the attention matrices that align word representations of high-resource language space and low-resource language space. Furthermore, CAN augments word representations learned in low-resource language space with word representations learned in high-resource language space using the attention matrices. Experiments on four low-resource NER datasets show that CAN achieves consistent and significant performance improvements, which indicates the effectiveness of CAN.","PeriodicalId":32381,"journal":{"name":"International Journal of Crowd Science","volume":"8 3","pages":"140-148"},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10638678","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142013156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：