Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949252
Mingli Wu, Yafei Huang, Jianyong Duan
As there is an increasing trend of people consuming by debit in China, financial organizations deal with a lot of loan applications. If customers cannot repay the loans on time, the organizations have to cover the loss. Therefore it is important to predict correctly whether a customer will repay the loan on time. Typical machine learning methods can be employed to exploit customers' financial information and give valuable judgements. We investigated the function of Deep Neural Network (DNN) in this work, as it achieves high successful rate in fields of image recognition, speech recognition and natural language processing. We compared it with traditional learning methods, such as Naïve Bayes, decision tree and K-Nearest Neighbor. Experiments showed that DNN achieves better performance than its traditional competitors. The accuracy and recall of DNN are 0.73 and 0.42 respectively. Its It-score is 25% higher than the best one of traditional methods.
{"title":"Investigations on Classification Methods for Loan Application Based on Machine Learning","authors":"Mingli Wu, Yafei Huang, Jianyong Duan","doi":"10.1109/ICMLC48188.2019.8949252","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949252","url":null,"abstract":"As there is an increasing trend of people consuming by debit in China, financial organizations deal with a lot of loan applications. If customers cannot repay the loans on time, the organizations have to cover the loss. Therefore it is important to predict correctly whether a customer will repay the loan on time. Typical machine learning methods can be employed to exploit customers' financial information and give valuable judgements. We investigated the function of Deep Neural Network (DNN) in this work, as it achieves high successful rate in fields of image recognition, speech recognition and natural language processing. We compared it with traditional learning methods, such as Naïve Bayes, decision tree and K-Nearest Neighbor. Experiments showed that DNN achieves better performance than its traditional competitors. The accuracy and recall of DNN are 0.73 and 0.42 respectively. Its It-score is 25% higher than the best one of traditional methods.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125717278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949205
Shu-Chen Cheng, Chun Lu
When users input keywords into the search engine, a massive search results will be retrieved. However, it becomes difficult for the users to learn as it is unreadable with the excessive amount of results. This study establishes an information retrieval system for computer science related articles. It firstly collects articles by running a web crawler, and uses TF-IDF (Term Frequency-Inverse Document Frequency) method to extract keywords to acquire the focus of the article. And with the use of association rules and cosine similarity, the articles are classified by their relevance. Finally, according to users' feedbacks, the system provides appropriate resources to improve the motivation and willingness to learn. In addition, the pictures in the articles are also a basis for analyzing the articles. This study uses image semantic analysis to label the pictures so as to improve the accuracy in analyzing the articles.
{"title":"Retrieving Articles and Image Labeling Based on Relevance of Keywords","authors":"Shu-Chen Cheng, Chun Lu","doi":"10.1109/ICMLC48188.2019.8949205","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949205","url":null,"abstract":"When users input keywords into the search engine, a massive search results will be retrieved. However, it becomes difficult for the users to learn as it is unreadable with the excessive amount of results. This study establishes an information retrieval system for computer science related articles. It firstly collects articles by running a web crawler, and uses TF-IDF (Term Frequency-Inverse Document Frequency) method to extract keywords to acquire the focus of the article. And with the use of association rules and cosine similarity, the articles are classified by their relevance. Finally, according to users' feedbacks, the system provides appropriate resources to improve the motivation and willingness to learn. In addition, the pictures in the articles are also a basis for analyzing the articles. This study uses image semantic analysis to label the pictures so as to improve the accuracy in analyzing the articles.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128027683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949260
Jian Zhang, Zijiang Yang, Y. Benslimane
The combination of data mining and machine learning technology with web-based education system is becoming an imperative research area to enhance the quality of education beyond the traditional concept. With the worldwide fast growth of the Information Communication Technology (ICT), data come with significant large volume, high velocity and extensive variety. In this paper, four popular data mining methods are applied on Apache Spark using large volume of datasets from Online Cognitive Learning Systems to explore the scalability and efficiency of Spark. Various volumes of datasets are tested on Spark MLlib with different running configurations and parameter tunings. The output of the paper convincingly presents useful strategies of computing resource allocation and tuning to make full advantage of the in-memory system of Apache Spark with the tasks of data mining and machine learning on educational datasets.
{"title":"Exploring and Evaluating the Scalability and Eficinecy of Apache Spark Using Educational Datasets","authors":"Jian Zhang, Zijiang Yang, Y. Benslimane","doi":"10.1109/ICMLC48188.2019.8949260","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949260","url":null,"abstract":"The combination of data mining and machine learning technology with web-based education system is becoming an imperative research area to enhance the quality of education beyond the traditional concept. With the worldwide fast growth of the Information Communication Technology (ICT), data come with significant large volume, high velocity and extensive variety. In this paper, four popular data mining methods are applied on Apache Spark using large volume of datasets from Online Cognitive Learning Systems to explore the scalability and efficiency of Spark. Various volumes of datasets are tested on Spark MLlib with different running configurations and parameter tunings. The output of the paper convincingly presents useful strategies of computing resource allocation and tuning to make full advantage of the in-memory system of Apache Spark with the tasks of data mining and machine learning on educational datasets.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116601904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949265
C. Hsieh, Dung-Ching Lin, Chengjia Wang, Zong-Ting Chen, Jiun-Jian Liaw
Car accident is a serious social problem which often results in both life loss and financial loss. Most of car accidents are caused by a lack of safe distance between cars. To relieve this problem, in this paper we propose a real-time car detection and safety alarm system. The proposed system consists of two modules: real-time car detection module and safety alarm module. The proposed system is supposed to apply in a normal highway driving scenario. In the car detection module, the Google Tensorflow Object Detection (GTOD) API is employed. The function of GTOD API is to detect frontal cars in real-time and then mark them with rectangular boxes. As for the safety alarm module, it consists of three phases: to calculate the box width of detected cars; to calculate the safety factor; to determine the driving state. To justify the proposed system, a real highway experiment is conducted. The results show that the proposed system is able to appropriately indicate driving states: safe, dangerous and warning. By the given experimental results, it implies that the proposed system is feasible and applicable in the real-world applications.
{"title":"Real-Time Car Detection and Driving Safety Alarm System With Google Tensorflow Object Detection API","authors":"C. Hsieh, Dung-Ching Lin, Chengjia Wang, Zong-Ting Chen, Jiun-Jian Liaw","doi":"10.1109/ICMLC48188.2019.8949265","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949265","url":null,"abstract":"Car accident is a serious social problem which often results in both life loss and financial loss. Most of car accidents are caused by a lack of safe distance between cars. To relieve this problem, in this paper we propose a real-time car detection and safety alarm system. The proposed system consists of two modules: real-time car detection module and safety alarm module. The proposed system is supposed to apply in a normal highway driving scenario. In the car detection module, the Google Tensorflow Object Detection (GTOD) API is employed. The function of GTOD API is to detect frontal cars in real-time and then mark them with rectangular boxes. As for the safety alarm module, it consists of three phases: to calculate the box width of detected cars; to calculate the safety factor; to determine the driving state. To justify the proposed system, a real highway experiment is conducted. The results show that the proposed system is able to appropriately indicate driving states: safe, dangerous and warning. By the given experimental results, it implies that the proposed system is feasible and applicable in the real-world applications.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129726629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949270
Hengdong Yang, Shiang-Lin Lin, Jui-Yen Chang
Location-based advertising (LBA) is a mobile phone service to apply customers' geographic location to provide suitable advertisement. It was proved in literature to be effective to motivate purchasing intention. However, perceived benefits would be accompanied by perceived risks of privacy concerns to users. The precondition of receiving the LBA is to turn on the location functions in the mobile device. This study applies the Fuzzy Analytic Hierarchy Process (FAHP) method to analyze the factors evaluated by consumers while considering to turn on the GPS functions for receiving LBA. The analytic results reported that “functional value”, “privacy considerations” and “inertia and usage of other 3C habits” are the top three important decision dimensions. In terms of decision factors, the most top three important evaluation factors are “habit of using 3C device”, “getting money-saving opportunities” and “whereabouts”. The findings are useful not only for LBA providers to design and manage their advertising practices but also for consumers to understand the critical factors while considering to turn on the GPS functions for receiving LBA.
基于位置的广告(location based advertising, LBA)是一种利用手机用户的地理位置提供合适广告的服务。文献证明，这对激发购买意愿是有效的。然而，感知到的好处将伴随着用户隐私担忧的感知风险。接收LBA的前提是打开移动设备的定位功能。本研究采用模糊层次分析法(FAHP)对消费者在考虑开启GPS接收LBA功能时所评价的因素进行分析。分析结果显示，“功能价值”、“隐私考虑”和“惯性和其他3C习惯的使用”是最重要的三个决策维度。在决策因素方面，排名前三位的重要评价因素分别是“使用3C设备的习惯”、“获得省钱机会”和“去向”。研究结果不仅有助于LBA提供商设计和管理他们的广告实践，也有助于消费者在考虑打开GPS功能以接收LBA时了解关键因素。
{"title":"Would you Turn-On GPS for LBA? Fuzzy AHP Approach","authors":"Hengdong Yang, Shiang-Lin Lin, Jui-Yen Chang","doi":"10.1109/ICMLC48188.2019.8949270","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949270","url":null,"abstract":"Location-based advertising (LBA) is a mobile phone service to apply customers' geographic location to provide suitable advertisement. It was proved in literature to be effective to motivate purchasing intention. However, perceived benefits would be accompanied by perceived risks of privacy concerns to users. The precondition of receiving the LBA is to turn on the location functions in the mobile device. This study applies the Fuzzy Analytic Hierarchy Process (FAHP) method to analyze the factors evaluated by consumers while considering to turn on the GPS functions for receiving LBA. The analytic results reported that “functional value”, “privacy considerations” and “inertia and usage of other 3C habits” are the top three important decision dimensions. In terms of decision factors, the most top three important evaluation factors are “habit of using 3C device”, “getting money-saving opportunities” and “whereabouts”. The findings are useful not only for LBA providers to design and manage their advertising practices but also for consumers to understand the critical factors while considering to turn on the GPS functions for receiving LBA.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133182872","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949192
C. Teng, Ben-Jian Dong
Image feature matching is a very important and fundamental task in computer vision. In this paper, a spatial-order based progressive feature matching framework is proposed. With the model of spatial order, the searching space is partitioned into many intervals with each interval associated with a probability that a correct match is occurred in this interval. Using this information, many incorrect features could be filtered out and only the survived features are passed for subsequent matching. As the features are progressively matched, the model of spatial order is also progressively updated and the lengths of partitioned intervals are further shortened to filter out more features. To demonstrate the feasibility of proposed system, a series of experiments were conducted. A standard benchmark image data set was used to test the proposed system and the results showed that the proposed framework can indeed produce more efficient and accurate feature matching compared with traditional brute force technique.
{"title":"Using Feature Spatial Order in Progressive Image Feature Matching","authors":"C. Teng, Ben-Jian Dong","doi":"10.1109/ICMLC48188.2019.8949192","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949192","url":null,"abstract":"Image feature matching is a very important and fundamental task in computer vision. In this paper, a spatial-order based progressive feature matching framework is proposed. With the model of spatial order, the searching space is partitioned into many intervals with each interval associated with a probability that a correct match is occurred in this interval. Using this information, many incorrect features could be filtered out and only the survived features are passed for subsequent matching. As the features are progressively matched, the model of spatial order is also progressively updated and the lengths of partitioned intervals are further shortened to filter out more features. To demonstrate the feasibility of proposed system, a series of experiments were conducted. A standard benchmark image data set was used to test the proposed system and the results showed that the proposed framework can indeed produce more efficient and accurate feature matching compared with traditional brute force technique.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"24 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123183934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949166
Cheng-Bin Wang, Fachao Li
Comprehensive quality evaluation is the measure of the members' comprehensive ability and the premise and foundation of improving the team's operation efficiency. How to accurately obtain the comprehensive quality of members has been a widely concerned issue in the academic and application fields. Taking cooperative performance as the main observation index, this paper proposes a comprehensive quality evaluation model based on cooperative performance, and analyzes the characteristics and shortcomings of this model. In order to solve the problem that the solution cannot be guaranteed, the method of multi objective programming is applied to give the solution strategy based on the deviation variable. Finally, the feasibility and effectiveness of the model are analyzed with a case study. Theoretical analysis and example calculation show that the model has good interpretability and operability, which not only improves the existing evaluation methods to a certain extent, but also has wide application value in the fields of resource allocation, artificial intelligence and recommendation system.
{"title":"Research on Comprehensive Quality Evaluation Method Based on Cooperative Performance","authors":"Cheng-Bin Wang, Fachao Li","doi":"10.1109/ICMLC48188.2019.8949166","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949166","url":null,"abstract":"Comprehensive quality evaluation is the measure of the members' comprehensive ability and the premise and foundation of improving the team's operation efficiency. How to accurately obtain the comprehensive quality of members has been a widely concerned issue in the academic and application fields. Taking cooperative performance as the main observation index, this paper proposes a comprehensive quality evaluation model based on cooperative performance, and analyzes the characteristics and shortcomings of this model. In order to solve the problem that the solution cannot be guaranteed, the method of multi objective programming is applied to give the solution strategy based on the deviation variable. Finally, the feasibility and effectiveness of the model are analyzed with a case study. Theoretical analysis and example calculation show that the model has good interpretability and operability, which not only improves the existing evaluation methods to a certain extent, but also has wide application value in the fields of resource allocation, artificial intelligence and recommendation system.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"148 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122282117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949312
Yuya Kinishi, T. Maekawa, S. Mizuta, T. Ishikawa, Y. Hata
This paper aims to determine the optimal puncture position of ovum by evaluating rupture membrane of cytoplasm. We employed 139 ovum images on the Piezo-ICSI (Intracytoplasmic sperm injection). In it, grayscale images before puncture and their actual puncture position were obtained from the movie file (Rupture:31, No Rupture:108), and Local Binary Pattern (LBP) feature is calculated at analysis area around the puncture position. LBP feature dimensions are reduced, and data are classified by hierarchical clustering method using feature of three dimensions. As a result, the data classified into two clusters (Clusters A and B). Cluster A has 7 Ruptures and 50 No Ruptures, Cluster B has 24 Ruptures and 58 No Ruptures. Then, the sensitivity is 0.77. Therefore, it is possible to evaluate rupture membrane of cytoplasm from shape feature of membrane. The optimal puncture position could be determined by the features.
{"title":"Detection of Optimal Puncture Position in OVUM Images for Artificial Insemination","authors":"Yuya Kinishi, T. Maekawa, S. Mizuta, T. Ishikawa, Y. Hata","doi":"10.1109/ICMLC48188.2019.8949312","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949312","url":null,"abstract":"This paper aims to determine the optimal puncture position of ovum by evaluating rupture membrane of cytoplasm. We employed 139 ovum images on the Piezo-ICSI (Intracytoplasmic sperm injection). In it, grayscale images before puncture and their actual puncture position were obtained from the movie file (Rupture:31, No Rupture:108), and Local Binary Pattern (LBP) feature is calculated at analysis area around the puncture position. LBP feature dimensions are reduced, and data are classified by hierarchical clustering method using feature of three dimensions. As a result, the data classified into two clusters (Clusters A and B). Cluster A has 7 Ruptures and 50 No Ruptures, Cluster B has 24 Ruptures and 58 No Ruptures. Then, the sensitivity is 0.77. Therefore, it is possible to evaluate rupture membrane of cytoplasm from shape feature of membrane. The optimal puncture position could be determined by the features.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115437624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949254
Guihai Li, Gang Liu, Yu-Xuan Li, Song-Lin Chen
The endurance ability is an important factor to be considered in the practical application of bionic robotic fish. By designing an optimal energy consumption control method, the energy consumption of robotic fish can be effectively reduced. In this paper, the structural characteristics of the tail fin bionic robotic fish are abstracted through the motion analysis of the tail fin fish. On this basis, a simplified dynamic and kinematic model of the robotic fish and a calculation method of energy consumption are established. Then, by changing the oscillation amplitude and frequency of the tail, the change law of the swimming speed is obtained. It is also found that the energy consumption is positively correlated with the swimming speed in general. In order to get the lowest energy consumption swimming mode of robotic fish at different swimming speeds, a series of optimal energy consumption points are obtained at the interval of 0.05m/s. The control method of optimal energy consumption of robotic fish is designed by analyzing its distribution law.
{"title":"On Optimal Energy Consumption Control Method for Tail Fin Bionic Robotic Fish","authors":"Guihai Li, Gang Liu, Yu-Xuan Li, Song-Lin Chen","doi":"10.1109/ICMLC48188.2019.8949254","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949254","url":null,"abstract":"The endurance ability is an important factor to be considered in the practical application of bionic robotic fish. By designing an optimal energy consumption control method, the energy consumption of robotic fish can be effectively reduced. In this paper, the structural characteristics of the tail fin bionic robotic fish are abstracted through the motion analysis of the tail fin fish. On this basis, a simplified dynamic and kinematic model of the robotic fish and a calculation method of energy consumption are established. Then, by changing the oscillation amplitude and frequency of the tail, the change law of the swimming speed is obtained. It is also found that the energy consumption is positively correlated with the swimming speed in general. In order to get the lowest energy consumption swimming mode of robotic fish at different swimming speeds, a series of optimal energy consumption points are obtained at the interval of 0.05m/s. The control method of optimal energy consumption of robotic fish is designed by analyzing its distribution law.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127878718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949315
Wei-Ho Tsai, Van-Thuan Tran, Shiang-Shiun Kung
In this study, we propose an automatic system for detecting mispronounced lyrics in singing, thereby providing information for singing performance assessment. The system is built upon the basis of speech utterance verification and further improved by considering the difference between singing and speech. We recognize that the vowels are often lengthened during singing and thus include a duration modeling concept in the acoustic modeling to absorb the variation of the length of a vowel in singing. Our experiments show that the proposed methods can achieve 11.3% equal error rate in detecting the mispronounced lyrics in singing.
{"title":"Automatic Detection of Mispronounced Lyrics in Singing","authors":"Wei-Ho Tsai, Van-Thuan Tran, Shiang-Shiun Kung","doi":"10.1109/ICMLC48188.2019.8949315","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949315","url":null,"abstract":"In this study, we propose an automatic system for detecting mispronounced lyrics in singing, thereby providing information for singing performance assessment. The system is built upon the basis of speech utterance verification and further improved by considering the difference between singing and speech. We recognize that the vowels are often lengthened during singing and thus include a duration modeling concept in the acoustic modeling to absorb the variation of the length of a vowel in singing. Our experiments show that the proposed methods can achieve 11.3% equal error rate in detecting the mispronounced lyrics in singing.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":"147 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125870879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：