Pub Date : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00170
Mikio Yamamoto, Akiko Sakamoto, Nahoko Okada
Real-time captioning is one of the effective support for ensuring the accessibility of lectures for students with hearing disabilities. Speech recognition technology is expected as a technology for automating real-time captioning support in higher education. However, the recognition rate of Japanese is often insufficient for hearing-impaired students to accurately understand the content of lectures in real time, and it is necessary to correct the misrecognition results. For this reason, the mainstream of real-time captioning support for hearing-impaired students in Japanese higher education is so-called PC note-taking, in which the supporter summarizes the content of the lecture and inputs it to a personal computer. We have classified the automation level by using speech recognition technology into four stages, and by optimizing the automation level for each class, we are gradually promoting remote automation of real-time captioning support at Hiroshima University from 2019. In this paper, we report on the practice of step-by-step automation of real-time captioning support at Hiroshima University, and discuss the advantages of the remote summary dictation we have adopted, including cost-effectiveness.
{"title":"Practice and automation of remote real-time captioning support for students with disabilities using speech recognition technology","authors":"Mikio Yamamoto, Akiko Sakamoto, Nahoko Okada","doi":"10.1109/iiai-aai53430.2021.00170","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00170","url":null,"abstract":"Real-time captioning is one of the effective support for ensuring the accessibility of lectures for students with hearing disabilities. Speech recognition technology is expected as a technology for automating real-time captioning support in higher education. However, the recognition rate of Japanese is often insufficient for hearing-impaired students to accurately understand the content of lectures in real time, and it is necessary to correct the misrecognition results. For this reason, the mainstream of real-time captioning support for hearing-impaired students in Japanese higher education is so-called PC note-taking, in which the supporter summarizes the content of the lecture and inputs it to a personal computer. We have classified the automation level by using speech recognition technology into four stages, and by optimizing the automation level for each class, we are gradually promoting remote automation of real-time captioning support at Hiroshima University from 2019. In this paper, we report on the practice of step-by-step automation of real-time captioning support at Hiroshima University, and discuss the advantages of the remote summary dictation we have adopted, including cost-effectiveness.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114657132","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00091
Shangyang Mou, Hugang Han
While the affine T-S fuzzy model is much more effective for describing the dynamics of the system concerned in comparison with the regular T-S fuzzy model, it encounters some difficulties when designing a controller based on the model because of the affine term within. However the difficulties can be overcome by introduction of a kind of quadratic inequalities. This paper proposes a way to yield such quadratic inequalities based on the information over the antecedent part of affine T-S fuzzy model rules. Although the introduction of such quadratic inequalities is motivated by controller designs based on the affine T-S fuzzy model, the way of the introduction can be applied to other fuzzy models.
{"title":"Quadratic Inequalities over Affine T-S Fuzzy Models","authors":"Shangyang Mou, Hugang Han","doi":"10.1109/iiai-aai53430.2021.00091","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00091","url":null,"abstract":"While the affine T-S fuzzy model is much more effective for describing the dynamics of the system concerned in comparison with the regular T-S fuzzy model, it encounters some difficulties when designing a controller based on the model because of the affine term within. However the difficulties can be overcome by introduction of a kind of quadratic inequalities. This paper proposes a way to yield such quadratic inequalities based on the information over the antecedent part of affine T-S fuzzy model rules. Although the introduction of such quadratic inequalities is motivated by controller designs based on the affine T-S fuzzy model, the way of the introduction can be applied to other fuzzy models.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121096530","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00025
Masateru Kishikawa, T. Kakeshita
We have proposed a programming education support tool pgtracer for effective programming education. Pgtracer utilizes source codes and trace tables that represent execution results of the source code and creates programming exercises by defining blanks in some of them. The difficulty level of the exercises can be freely adjusted since the fill-in-the-blanks in the programs and trace tables can be defined freely. Furthermore, we can realize learning analytics by collecting the data of students' response process. In this paper, we extend pgtracer to incorporate Java for the effective education of object-oriented programming. We convert Java programs into an XML format that can be processed by pgtracer to effectively teach object-oriented programming using Java. By expressing the source code in XML format, it becomes possible to create flexible fill-in-the-blank questions. For this purpose, we utilize JavaCC to create an XML generator, which implements a program for each syntactic rule of the Java grammar to convert the source code into the specified XML format. The XML generator is implemented by combining these programs.
{"title":"Converting Java Source Code for Programming Education Support Tool pgtracer","authors":"Masateru Kishikawa, T. Kakeshita","doi":"10.1109/iiai-aai53430.2021.00025","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00025","url":null,"abstract":"We have proposed a programming education support tool pgtracer for effective programming education. Pgtracer utilizes source codes and trace tables that represent execution results of the source code and creates programming exercises by defining blanks in some of them. The difficulty level of the exercises can be freely adjusted since the fill-in-the-blanks in the programs and trace tables can be defined freely. Furthermore, we can realize learning analytics by collecting the data of students' response process. In this paper, we extend pgtracer to incorporate Java for the effective education of object-oriented programming. We convert Java programs into an XML format that can be processed by pgtracer to effectively teach object-oriented programming using Java. By expressing the source code in XML format, it becomes possible to create flexible fill-in-the-blank questions. For this purpose, we utilize JavaCC to create an XML generator, which implements a program for each syntactic rule of the Java grammar to convert the source code into the specified XML format. The XML generator is implemented by combining these programs.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"74 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127292399","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00002
Kyohei Yoshihara, K. Hirata
In this paper, we develop the method to detect Campylobacter bacteria and phagocytotic activity of leukocytes from Gram stained smears images. First, we improve VGG16 by adding batch normalization and by replacing flatten with global average pooling and construct the classifier by using transfer learning model based on the improved VGG16. Then, by comparing the detection by the VGG16 with that by the improved VGG16, for the classification of Camphylobacter bacteria, we give experimental results of classifying Campylobacter images with non-Campylobacter images. On the other hand, for the classification of phagocytotic activity of leukocytes, we give experimental results of classifying phagocytotic images with quasi- and non-phagocytotic images and of classifying phagocytotic images, quasi-phagocytotic images and non-phagocytotic images.
{"title":"Detecting Campylobacter Bacteria and Phagocytotic Activity of Leukocytes from Gram Stained Smears Images","authors":"Kyohei Yoshihara, K. Hirata","doi":"10.1109/iiai-aai53430.2021.00002","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00002","url":null,"abstract":"In this paper, we develop the method to detect Campylobacter bacteria and phagocytotic activity of leukocytes from Gram stained smears images. First, we improve VGG16 by adding batch normalization and by replacing flatten with global average pooling and construct the classifier by using transfer learning model based on the improved VGG16. Then, by comparing the detection by the VGG16 with that by the improved VGG16, for the classification of Camphylobacter bacteria, we give experimental results of classifying Campylobacter images with non-Campylobacter images. On the other hand, for the classification of phagocytotic activity of leukocytes, we give experimental results of classifying phagocytotic images with quasi- and non-phagocytotic images and of classifying phagocytotic images, quasi-phagocytotic images and non-phagocytotic images.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123232238","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00151
Keiya Kadota, Tetsuya Sato, T. Shiina
Supply chain management is a large-scale planning under uncertainty. It is significant that building an efficient supply chain under uncertain conditions is a challenge. There are numerous traditional inventory transshipment models targeting only demands but do not include rentals and returns. This study provides the uncertainty between rentals and returns by scenarios using a multi-period stochastic programming model of inventory transshipment problems. The moment matching method was used to reduce the number of scenarios, and the comparative experiment demonstrated the utility of this model.
{"title":"Solution Algorithm for The Multi-period Inventory Transshipment Problem Considering Rentals and Returns","authors":"Keiya Kadota, Tetsuya Sato, T. Shiina","doi":"10.1109/iiai-aai53430.2021.00151","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00151","url":null,"abstract":"Supply chain management is a large-scale planning under uncertainty. It is significant that building an efficient supply chain under uncertain conditions is a challenge. There are numerous traditional inventory transshipment models targeting only demands but do not include rentals and returns. This study provides the uncertainty between rentals and returns by scenarios using a multi-period stochastic programming model of inventory transshipment problems. The moment matching method was used to reduce the number of scenarios, and the comparative experiment demonstrated the utility of this model.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125550442","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00097
Hidehiko Hayashi, H. Hoshino
Focusing on the development and maintenance of ski resort equipment, this study describes applications using GNSS receivers. We propose the facility management system framework based on the extraction of the process of getting on and off ski lifts equipment. As a measurement model, we designed the interaction between physical environment information and human behavior as a trigger and adopted signifiers of Norman's framework. The feasibility is realized by applying GNSS analysis technology. The novelty of this study is to propose the application of a new interaction design framework that extracts the characteristics of physical environment information and analysis using human location information in the area of facility management systems. The scope of this paper covers chairlift-type equipment among ski resorts. As a result, we found the possibility of monitoring the vibration information given to the skier as the state of the ski lift and applying it to improve the service of the ski resort. This study will increase the momentum for the development of large-scale data collection systems for ski resorts and contribute to the development of ski resort facilities.
{"title":"Facility management system framework for ski resort equipment using GNSS analysis technology","authors":"Hidehiko Hayashi, H. Hoshino","doi":"10.1109/iiai-aai53430.2021.00097","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00097","url":null,"abstract":"Focusing on the development and maintenance of ski resort equipment, this study describes applications using GNSS receivers. We propose the facility management system framework based on the extraction of the process of getting on and off ski lifts equipment. As a measurement model, we designed the interaction between physical environment information and human behavior as a trigger and adopted signifiers of Norman's framework. The feasibility is realized by applying GNSS analysis technology. The novelty of this study is to propose the application of a new interaction design framework that extracts the characteristics of physical environment information and analysis using human location information in the area of facility management systems. The scope of this paper covers chairlift-type equipment among ski resorts. As a result, we found the possibility of monitoring the vibration information given to the skier as the state of the ski lift and applying it to improve the service of the ski resort. This study will increase the momentum for the development of large-scale data collection systems for ski resorts and contribute to the development of ski resort facilities.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"213 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115454260","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00080
Jeongbin Kim, Jiyoon Kim, Hyeonseong Jo, Jason J. Jung, David Camacho
Question answering (QA) system is widely used because it finds the information that users want. Users can ask a variety of questions to the QA system, and the type of question is divided into a simple question and a complex question. A simple question is one query with one question while a complex question is one query with two or more questions. Although many studies have greatly improved the ability of the QA system to answer simple questions, answering complex questions remains a difficult problem. To solve the problem, we use the information from multiple domains and propose a method to quickly retrieve information from many data through a heuristic algorithm. For experiments, we use the movie, person, and country domain data as a knowledge base for the QA system, which is able to help solve the complex question. Compared to the widely used graph search algorithm Breadth-First Search (BFS) and Depth-First Search (DFS), the proposed method reduces the search time by 62% and 44% for finding the answers, respectively.
{"title":"Multiple Domains Knowledge Graph Search via Heuristic Algorithm for Answering Complex Questions","authors":"Jeongbin Kim, Jiyoon Kim, Hyeonseong Jo, Jason J. Jung, David Camacho","doi":"10.1109/iiai-aai53430.2021.00080","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00080","url":null,"abstract":"Question answering (QA) system is widely used because it finds the information that users want. Users can ask a variety of questions to the QA system, and the type of question is divided into a simple question and a complex question. A simple question is one query with one question while a complex question is one query with two or more questions. Although many studies have greatly improved the ability of the QA system to answer simple questions, answering complex questions remains a difficult problem. To solve the problem, we use the information from multiple domains and propose a method to quickly retrieve information from many data through a heuristic algorithm. For experiments, we use the movie, person, and country domain data as a knowledge base for the QA system, which is able to help solve the complex question. Compared to the widely used graph search algorithm Breadth-First Search (BFS) and Depth-First Search (DFS), the proposed method reduces the search time by 62% and 44% for finding the answers, respectively.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116215249","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00051
Satoshi Ozeki, Toru Hayashi, Masa Fukano, Shinichi Yamazaki, Andrea L. Beach, M. D. Sorcinelli
Establishing a management system of teaching and learning has become an essential component for quality assurance in Japanese higher education. An effective system requires several key ingredients, such as clarifying educational goals, organizing curriculums that align with those goals, and assessing learning outcomes. Institutional research for educational data management has been recognized as a key factor in the management system. Although it is imperative to develop teaching skills and capabilities of faculty, who are directly responsible for student learning outcomes, faculty development (FD) receives little attention. This paper aims to discuss how FD should be conducted to further advance Japanese higher education. Specifically, it will present results of a survey conducted to explore future trends in FD. Then, by comparing FD in the United States and Japan, this paper highlights similarities and differences in both countries and suggests future directions for FD in Japanese higher education.
{"title":"Exploring the Future Trends of Faculty Development in Japanese Higher Education","authors":"Satoshi Ozeki, Toru Hayashi, Masa Fukano, Shinichi Yamazaki, Andrea L. Beach, M. D. Sorcinelli","doi":"10.1109/iiai-aai53430.2021.00051","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00051","url":null,"abstract":"Establishing a management system of teaching and learning has become an essential component for quality assurance in Japanese higher education. An effective system requires several key ingredients, such as clarifying educational goals, organizing curriculums that align with those goals, and assessing learning outcomes. Institutional research for educational data management has been recognized as a key factor in the management system. Although it is imperative to develop teaching skills and capabilities of faculty, who are directly responsible for student learning outcomes, faculty development (FD) receives little attention. This paper aims to discuss how FD should be conducted to further advance Japanese higher education. Specifically, it will present results of a survey conducted to explore future trends in FD. Then, by comparing FD in the United States and Japan, this paper highlights similarities and differences in both countries and suggests future directions for FD in Japanese higher education.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"669 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122968497","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}
In the field of online food delivery, which is expanding worldwide, the increasing number of traffic accidents during delivery is becoming problematic. To ensure the safety of delivery workers and residents, it is necessary to understand the incentives of workers for behavior choices. While many existing sharing platforms pull workers into the online labor process by incentives like on-peak/off-peak surcharges, workers try accomplishing more than their goals within a limited time. In this study, to prevent speeding during delivery, we compared and evaluated task allocation methods and incentive schemes using multi-agent simulation. We model workers' rational choices of behaviors based on reinforcement learning, considering the profit and speeding of delivery workers.
{"title":"Incentive Design and Task Allocation for Safe Delivery to Prevent Traffic Accidents","authors":"Takashi Nishino, Sin Syo, Yuka Yamanari, Masaru Miyao, Peng Liu, Hisashi Hayashi","doi":"10.1109/iiai-aai53430.2021.00087","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00087","url":null,"abstract":"In the field of online food delivery, which is expanding worldwide, the increasing number of traffic accidents during delivery is becoming problematic. To ensure the safety of delivery workers and residents, it is necessary to understand the incentives of workers for behavior choices. While many existing sharing platforms pull workers into the online labor process by incentives like on-peak/off-peak surcharges, workers try accomplishing more than their goals within a limited time. In this study, to prevent speeding during delivery, we compared and evaluated task allocation methods and incentive schemes using multi-agent simulation. We model workers' rational choices of behaviors based on reinforcement learning, considering the profit and speeding of delivery workers.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122990502","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 : 2021-07-01DOI: 10.1109/iiai-aai53430.2021.00161
Kazuki Oda, Minoru Sasaki
In this study, we focus on extracting person names from text contained in from TV programs and TV-CMs broadcast in Japan. Person name extraction from TV programs is a challenging task because person names are often recognized with other entity names such as locations and organizations. To tackle this problem, we experiment with an existing named entity extraction method using a list of names extracted from Wikipedia. However, this method results in low precision in extracting the person names, which leads to the problem that the list of person names is not effective for TV program texts. In this paper, to solve this problem, we propose a person name extraction method using Conditional Random Fields (CRF) with ELMo pre-training language model. Also, we propose to use news headlines to construct the person name extraction model for effective extraction of person names performing on the TV programs. As a result of our experiments, the proposed model trained with training data and news headlines provides the highest precision. These results show that adding the news headlines as external information to TV program data is effective for person name extraction.
{"title":"Person Name Extraction from TV program Using Pre-trained Language Model and News Headline","authors":"Kazuki Oda, Minoru Sasaki","doi":"10.1109/iiai-aai53430.2021.00161","DOIUrl":"https://doi.org/10.1109/iiai-aai53430.2021.00161","url":null,"abstract":"In this study, we focus on extracting person names from text contained in from TV programs and TV-CMs broadcast in Japan. Person name extraction from TV programs is a challenging task because person names are often recognized with other entity names such as locations and organizations. To tackle this problem, we experiment with an existing named entity extraction method using a list of names extracted from Wikipedia. However, this method results in low precision in extracting the person names, which leads to the problem that the list of person names is not effective for TV program texts. In this paper, to solve this problem, we propose a person name extraction method using Conditional Random Fields (CRF) with ELMo pre-training language model. Also, we propose to use news headlines to construct the person name extraction model for effective extraction of person names performing on the TV programs. As a result of our experiments, the proposed model trained with training data and news headlines provides the highest precision. These results show that adding the news headlines as external information to TV program data is effective for person name extraction.","PeriodicalId":414070,"journal":{"name":"2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122998457","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：