Jake Gonzalez, Chau Pham, Afamefuna Umejiaku, Juanita Benjamin, Tommy Dang
This paper proposes SkeletonVR, a VR puzzle assembly application, to facilitate the education of the human skeletal system. With the use of motion-tracked controllers, the application allows users to grab and assemble bones within the virtual environment to learn about the location and orientation of the different bones within a human skeleton while also educating them on the names of the bones being interacted with. We aim to bring a new experience to users by providing an interactive and immersive environment that makes learning more intriguing while also providing different difficulty modes to keep users engaged and challenged. We further discuss students’ feedback in a VR class to identify the limitations of our approach and evaluate its usefulness. While aiming at human anatomy, our interactive puzzle assembly application can be extended to other research and application domains such as chemical compound structure assembly and structure-based drug design.
{"title":"SkeletonVR: Educating Human Anatomy Through An Interactive Puzzle Assembly","authors":"Jake Gonzalez, Chau Pham, Afamefuna Umejiaku, Juanita Benjamin, Tommy Dang","doi":"10.1145/3468784.3471605","DOIUrl":"https://doi.org/10.1145/3468784.3471605","url":null,"abstract":"This paper proposes SkeletonVR, a VR puzzle assembly application, to facilitate the education of the human skeletal system. With the use of motion-tracked controllers, the application allows users to grab and assemble bones within the virtual environment to learn about the location and orientation of the different bones within a human skeleton while also educating them on the names of the bones being interacted with. We aim to bring a new experience to users by providing an interactive and immersive environment that makes learning more intriguing while also providing different difficulty modes to keep users engaged and challenged. We further discuss students’ feedback in a VR class to identify the limitations of our approach and evaluate its usefulness. While aiming at human anatomy, our interactive puzzle assembly application can be extended to other research and application domains such as chemical compound structure assembly and structure-based drug design.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127796310","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}
Skin diseases are becoming the most prevalent health concern among all nations worldwide. Recognition of skin lesion, abnormal change usually caused by disease or infection in the skin is the first step in diagnosing skin diseases. In dermatology, morphology skin lesions occur due to the disease process's direct result and indicate categorizing a skin lesions' structure and appearance. In this work, we focus on primary skin lesion classification, particularly early-stage detection, and present a deep learning approach to classify images containing skin lesions, macule, nodule, papule, plaque pustule, wheal, and bulla. We applied deep learning techniques for classifying such images into seven classes covering the aforementioned types of lesion. In particular, we performed experiments on pre-trained deep convolutional neural network models to find the most accuracy one. The result shows that the pre-trained model ResNet-50 after the training and testing can achieve satisfactory accuracy of 85.95%.
{"title":"Multi-Class Primary Morphology Lesions Classification Using Deep Convolutional Neural Network","authors":"Naqibullah Vakili, Worarat Krathu, Nongnuch Laomaneerattanaporn","doi":"10.1145/3468784.3468887","DOIUrl":"https://doi.org/10.1145/3468784.3468887","url":null,"abstract":"Skin diseases are becoming the most prevalent health concern among all nations worldwide. Recognition of skin lesion, abnormal change usually caused by disease or infection in the skin is the first step in diagnosing skin diseases. In dermatology, morphology skin lesions occur due to the disease process's direct result and indicate categorizing a skin lesions' structure and appearance. In this work, we focus on primary skin lesion classification, particularly early-stage detection, and present a deep learning approach to classify images containing skin lesions, macule, nodule, papule, plaque pustule, wheal, and bulla. We applied deep learning techniques for classifying such images into seven classes covering the aforementioned types of lesion. In particular, we performed experiments on pre-trained deep convolutional neural network models to find the most accuracy one. The result shows that the pre-trained model ResNet-50 after the training and testing can achieve satisfactory accuracy of 85.95%.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128835624","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}
Tashfiq Rahman, Rohani Rohan, Debajyoti Pal, P. Kanthamanon
Humans are often considered to be the weakest link in the cybersecurity chain. However, traditionally the Computer Science (CS) researchers have investigated the technical aspects of cybersecurity, focusing on the encryption and network security mechanisms. The human aspect although very important is often neglected. In this work we carry out a scoping review to investigate the take of the CS community on the human-centric cybersecurity paradigm by considering the top conferences on network and computer security for the past six years. Results show that broadly two types of users are considered: expert and non-expert users. Qualitative techniques dominate the research methodology employed, however, there is a lack of focus on the theoretical aspects. Moreover, the samples have a heavy bias towards the Western community, due to which the results cannot be generalized, and the effect of culture on cybersecurity is a lesser known aspect. Another issue is with respect to the unavailability of standardized security-specific scales that can measure the cybersecurity perception of the users. New insights are obtained and avenues for future research are presented.
{"title":"Human Factors in Cybersecurity: A Scoping Review","authors":"Tashfiq Rahman, Rohani Rohan, Debajyoti Pal, P. Kanthamanon","doi":"10.1145/3468784.3468789","DOIUrl":"https://doi.org/10.1145/3468784.3468789","url":null,"abstract":"Humans are often considered to be the weakest link in the cybersecurity chain. However, traditionally the Computer Science (CS) researchers have investigated the technical aspects of cybersecurity, focusing on the encryption and network security mechanisms. The human aspect although very important is often neglected. In this work we carry out a scoping review to investigate the take of the CS community on the human-centric cybersecurity paradigm by considering the top conferences on network and computer security for the past six years. Results show that broadly two types of users are considered: expert and non-expert users. Qualitative techniques dominate the research methodology employed, however, there is a lack of focus on the theoretical aspects. Moreover, the samples have a heavy bias towards the Western community, due to which the results cannot be generalized, and the effect of culture on cybersecurity is a lesser known aspect. Another issue is with respect to the unavailability of standardized security-specific scales that can measure the cybersecurity perception of the users. New insights are obtained and avenues for future research are presented.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128456099","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}
R. Kannan, Sridhar Swaminathan, Chutiporn Anutariya, V. Saravanan
Social media is more and more utilized by people around the world to express their feelings and opinions in the kind of short text messages. Twitter has been a rapidly growing microblogging social networking website where people express their opinions in a precise and simple manner of expressions. It has also become a platform for governmental, non-governmental and individual opinions and policy announcements. Detecting sentiments from tweets has a wide range of applications including identifying the anxiety or depression of individuals and measuring the well-being or mood of a community. In addition, the sentiment classification becomes complex when the tweet is written in codemix language which is a mix of two different languages. The main objective of this paper is to classify tweets written in mix of Tamil and English language into positive, negative, or neutral. This is achieved by fine tuning a pretrained multilingual text representation model as well as deep learning transformers. The proposed approach is experimented with large scale of tweets collected for societal issues in India. We also provide a comparative study of different machine learning and deep learning models. The proposed architecture based on neural linguistic representation provides significant accuracy in classifying both Tamil and codemix tweets.
{"title":"Exploiting Multilingual Neural Linguistic Representation for Sentiment Classification of Political Tweets in Code-mix Language","authors":"R. Kannan, Sridhar Swaminathan, Chutiporn Anutariya, V. Saravanan","doi":"10.1145/3468784.3470787","DOIUrl":"https://doi.org/10.1145/3468784.3470787","url":null,"abstract":"Social media is more and more utilized by people around the world to express their feelings and opinions in the kind of short text messages. Twitter has been a rapidly growing microblogging social networking website where people express their opinions in a precise and simple manner of expressions. It has also become a platform for governmental, non-governmental and individual opinions and policy announcements. Detecting sentiments from tweets has a wide range of applications including identifying the anxiety or depression of individuals and measuring the well-being or mood of a community. In addition, the sentiment classification becomes complex when the tweet is written in codemix language which is a mix of two different languages. The main objective of this paper is to classify tweets written in mix of Tamil and English language into positive, negative, or neutral. This is achieved by fine tuning a pretrained multilingual text representation model as well as deep learning transformers. The proposed approach is experimented with large scale of tweets collected for societal issues in India. We also provide a comparative study of different machine learning and deep learning models. The proposed architecture based on neural linguistic representation provides significant accuracy in classifying both Tamil and codemix tweets.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"165 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116047820","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}
AI-driven automation is the game-changer in this decade. The one concept that belongs to this domain is to simulate human working processes by using machine learning. An adaptation of this knowledge in web development is popularized topic in the web developer society. Moreover, Web Components, the new paradigm in software engineering practices in web development, becomes the new standard defined by World Wide Web Consortium (W3C). It is an essential building block for modularizing large and complex web applications into smaller pieces and then presenting them via the web browser on the user's computer or mobile. We combine knowledge between Computer Vision (CV) with deep learning and Web Components developer framework together to train the machine to recognize bounding boxes and category labels for each object of interest in an image. This paper introduces the methodology to automatically generate a website by neuron network model composite with many small web components. Our work's best result has a validation loss of 1.873, which can recognize the web object and transform it into the Web Components Template by React web framework.
{"title":"Web Components Template Generation from Web Screenshot","authors":"Pattana Anunphop, P. Chongstitvatana","doi":"10.1145/3468784.3468787","DOIUrl":"https://doi.org/10.1145/3468784.3468787","url":null,"abstract":"AI-driven automation is the game-changer in this decade. The one concept that belongs to this domain is to simulate human working processes by using machine learning. An adaptation of this knowledge in web development is popularized topic in the web developer society. Moreover, Web Components, the new paradigm in software engineering practices in web development, becomes the new standard defined by World Wide Web Consortium (W3C). It is an essential building block for modularizing large and complex web applications into smaller pieces and then presenting them via the web browser on the user's computer or mobile. We combine knowledge between Computer Vision (CV) with deep learning and Web Components developer framework together to train the machine to recognize bounding boxes and category labels for each object of interest in an image. This paper introduces the methodology to automatically generate a website by neuron network model composite with many small web components. Our work's best result has a validation loss of 1.873, which can recognize the web object and transform it into the Web Components Template by React web framework.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126247677","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}
Abnash Bassi, Anika Shenoy, Arjun Sharma, Hanna Sigurdson, Connor Glossop, Jonathan H. Chan
Abstract: Building energy consumption forecasting is essential for improving the sustainability of buildings in the context of addressing climate change. Accurate building load predictions are useful for energy efficient building design selection and demand-side management initiatives. Using historical building energy consumption data has allowed researchers to develop machine learning models to improve the accuracy of such predictions, beyond inefficient traditional approaches otherwise used by the building sector. This work examines gradient boosting machine learning models, namely LightGBM, CatBoost, and XGBoost, for the purpose of comparing their performance on a select dataset. These gradient boosting models are popular in Kaggle machine learning contest solutions but have not been compared formally for the application of building energy consumption predictions. This work applies the three gradient boosting algorithms to a synthesized dataset for a large office building in Chicago. Preliminary results from the presented comparison demonstrate that XGBoost performs better than LightGBM and CatBoost when trained on the selected dataset.
{"title":"Building Energy Consumption Forecasting: A Comparison of Gradient Boosting Models","authors":"Abnash Bassi, Anika Shenoy, Arjun Sharma, Hanna Sigurdson, Connor Glossop, Jonathan H. Chan","doi":"10.1145/3468784.3470656","DOIUrl":"https://doi.org/10.1145/3468784.3470656","url":null,"abstract":"Abstract: Building energy consumption forecasting is essential for improving the sustainability of buildings in the context of addressing climate change. Accurate building load predictions are useful for energy efficient building design selection and demand-side management initiatives. Using historical building energy consumption data has allowed researchers to develop machine learning models to improve the accuracy of such predictions, beyond inefficient traditional approaches otherwise used by the building sector. This work examines gradient boosting machine learning models, namely LightGBM, CatBoost, and XGBoost, for the purpose of comparing their performance on a select dataset. These gradient boosting models are popular in Kaggle machine learning contest solutions but have not been compared formally for the application of building energy consumption predictions. This work applies the three gradient boosting algorithms to a synthesized dataset for a large office building in Chicago. Preliminary results from the presented comparison demonstrate that XGBoost performs better than LightGBM and CatBoost when trained on the selected dataset.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"272 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131691590","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}
Huyen N. Nguyen, Caleb M. Trujillo, Kevin Wee, Kathleen A. Bowe
Data visualization accelerates the communication of quantitative measures across many fields, including education, but few visualization methods exist for qualitative data in educational fields that capture both the context-specific information and summarize trends for instructors. In this paper, we design an interface to visualize students’ weekly journal entries collected as formative educational assessments from an undergraduate data visualization course and a statistics course. Using these qualitative data, we present an interactive WordStream and word cloud to show the temporal and topic-based organization of students’ development during instruction and explore the patterns, trends, and diversity of student ideas in a context-specific way. Informed by the Technology Acceptance Model, we used an informal user study to evaluate the perceived ease of use and usefulness of the tool for instructors using journal entries. Our evaluation found the tool to be intuitive, clear, and easy-to-use to explore student entries, especially words of interest, but might be limited by focusing on word frequencies rather than underlying relationships among the student’s ideas or other measures in assessment. Implications and challenges for bridging qualitative data for educational assessment with data visualization methods are discussed.
{"title":"Interactive Qualitative Data Visualization for Educational Assessment","authors":"Huyen N. Nguyen, Caleb M. Trujillo, Kevin Wee, Kathleen A. Bowe","doi":"10.1145/3468784.3469851","DOIUrl":"https://doi.org/10.1145/3468784.3469851","url":null,"abstract":"Data visualization accelerates the communication of quantitative measures across many fields, including education, but few visualization methods exist for qualitative data in educational fields that capture both the context-specific information and summarize trends for instructors. In this paper, we design an interface to visualize students’ weekly journal entries collected as formative educational assessments from an undergraduate data visualization course and a statistics course. Using these qualitative data, we present an interactive WordStream and word cloud to show the temporal and topic-based organization of students’ development during instruction and explore the patterns, trends, and diversity of student ideas in a context-specific way. Informed by the Technology Acceptance Model, we used an informal user study to evaluate the perceived ease of use and usefulness of the tool for instructors using journal entries. Our evaluation found the tool to be intuitive, clear, and easy-to-use to explore student entries, especially words of interest, but might be limited by focusing on word frequencies rather than underlying relationships among the student’s ideas or other measures in assessment. Implications and challenges for bridging qualitative data for educational assessment with data visualization methods are discussed.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133075477","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}
There are many applications for creating an indoor map by a single robot already. When using a single robot in a large working space like a factory, the performance and robustness are needed to be increased. Multi-Agent Robot System (MAR) is introduced to meet this requirement. MAR could increase productivity and flexibility while works in a dynamic environment because it is modular and can work simultaneously. When MAR combines with Simultaneous Localization and Mapping (SLAM) technology, it can explore and discover the indoor environment cooperatively and simultaneously. Each robot creates its map with different initial poses and path planning. The main issue of a Multi-Robot SLAM (MRSLAM) is how to combine maps from different robots correctly. In this research, we will focus on algorithms of map merging. SIFT and ORB descriptors are selected along with some image processing techniques, and a proposed approach including the algorithms is verified by general benchmark map data. The results will be shown and discussed. Then, the proposed approach will be deployed into a real robot platform based on Robot Operating System (ROS). Experiments will be conducted to prove the feasibility and the limitation of the proposed approach in the real-world scenario.
已经有很多应用程序可以通过单个机器人创建室内地图。当在工厂等大型工作空间中使用单个机器人时，需要提高其性能和鲁棒性。多智能体机器人系统(Multi-Agent Robot System, MAR)就是为了满足这一需求而引入的。MAR可以在动态环境中工作时提高生产力和灵活性，因为它是模块化的，可以同时工作。当MAR与SLAM (Simultaneous Localization and Mapping)技术相结合时，可以协同、同步地对室内环境进行探索和发现。每个机器人用不同的初始姿态和路径规划创建自己的地图。多机器人SLAM (MRSLAM)的主要问题是如何正确地组合来自不同机器人的地图。在本研究中，我们将重点研究地图合并算法。选择SIFT和ORB描述符以及一些图像处理技术，并通过一般基准地图数据验证了包含这些算法的方法。结果将被展示和讨论。然后，将该方法部署到基于机器人操作系统(ROS)的真实机器人平台中。将进行实验来证明在现实世界场景中提出的方法的可行性和局限性。
{"title":"Investigation of SIFT and ORB descriptors for Indoor Maps Fusion for the Multi-agent mobile robots","authors":"Ming-Hsien Chuang, K. Sukvichai","doi":"10.1145/3468784.3469950","DOIUrl":"https://doi.org/10.1145/3468784.3469950","url":null,"abstract":"There are many applications for creating an indoor map by a single robot already. When using a single robot in a large working space like a factory, the performance and robustness are needed to be increased. Multi-Agent Robot System (MAR) is introduced to meet this requirement. MAR could increase productivity and flexibility while works in a dynamic environment because it is modular and can work simultaneously. When MAR combines with Simultaneous Localization and Mapping (SLAM) technology, it can explore and discover the indoor environment cooperatively and simultaneously. Each robot creates its map with different initial poses and path planning. The main issue of a Multi-Robot SLAM (MRSLAM) is how to combine maps from different robots correctly. In this research, we will focus on algorithms of map merging. SIFT and ORB descriptors are selected along with some image processing techniques, and a proposed approach including the algorithms is verified by general benchmark map data. The results will be shown and discussed. Then, the proposed approach will be deployed into a real robot platform based on Robot Operating System (ROS). Experiments will be conducted to prove the feasibility and the limitation of the proposed approach in the real-world scenario.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115126199","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}
Neural networks are known for their predictive capability, leading to vast applications in various domains. However, the explainability of a neural network model remains enigmatic, especially when the model comes short in learning a particular pattern or features. This work introduces a visual explainable LSTM network framework focusing on temporal prediction. The hindrance to the training process is highlighted by the irregular instances throughout the entire architecture, from input to intermediate layers and output. The framework provides interactive features to support users in customizing and rearranging the structure to obtain different network representations and perform what-if analysis. To evaluate the usefulness of our approach, we demonstrate the application of VixLSTM on the various datasets generated from different domains.
{"title":"VixLSTM: Visual Explainable LSTM for Multivariate Time Series","authors":"Tommy Dang, Huyen N. Nguyen, Ngan V. T. Nguyen","doi":"10.1145/3468784.3471603","DOIUrl":"https://doi.org/10.1145/3468784.3471603","url":null,"abstract":"Neural networks are known for their predictive capability, leading to vast applications in various domains. However, the explainability of a neural network model remains enigmatic, especially when the model comes short in learning a particular pattern or features. This work introduces a visual explainable LSTM network framework focusing on temporal prediction. The hindrance to the training process is highlighted by the irregular instances throughout the entire architecture, from input to intermediate layers and output. The framework provides interactive features to support users in customizing and rearranging the structure to obtain different network representations and perform what-if analysis. To evaluate the usefulness of our approach, we demonstrate the application of VixLSTM on the various datasets generated from different domains.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"05 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124458810","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}
The number of streaming service providers has been increasing dramatically every year. Hence, users may prefer to publish their stream to multiple service endpoints simultaneously to increase visibility. However, most service providers prefer to monopolize their services. Hence, a study of a suitable architectural design of a streaming service that supports multiple streaming endpoints has not gained lots of attention. In this study, the effect of adopting different architectural design on developing a live streaming service over LTE network which can supports multiple streaming endpoints are investigated. Two major designs are selected which are a selective forwarding unit based architecture, and a non-selective forwarding unit based architecture. The results suggest that a selective forwarding unit architecture has an advantage over a non-selective forwarding unit based architecture on keeping overall average streaming end-to-end delay to be minimum., while a fluctuation in an end-to-end delay occurs in a non-selective forwarding unit based architecture in our experiment testbed. The results, discussions, and suggestions on future studies are given at the end of this study.
{"title":"A Study of Effect of Architectural Design on Quality of Service of a Live Streaming Application with Multiple Endpoints over LTE Network","authors":"Charlif Prapawit, C. Angsuchotmetee","doi":"10.1145/3468784.3469855","DOIUrl":"https://doi.org/10.1145/3468784.3469855","url":null,"abstract":"The number of streaming service providers has been increasing dramatically every year. Hence, users may prefer to publish their stream to multiple service endpoints simultaneously to increase visibility. However, most service providers prefer to monopolize their services. Hence, a study of a suitable architectural design of a streaming service that supports multiple streaming endpoints has not gained lots of attention. In this study, the effect of adopting different architectural design on developing a live streaming service over LTE network which can supports multiple streaming endpoints are investigated. Two major designs are selected which are a selective forwarding unit based architecture, and a non-selective forwarding unit based architecture. The results suggest that a selective forwarding unit architecture has an advantage over a non-selective forwarding unit based architecture on keeping overall average streaming end-to-end delay to be minimum., while a fluctuation in an end-to-end delay occurs in a non-selective forwarding unit based architecture in our experiment testbed. The results, discussions, and suggestions on future studies are given at the end of this study.","PeriodicalId":341589,"journal":{"name":"The 12th International Conference on Advances in Information Technology","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134254831","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：