Yung-Cheng Yao, Xing-Rui Huang, Cheng-Yi Liu, Po-Wei Lin
In traditional fish farms, all work requires a lot of human resources. Whether it is feeding fish, temperature monitoring, etc. These tasks require a lot of manpower. Due to lack of manpower or human negligence, the fish dies, and then cause economic losses to the owners of the fish farms. In this study, two computers are used for image processing. One is to run image processing to detect objects when an object intrudes in the fish farm and use the Arduino to control the pan/tilt motor to rotate the lens and laser to drive away the invading object. Another computer is used as a real-time image server, and the user can use the APP to receive real-time image of fish farm.
{"title":"Intelligent Controlling System in Aquaculture","authors":"Yung-Cheng Yao, Xing-Rui Huang, Cheng-Yi Liu, Po-Wei Lin","doi":"10.1145/3440943.3444351","DOIUrl":"https://doi.org/10.1145/3440943.3444351","url":null,"abstract":"In traditional fish farms, all work requires a lot of human resources. Whether it is feeding fish, temperature monitoring, etc. These tasks require a lot of manpower. Due to lack of manpower or human negligence, the fish dies, and then cause economic losses to the owners of the fish farms. In this study, two computers are used for image processing. One is to run image processing to detect objects when an object intrudes in the fish farm and use the Arduino to control the pan/tilt motor to rotate the lens and laser to drive away the invading object. Another computer is used as a real-time image server, and the user can use the APP to receive real-time image of fish farm.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"100 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120978984","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}
A precise forecasting for the future short-term electricity consumption will be quite useful for making a good plan for the power demand management. Since deep neural network (DNN) provides an effective way for the short-term load forecasting, one of the focuses of this research is thus to use it to construct the prediction model. The gradient-based optimizers (GBOs) have been widely used in DNN algorithms in recent years; however, they make it easy for DNN to trap in poor local regions during the training process. In this research, we propose a new optimizer that is based on the searched experiences to solve this problem to enhance the performance of GBOs. More precisely, the proposed optimizer integrates the best position searched, Lévy flight, and gradient descent to preserve not only the diversification but also the intensification of search during the training process. To evaluate the performance of the proposed optimizer, we compare it with several state-of-the-art GBOs for DNN; namely, Adagrad, RMSprop, and Adam, for training a forecasting model for the short-term electricity consumption forecasting problem. The simulation results show that the proposed optimizer outperforms all the other GBOs in terms of the mean absolute percentage error.
{"title":"An Effective Optimizer based on Global and Local Searched Experiences for Short-term Electricity Consumption Forecasting","authors":"Zhe Xiao, Zhi-Yan Fang, Chun-Wei Tsai","doi":"10.1145/3440943.3444723","DOIUrl":"https://doi.org/10.1145/3440943.3444723","url":null,"abstract":"A precise forecasting for the future short-term electricity consumption will be quite useful for making a good plan for the power demand management. Since deep neural network (DNN) provides an effective way for the short-term load forecasting, one of the focuses of this research is thus to use it to construct the prediction model. The gradient-based optimizers (GBOs) have been widely used in DNN algorithms in recent years; however, they make it easy for DNN to trap in poor local regions during the training process. In this research, we propose a new optimizer that is based on the searched experiences to solve this problem to enhance the performance of GBOs. More precisely, the proposed optimizer integrates the best position searched, Lévy flight, and gradient descent to preserve not only the diversification but also the intensification of search during the training process. To evaluate the performance of the proposed optimizer, we compare it with several state-of-the-art GBOs for DNN; namely, Adagrad, RMSprop, and Adam, for training a forecasting model for the short-term electricity consumption forecasting problem. The simulation results show that the proposed optimizer outperforms all the other GBOs in terms of the mean absolute percentage error.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125324305","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}
This study aims at summarizing narrative works (i.e., creative works that contain stories) in the consideration of their stories and types of required summaries. Various methods for story-based summarization have been proposed as a practical application of the character network analysis (i.e., a social network among characters that appeared in a story). However, the existing methods do not consider that summaries have different requirements according to their types (e.g., trailers, highlights, and recaps). These methods consist of three parts: (i) discretizing narrative works into regular units (e.g., scenes or shots), (ii) measuring the narrative significance of each unit, and (iii) generating summaries based on the narrative significance. Most of the existing studies have proposed their unique significance measurements based on individual narrative features. Also, since these methods have not considered the diverse types of summaries, they have simply selected top-N narrative units according to the measurements. In this study, we first introduce and redefine the narrative significance measurements. Subsequently, we propose a method for summarizing a narrative work regarding the requirements of the summaries by integrating the various significance measurements.
{"title":"Towards Story-based Summarization of Narrative Multimedia","authors":"O-Joun Lee, Jin-Taek Kim, Eunsoon You","doi":"10.1145/3440943.3444719","DOIUrl":"https://doi.org/10.1145/3440943.3444719","url":null,"abstract":"This study aims at summarizing narrative works (i.e., creative works that contain stories) in the consideration of their stories and types of required summaries. Various methods for story-based summarization have been proposed as a practical application of the character network analysis (i.e., a social network among characters that appeared in a story). However, the existing methods do not consider that summaries have different requirements according to their types (e.g., trailers, highlights, and recaps). These methods consist of three parts: (i) discretizing narrative works into regular units (e.g., scenes or shots), (ii) measuring the narrative significance of each unit, and (iii) generating summaries based on the narrative significance. Most of the existing studies have proposed their unique significance measurements based on individual narrative features. Also, since these methods have not considered the diverse types of summaries, they have simply selected top-N narrative units according to the measurements. In this study, we first introduce and redefine the narrative significance measurements. Subsequently, we propose a method for summarizing a narrative work regarding the requirements of the summaries by integrating the various significance measurements.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126854617","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}
Based on the 4th Industrial Revolution, numerous ICT technologies are developing, and for this reason, IoT devices are formed around us. Accordingly, hackers are inflicting financial and physical damage to our lives by using software vulnerabilities in IoT devices around us based on intelligent hacking technology. Automated security vulnerability response systems are required to respond to attacks through continuously occurring software vulnerabilities. In this paper, we analyze the hybrid fuzzing system that complements the technical limitations of the existing fuzzing technology as the base technology for an automated security vulnerability response system. In addition, we propose a hybrid fuzzing system based on an automatic seed generation mechanism for coverage efficiency in order to find vulnerabilities inherent in software quickly and efficiently.
{"title":"Automatic Seed Generation based Hybrid Fuzzing for Code Coverage Efficiency","authors":"Soonhong Kwon, Yangseo Choi, Jong‐Hyouk Lee","doi":"10.1145/3440943.3444744","DOIUrl":"https://doi.org/10.1145/3440943.3444744","url":null,"abstract":"Based on the 4th Industrial Revolution, numerous ICT technologies are developing, and for this reason, IoT devices are formed around us. Accordingly, hackers are inflicting financial and physical damage to our lives by using software vulnerabilities in IoT devices around us based on intelligent hacking technology. Automated security vulnerability response systems are required to respond to attacks through continuously occurring software vulnerabilities. In this paper, we analyze the hybrid fuzzing system that complements the technical limitations of the existing fuzzing technology as the base technology for an automated security vulnerability response system. In addition, we propose a hybrid fuzzing system based on an automatic seed generation mechanism for coverage efficiency in order to find vulnerabilities inherent in software quickly and efficiently.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125494951","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 various potential hazards that may occur in diving, but some risks can be prevented through advance planning, such as diving decompression risk is one of them. This article construction of diving record wearable device based on IOT. The instructor can set the diving height and time according to the situation of each student, and there is an emergency button set on the wearable device. When students have any problems, they can use AIS communication to send messages to the coach ship, and the coach ship can immediately locate and rescue. This article mainly uses GPS and Beacon for positioning to understand the student's position and height. This article uses the IoT sensor to understand the student's movement rate and height to avoid the danger that the student's ascent rate is too fast or the altitude is too high.
{"title":"Construction of Diving Record Wearable Device Based on IOT","authors":"Yen-Chiao Chuang, Cheng-Lung Wu, I-Yun Chen, Chien-Hung Wu","doi":"10.1145/3440943.3444345","DOIUrl":"https://doi.org/10.1145/3440943.3444345","url":null,"abstract":"There are various potential hazards that may occur in diving, but some risks can be prevented through advance planning, such as diving decompression risk is one of them. This article construction of diving record wearable device based on IOT. The instructor can set the diving height and time according to the situation of each student, and there is an emergency button set on the wearable device. When students have any problems, they can use AIS communication to send messages to the coach ship, and the coach ship can immediately locate and rescue. This article mainly uses GPS and Beacon for positioning to understand the student's position and height. This article uses the IoT sensor to understand the student's movement rate and height to avoid the danger that the student's ascent rate is too fast or the altitude is too high.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131561026","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 input image scale must be considered in the microsatellite instability recognition method through deep learning image analysis. Since pathological images can observe various features through high magnification, an image analysis method capable of analyzing high-resolution images is required. Although CNN has excellent image analysis capabilities, the size of input images is limited. If we want to analyze an area bigger than the input image size of the CNN, the area should be reduced or crop. In this paper, we propose a recombination block that extracts and combines features in patch units to handle microsatellite images made up of high-resolution images.
{"title":"Reunion Block for High Magnification Histopathology Microscopic Image Analysis","authors":"Hyun-Cheol Park, Sang-Woong Lee","doi":"10.1145/3440943.3444356","DOIUrl":"https://doi.org/10.1145/3440943.3444356","url":null,"abstract":"The input image scale must be considered in the microsatellite instability recognition method through deep learning image analysis. Since pathological images can observe various features through high magnification, an image analysis method capable of analyzing high-resolution images is required. Although CNN has excellent image analysis capabilities, the size of input images is limited. If we want to analyze an area bigger than the input image size of the CNN, the area should be reduced or crop. In this paper, we propose a recombination block that extracts and combines features in patch units to handle microsatellite images made up of high-resolution images.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127417402","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}
Since there are way too many possible combinations of hyperparameters for training a desired deep neural network (DNN) model, finding out a set of suitable values for them is typically a difficult topic for the researchers when they use DNN for solving forecasting problems. In addition to manual tuning and trial-and-error for hyperparameters, how to automatically determine the values of hyperparameters has become a critical problem in recent years. In this study, we present a metaheuristic algorithm based on the whale optimization algorithm (WOA) to select suitable hyperparameters for the DNN because WOA demonstrates brilliant convergence speed in many optimization problems and the local optima avoidance mechanism is devised to prevent the searches from trapping into suboptimal solution easily. To validate the feasibility of the proposed algorithm, we compared it with several state-of-the-art hyperparameter selection algorithms for DNN in solving the network-wide traffic speed prediction problem. The experimental results show that WOA not only behaves much more stable but also outperforms all the other hyperparameter selection algorithms compared in this study in terms of the mean square error, mean average error, and mean average percentage error.
{"title":"The Whale Optimization Algorithm for Hyperparameter Optimization in Network-Wide Traffic Speed Prediction","authors":"Zhang-Han Zhuang, Ming-Chao Chiang","doi":"10.1145/3440943.3444729","DOIUrl":"https://doi.org/10.1145/3440943.3444729","url":null,"abstract":"Since there are way too many possible combinations of hyperparameters for training a desired deep neural network (DNN) model, finding out a set of suitable values for them is typically a difficult topic for the researchers when they use DNN for solving forecasting problems. In addition to manual tuning and trial-and-error for hyperparameters, how to automatically determine the values of hyperparameters has become a critical problem in recent years. In this study, we present a metaheuristic algorithm based on the whale optimization algorithm (WOA) to select suitable hyperparameters for the DNN because WOA demonstrates brilliant convergence speed in many optimization problems and the local optima avoidance mechanism is devised to prevent the searches from trapping into suboptimal solution easily. To validate the feasibility of the proposed algorithm, we compared it with several state-of-the-art hyperparameter selection algorithms for DNN in solving the network-wide traffic speed prediction problem. The experimental results show that WOA not only behaves much more stable but also outperforms all the other hyperparameter selection algorithms compared in this study in terms of the mean square error, mean average error, and mean average percentage error.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"224 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128284330","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}
Whole slide imaging is now being used across the world in pathology labs for an accurate diagnosis of biopsy specimens. However, due to the large size of these images, an automatic deep learning-based method is highly desirable for diagnosing. Herein, we propose a two-step methodology for the classification and segmentation of whole-slide image (WSI). First, the patches are extracted from the image and fed into deep learning based techniques like U-Net with its corresponding mask for the accurate segmentation. Further, the cancerous patches are trained for the classification task. During inference, the predicted segmented mask are evaluated in the classification model. Our experimental results demonstrated that the proposed methodology can be used for accurate segmentation and classification.
{"title":"Whole Slide Image Classification and Segmentation using Deep Learning","authors":"S. Poudel, Sang-Woong Lee","doi":"10.1145/3440943.3444357","DOIUrl":"https://doi.org/10.1145/3440943.3444357","url":null,"abstract":"Whole slide imaging is now being used across the world in pathology labs for an accurate diagnosis of biopsy specimens. However, due to the large size of these images, an automatic deep learning-based method is highly desirable for diagnosing. Herein, we propose a two-step methodology for the classification and segmentation of whole-slide image (WSI). First, the patches are extracted from the image and fed into deep learning based techniques like U-Net with its corresponding mask for the accurate segmentation. Further, the cancerous patches are trained for the classification task. During inference, the predicted segmented mask are evaluated in the classification model. Our experimental results demonstrated that the proposed methodology can be used for accurate segmentation and classification.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129324652","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}
I. Liu, Yung-Lin Chang, Jung-Shian Li, Chuan-Gang Liu
Due to the rapid spread of tablet computers, smartphones, and other mobile information devices, wireless communication technology is fully developed and deployed widely. Mobile Internet access has been a main and important way to communicate the world in daily and it privacy protection also catches much attentions. The Onion Router, better known as Tor, is a technique for anonymous communication over internet without regional restrictions for privacy protection. Apart from this, Tor can reach sites that normal search engine cannot search. As the opinion of Tor, the transmitted data has been encrypted layer by layer, just like onion, before it reaching server. Our research proposed a system predicting URL's category with the use of machine learning technique before visiting. According to the category prediction, we represent different privacy level with three kinds of RSA key lengths on onion routing. Depending on various situations, our system can obtain the balance between security and time cost. Hence, our proposed scheme can make onion routing more flexible and efficiently.
{"title":"Differential Privacy Protection with Group Onion Routing based on AI-based URL Classification","authors":"I. Liu, Yung-Lin Chang, Jung-Shian Li, Chuan-Gang Liu","doi":"10.1145/3440943.3444717","DOIUrl":"https://doi.org/10.1145/3440943.3444717","url":null,"abstract":"Due to the rapid spread of tablet computers, smartphones, and other mobile information devices, wireless communication technology is fully developed and deployed widely. Mobile Internet access has been a main and important way to communicate the world in daily and it privacy protection also catches much attentions. The Onion Router, better known as Tor, is a technique for anonymous communication over internet without regional restrictions for privacy protection. Apart from this, Tor can reach sites that normal search engine cannot search. As the opinion of Tor, the transmitted data has been encrypted layer by layer, just like onion, before it reaching server. Our research proposed a system predicting URL's category with the use of machine learning technique before visiting. According to the category prediction, we represent different privacy level with three kinds of RSA key lengths on onion routing. Depending on various situations, our system can obtain the balance between security and time cost. Hence, our proposed scheme can make onion routing more flexible and efficiently.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127268850","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}
Ming-Ying Lee, Hanbyeol Kang, Jun-Seok Kwak, Donghyun Kim, Hye-Lim Jeong, Jung-Taek Seo
With the advancement of information and communication technology, online file exchange such as document delivery, business reporting, and submission of reports within or between companies has been widely used. With the increase in convenience, however, security threats have also increased due to the high risk of external leak of files exchanged online. Although companies employ network separation by building an internal corporate network, this has a drawback, i.e., files cannot be used outside such as a work-from-home environment. To solve this problem, this paper proposes a secure file sharing system using the Fast IDentify Online (FIDO) technique. This service is expected to contribute to the improvement of file sharing security because it can encrypt and decrypt files through FIDO authentication when viewing internal files from outside.
{"title":"Proposed On-site Document Sharing System using FIDO","authors":"Ming-Ying Lee, Hanbyeol Kang, Jun-Seok Kwak, Donghyun Kim, Hye-Lim Jeong, Jung-Taek Seo","doi":"10.1145/3440943.3444740","DOIUrl":"https://doi.org/10.1145/3440943.3444740","url":null,"abstract":"With the advancement of information and communication technology, online file exchange such as document delivery, business reporting, and submission of reports within or between companies has been widely used. With the increase in convenience, however, security threats have also increased due to the high risk of external leak of files exchanged online. Although companies employ network separation by building an internal corporate network, this has a drawback, i.e., files cannot be used outside such as a work-from-home environment. To solve this problem, this paper proposes a secure file sharing system using the Fast IDentify Online (FIDO) technique. This service is expected to contribute to the improvement of file sharing security because it can encrypt and decrypt files through FIDO authentication when viewing internal files from outside.","PeriodicalId":310247,"journal":{"name":"Proceedings of the 2020 ACM International Conference on Intelligent Computing and its Emerging Applications","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127204628","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：