2020 International Symposium on Computer, Consumer and Control (IS3C)最新文献

{"title":"AI-based College Course Selection Recommendation System: Performance Prediction and Curriculum Suggestion","authors":"Yu-Hsuan Wu, Eric Hsiao-Kuang Wu","doi":"10.1109/IS3C50286.2020.00028","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00028","url":null,"abstract":"Recent advances of AI applications in various of industries have led to remarkable performance and efficiency. Driven by the great success of datasets and experience sharing, people are exploring more precious datasets with diverse features and longer time range. The promising reasoning information of well-curated student grade datasets is expected to assist young students to find the best of themselves and then improve their learning outcome and study experience. Through data and experience sharing, young students can have a better understanding of their learning condition and possible learning outcomes. Existing course selection systems in Taiwan which offer limited basic enrolling functions fail to provide performance prediction and course arrangement guidance based on their own learning condition. Students now selecting courses with unawareness of their expecting performance. A personalized guide for students on course selection is crucial for how they structure professional knowledge and arrange study schedule. In this paper, we first analyzed what factors can be used on defining learning curve, and discovered the difference between students with different properties and background. Second, we developed a recommendation system based on great amount of grade datasets of past students, and the system can give students suggestions on how to assign their credits based on their own learning curve and students that had similar learning curve. The result of our research demonstrates the feasibility of a new approach on applying big data and AI technology on learning analysis and course selection.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121490936","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}

{"title":"Classification of driving fatigue based on EEG signals","authors":"Xuebin Qin, Peijiao Yang, Yutong Shen, Mingqiao Li, Jiachen Hu, Janhong Yun","doi":"10.1109/IS3C50286.2020.00138","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00138","url":null,"abstract":"traffic accidents bring serious harm to individuals and society. Fatigue driving has many potential safety hazards, which is the main factor causing road traffic accidents. Therefore, it is urgent to monitor the fatigue system. Firstly, the EEG signals are preprocessed by Butterworth band-pass filter, and then the features are extracted by wavelet transform. The classification results of fatigue EEG signals based on support vector machine are used as the initial fatigue value. Then RANSAC method is used to select fatigue signal. Finally, according to the average value of signals screened by RANSAC method as the standard value, the driver's fatigue state is determined by calculating the Euclidean distance between the standard value and the fatigue EEG signal. The experimental results show that the accuracy of the proposed method is better than that of the traditional method, which can reach 90%. It is easy to use and has wide application value.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"282 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132279432","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}

{"title":"Market Profile with Convolutional Neural Networks: Learning the Structure of Price Activities","authors":"Chern-Bin Ju, Min-Chih Hung, An-Pin Chen","doi":"10.1109/IS3C50286.2020.00123","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00123","url":null,"abstract":"Many studies tried to apply machine learning (ML) methods to forecast the financial time series during the past decade. Moreover, the emergence of deep learning has led to another researches with results that significantly outperform previous models. However, most of the deep learning researches used the raw financial time series price data which consists of open, close, high and low (OCHL) as learning features. According to statistics, Long Short Term Memory networks (LSTM) is the first choice to deal with the forecasting problems with OCHL datasets due to its feedback connection networks, resulting higher performances for price series prediction. Meanwhile, Convolutional Neural Networks (CNN) has increased its popularity since it outperforms traditional ML models in classification problems. In this paper, there are three types of future trend that are the ultimate targets to be discovered. Nevertheless, OCHL features may be too sensitive to learn the large future trend in financial time series. This study proposes a novel approach: Convolutional Neural Networks with Market Profiles (CNN-MPs) which includes (1) adapting Market Profile to covert financial time series data to grey-scale image method, (2) generating two types of learning images: stacked and sequential profile that can keep the interaction between continuous profiles, and (3) learning the structure of price activities with CNN. Market Profile is a concept that has been widely used in the financial decision-making by comparing the current price with the market fair value. In addition, the trend is well established at the accepting movement of fair value which can be confirmed from the structure of profiles. In experiments, one of the popular commodities, corn was selected to evaluate the proposed method. And the experimental results show that proposed sequential profile method obtained 17% higher accuracy and more profitability than LSTM networks and other methods. Therefore, the proposed CNN-MPs method can effectively discover the trend of corn providing those who need import corn with a reference.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130799003","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}

{"title":"A Systematic Evaluation for Paint Messages and Their Processing among C++/CLI Controls","authors":"Gao-Wei Chang","doi":"10.1109/IS3C50286.2020.00051","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00051","url":null,"abstract":"This paper is to propose a systematic evaluation for paint messages among C++/CLI controls, e.g., form, button, and picture box. Specifically, the occurrences of such messages and priorities of their processing are investigated to enhance the design of programming models for graphical user interface (GUI). To demonstrate the effectiveness of this approach, those models are progressively integrated with processing paint messages from various controls, in a graphically-demanding C++/CLI project.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117209304","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}

{"title":"Short-term Photovoltaic Power Forecasting Based on Improved Firefly Algorithm to optimize support vector machine","authors":"C. Nsengimana, Xiu Jun Shen, X. Han, Ling-ling Li, Haiyu Wang","doi":"10.1109/IS3C50286.2020.00095","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00095","url":null,"abstract":"With the current increasing demand in energy consumption, there is a huge increase of prominent energy problems that require us to imperatively seek for the new green energy sources. Photovoltaic power generation is one of the most feasible power generation methods due to its high cleanliness and static characteristics. This paper proposes a photoelectric power prediction method based on an improved firefly algorithm to optimize support vector machines (SVM) for short-term prediction. We effectively combine the regression support vector machine (SVR) with the modified firefly algorithm (MFFA) and use the firefly estimation method to determine the best fitness penalty factor c and kernel function g, so that the support vector machine can better predict the photovoltaic power. In order to make the firefly algorithm to optimize the support vector machine faster, we improved the firefly algorithm step factor $a$ and introduced a weight coefficient ϖ, Compared with conventional techniques, this method has better prediction results and prediction speed is also better than the traditional intelligent optimization models. Let's take the data from a photovoltaic base in the Desert Knowledge Australian Solar Energy Centre (DKASC) as an example.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128289340","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}

{"title":"Chi-Square Detection for PVD Steganography","authors":"I. Pan, Kung-Chin Liu, Chiang-Lung Liu","doi":"10.1109/IS3C50286.2020.00015","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00015","url":null,"abstract":"Although the Pixel-Value Differencing (PVD) steganography can avoid being detected by the RS steganalysis, the histogram of the pixel-value differences poses an abnormal distribution. Based on this hiding characteristic, this paper proposes a PVD steganalysis based on chi-Square statistics. The degrees of freedom were adopted to be tested for obtaining various detection accuracies (ACs). Experimental results demonstrate the detection accuracies are all above 80%. When the degrees of freedom are set as 10 while the accuracy is the best (AC = 83%). It means that the proposed Chi-Square based method is an efficient detection for PVD steganography.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134618242","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}

{"title":"End-to-End Deep Learning Model for Steering Angle Control of Autonomous Vehicles","authors":"Abida Khanum, Chao-Yang Lee, Chu-Sing Yang","doi":"10.1109/IS3C50286.2020.00056","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00056","url":null,"abstract":"Recently brilliant evolutions in the machine learning research area of autonomous self-driving vehicles. Unlike a modern rule-based method, this study has been supervised on the manipulate of images end-to-end, which is deep learning. The motivation of this paper where the input to the model is the camera image and the output is the steering angle target. The model trained a Residual Neural Network (ResNet) convolutional neural network(CNN) algorithm to drive an autonomous vehicle in the simulator. Therefore, the model trained and simulation are conducted using the UDACITY platform. The simulator has two choices one is the training and the second one is autonomous. The autonomous has two tracks track _1 considered as simple and track _2 complex as compare to track_1. In our paper, we used track_1 for autonomous driving in the simulator. The training option gives the recorded dataset its control through the keyboard in the simulator. We collected about 11655 images (left, center, right) with four attributes (steering, throttle, brake, speed) and also images dataset stored in a folder and attributes dataset save as CSV file in the same path. The stored raw images and steering angle data set used in this method. We divided 80–20 data set for training and Validation as shown in Table I. Images were sequentially fed into the convolutional neural network (ResNet)to predict the driving factors for making end planning decisions and execution of autonomous motion of vehicles. The loss value of the proposed model is 0.0418 as shown in Figure 2. The method trained takes succeeded precision of 0.81% is good consent with expected performance.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122969123","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}

{"title":"Smart Home Trash Can based on Artificial Intelligence Technologies","authors":"Yu-Huei Cheng, Hongning. Ruan, Jiun-Jian Liaw, Che-Nan Kuo","doi":"10.1109/IS3C50286.2020.00021","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00021","url":null,"abstract":"In recent years, with the development of wireless communication, network, Internet of things and artificial intelligence, the technology of smart home application in the environment is gradually mature, and the functions of related equipment are more and more diverse. Considering the application of smart home energy saving and home appliances. In this research, Arduino and NVIDIA Jetson Nano are used as the underlying hardware, and artificial intelligence technology are used to develop AI intelligent home trash can. AI intelligent home trash can is mainly designed based on the concept of home appliances. It uses artificial intelligence control and machine learning technology to achieve the function of automatically catching the falling trash.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"414 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124427518","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}

{"title":"Attention-LSTM Fused U-Net Architecture for Organ Segmentation in CT Images","authors":"Pin-Hsiu Chen, Cheng-Hsien Huang, S. Hung, Liang-Cheng Chen, Hui-Ling Hsieh, W. Chiou, Moon-Sing Lee, Hon-Yi Lin, Wei-Min Liu","doi":"10.1109/IS3C50286.2020.00085","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00085","url":null,"abstract":"During the treatment planning stage of the radiotherapy, the medical physicists or doctors have to delineate the contour of the tumor and the organs at risk in order to accurately deliver enough radiation energy to the tumor and reduce such exposure to the surrounding normal tissues. Organ contouring is a time-consuming and laborious task. An automatic contouring tool is definitely required to fulfill the needs of the increasing cancer population. In this work, we proposed a fusion model that combines the network characteristics of a sequential model (sensor3D) and Attention U-Net. In which the convolutional LSTM layer is applied to study the spatial correlation between different layers in a CT image data set. The attention mechanism suppresses the irrelevant features from the complex image content and focuses on the useful messages of target organs. The clinical data contained CT image series of 108 patients and was acquired from a local hospital with IRB approval. The proposed model segmented five types of organs, lung, liver, stomach, esophagus, and heart. The segmentation accuracy rates of Dice Similarity Coefficient (DSC) were 99.27%, 95.48%, 88.19%, 80.81%, and 93.8%, respectively. We further developed a user interface that converts the AI-generated results into DICOM-RT format. Therefore the radiologists can fine-tune the results under the software used to do the routine manual-delineation tasks.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127632990","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}

{"title":"An Artificial Intelligence-based Proactive Blind Spot Warning System for Motorcycles","authors":"Ing-Chau Chang, Wei-Rong Chen, Xun-Mei Kuo, Yajun Song, Ping-Hao Liao, Chunghui Kuo","doi":"10.1109/IS3C50286.2020.00110","DOIUrl":"https://doi.org/10.1109/IS3C50286.2020.00110","url":null,"abstract":"The goal of this research is to design a proactive bus blind spot warning (PBSW) system which will notify the motorcycle riders as soon as they enter the blind spot of a target vehicle, i.e., bus. The motorcycle side of this PBSW system, consisting of a Raspberry Pi 3B+ and a dual-lens stereo camera, will first transmit captured images to the Android phone using Wi-Fi and then to the cloud server through the cellular network. At the cloud server, the famous AI model, YOLOv4, is used to recognize the position of the rear-view mirror of the bus. By the principle of lens imaging, the distance between the bus and the motorcycle is estimated. Based on the estimated distance returned from the cloud server, the PBSW APP running in the Android phone illustrates the visible area/blind spot of the bus, the position of the rider and the estimated distance between the motorcycle and the bus. It further alarms the rider whenever the rider has entered the blind spot of the bus. According to performance evaluation on this implemented system, it recognizes the rear-view mirror with the average accuracy of 92.82%, the error rate of the estimated distance lower than 0.2% and the average round trip delay of 0.5 sec. It is concluded that this PBSW system keeps the motorcycle rider away from imminent dangers in real time.","PeriodicalId":143430,"journal":{"name":"2020 International Symposium on Computer, Consumer and Control (IS3C)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127914260","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}