Pub Date : 2021-05-01DOI: 10.6688/JISE.202105_37(3).0013
Jongbaeg Lee, Jong-Woo Lee, Sang-Won Lee
{"title":"Optimizing Read Operations of Hadoop Distributed File System on Heterogeneous Storages","authors":"Jongbaeg Lee, Jong-Woo Lee, Sang-Won Lee","doi":"10.6688/JISE.202105_37(3).0013","DOIUrl":"https://doi.org/10.6688/JISE.202105_37(3).0013","url":null,"abstract":"","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"2 1","pages":"709-729"},"PeriodicalIF":1.1,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83810824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-05-01DOI: 10.6688/JISE.20210537(3).0011
Chung-Ming Huang, Chung-Ming Wei
With the advance of wireless mobile communication technologies, video streaming has advanced much more on these years. However, the wireless mobile network usually suffers fluctuation in available bandwidth because mobile users usually keep moving and playing streaming video simultaneously. To overcome the situation, this work proposed a method called Stepwise Adaptive Streaming using Temporal-Geo Bandwidth Estimation (SASTGBE) for the wireless mobile networking environment based on MPEG Dynamic Adaptive Streaming over HTTP (MPEG-DASH). To have better Quality Of Experience (QoE), e.g., no or fewer quality switching, lower bitrate difference between two continuous video segments, no or fewer suspended times, no or shorter paused time, etc., (1) the proposed method considers location, time, and date for estimating the available bandwidth in the future and (2) the proposed stepwise adaptive streaming control scheme considers buffer level, video quality of the most recently downloaded segment, the downloading rate of the most recently downloaded segment and the estimated bandwidth to decide the video quality for the next downloaded segment. The proposed method has been implemented in the Android system for the client side and the Linux system for the server side. The experiments using SASTGBE in the real environment shown that SASTGBE has improvement in the performance of bandwidth utilization, suspended times, quality switch percentage, average bitrate difference considering suspending, and standard deviation of bitrate difference over the wireless mobile network.
{"title":"Stepwise Adaptive Video Streaming in the Wireless Mobile Network Using the Temporal-Geo Bandwidth Estimation Method","authors":"Chung-Ming Huang, Chung-Ming Wei","doi":"10.6688/JISE.20210537(3).0011","DOIUrl":"https://doi.org/10.6688/JISE.20210537(3).0011","url":null,"abstract":"With the advance of wireless mobile communication technologies, video streaming has advanced much more on these years. However, the wireless mobile network usually suffers fluctuation in available bandwidth because mobile users usually keep moving and playing streaming video simultaneously. To overcome the situation, this work proposed a method called Stepwise Adaptive Streaming using Temporal-Geo Bandwidth Estimation (SASTGBE) for the wireless mobile networking environment based on MPEG Dynamic Adaptive Streaming over HTTP (MPEG-DASH). To have better Quality Of Experience (QoE), e.g., no or fewer quality switching, lower bitrate difference between two continuous video segments, no or fewer suspended times, no or shorter paused time, etc., (1) the proposed method considers location, time, and date for estimating the available bandwidth in the future and (2) the proposed stepwise adaptive streaming control scheme considers buffer level, video quality of the most recently downloaded segment, the downloading rate of the most recently downloaded segment and the estimated bandwidth to decide the video quality for the next downloaded segment. The proposed method has been implemented in the Android system for the client side and the Linux system for the server side. The experiments using SASTGBE in the real environment shown that SASTGBE has improvement in the performance of bandwidth utilization, suspended times, quality switch percentage, average bitrate difference considering suspending, and standard deviation of bitrate difference over the wireless mobile network.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"78 1","pages":"679-696"},"PeriodicalIF":1.1,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81065813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-04-01DOI: 10.6688/JISE.202107_37(4).0001
Tsung-Pin Chen, Lin Chih-Lung, Kuo-Chin Fan, Lin Wan-Yu, Wan-Yu Kao
{"title":"Radar Automatic Target Recognition Based on Real-Life HRRP of Ship Target by Using Convolutional Neural Network","authors":"Tsung-Pin Chen, Lin Chih-Lung, Kuo-Chin Fan, Lin Wan-Yu, Wan-Yu Kao","doi":"10.6688/JISE.202107_37(4).0001","DOIUrl":"https://doi.org/10.6688/JISE.202107_37(4).0001","url":null,"abstract":"","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"27 1","pages":"733-752"},"PeriodicalIF":1.1,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88727415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.6688/JISE.202103_37(2).0011
Shan-Chun Liu, Da-Chun Wu, Wen-Hsiang Tsai
A new type of art image, called rectangle-based Neo-Plasticism-like image, is proposed, via which messages can be hidden for covert communication. Also proposed is an automatic method for creating such art images, which applies recursive binary partitioning to a source image by finding the maximum mutual information of the spatial positions and image intensities of the divided sub-regions. The resulting image consists of rectangular regions separated by horizontal and vertical lines, which show the abstraction style of the Neo-Plasticism art. Attracted by the artistic image content, a hacker hopefully will pay no attention to the hidden secret. Two data hiding techniques based on the binary partition tree constructed in the art image creation process are proposed, which embed messages by replacing the LSBs of each rectangular region's colors or by generating additional partition lines in the region. A message extraction process is also proposed. Data security is considered seriously by randomizing the message bits before being embedded, changing randomly the priorities of the sub-regions used in message hiding, and embedding fake messages to interfere with the hacker. Good experimental results show the feasibility of the proposed techniques for covert communication via the proposed type of art image.
{"title":"Rectangle-Based Neo-Plasticism-Like Image - A New Type of Art Image and its Application to Covert Communication","authors":"Shan-Chun Liu, Da-Chun Wu, Wen-Hsiang Tsai","doi":"10.6688/JISE.202103_37(2).0011","DOIUrl":"https://doi.org/10.6688/JISE.202103_37(2).0011","url":null,"abstract":"A new type of art image, called rectangle-based Neo-Plasticism-like image, is proposed, via which messages can be hidden for covert communication. Also proposed is an automatic method for creating such art images, which applies recursive binary partitioning to a source image by finding the maximum mutual information of the spatial positions and image intensities of the divided sub-regions. The resulting image consists of rectangular regions separated by horizontal and vertical lines, which show the abstraction style of the Neo-Plasticism art. Attracted by the artistic image content, a hacker hopefully will pay no attention to the hidden secret. Two data hiding techniques based on the binary partition tree constructed in the art image creation process are proposed, which embed messages by replacing the LSBs of each rectangular region's colors or by generating additional partition lines in the region. A message extraction process is also proposed. Data security is considered seriously by randomizing the message bits before being embedded, changing randomly the priorities of the sub-regions used in message hiding, and embedding fake messages to interfere with the hacker. Good experimental results show the feasibility of the proposed techniques for covert communication via the proposed type of art image.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"23 1","pages":"441-468"},"PeriodicalIF":1.1,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81834497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-03-01DOI: 10.6688/JISE.202103_37(2).0006
Che-Wen Chen, Shih-Pang Tseng, Jhing-Fa Wang
Outpatient text classification is an important problem in medical natural language processing. Existing research has conventionally focused on rule-based or knowledge-source-based feature engineering, but only a few studies have utilized the effective feature learning capabilities of deep learning methods. A long short-term memory (LSTM) model for the outpatient text classification system was proposed in this research. The system has the ability to classify outpatient categories according to textual content on website Taiwan E Hospital. The experimental results showed that our system has very well in the task. The success of the LSTM model applications in the outpatient system provide users to inquire about their health status as references.
{"title":"Outpatient Text Classification System Using LSTM","authors":"Che-Wen Chen, Shih-Pang Tseng, Jhing-Fa Wang","doi":"10.6688/JISE.202103_37(2).0006","DOIUrl":"https://doi.org/10.6688/JISE.202103_37(2).0006","url":null,"abstract":"Outpatient text classification is an important problem in medical natural language processing. Existing research has conventionally focused on rule-based or knowledge-source-based feature engineering, but only a few studies have utilized the effective feature learning capabilities of deep learning methods. A long short-term memory (LSTM) model for the outpatient text classification system was proposed in this research. The system has the ability to classify outpatient categories according to textual content on website Taiwan E Hospital. The experimental results showed that our system has very well in the task. The success of the LSTM model applications in the outpatient system provide users to inquire about their health status as references.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"75 1","pages":"365-379"},"PeriodicalIF":1.1,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76229215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.6688/JISE.202101_37(1).0006
Liang Wang, Ruixia Xue, Ning Cai, Pan Chen, Xiaobo Cui, Wei Wu, Miaomiao Niu, Dongliang Zhang, Zhao Zhang, Xiaosong Zhang
This paper puts forward a dynamic parallel meshless computing algorithm that efficiently solves flow fields with largescale motions of movable and deformable boundaries. The partition boundary is updated, as the moving boundary moves across the material interface. Meanwhile, the point clouds near the moving boundary are reconstructed. Our algorithm also solves the workload balance between nodes and information exchange in each region of the computational field, using the governing equations in the arbitrary Lagrangian-Eulerian (ALE) form. The AUFS scheme is extended to calculate the numerical convective flux. Take the interaction between a helium bubble and a shockwave as an example. Our algorithm is applied to compute the flow field with different numbers of discrete points (33,044 and 66,089) and partitions (2 and 4). The results show that our algorithm achieves an efficiency of over 80%, and captures the interaction between shockwaves and the bubble accurately. Hence, our parallel algorithm is suitable for solving problems with largescale motions of deformation boundaries. The research results shed new light on the calculation speed for similar problems.
{"title":"A Dynamic Parallel Meshless Method for the Problems with Large-Scale Movable and Deformable Boundary","authors":"Liang Wang, Ruixia Xue, Ning Cai, Pan Chen, Xiaobo Cui, Wei Wu, Miaomiao Niu, Dongliang Zhang, Zhao Zhang, Xiaosong Zhang","doi":"10.6688/JISE.202101_37(1).0006","DOIUrl":"https://doi.org/10.6688/JISE.202101_37(1).0006","url":null,"abstract":"This paper puts forward a dynamic parallel meshless computing algorithm that efficiently solves flow fields with largescale motions of movable and deformable boundaries. The partition boundary is updated, as the moving boundary moves across the material interface. Meanwhile, the point clouds near the moving boundary are reconstructed. Our algorithm also solves the workload balance between nodes and information exchange in each region of the computational field, using the governing equations in the arbitrary Lagrangian-Eulerian (ALE) form. The AUFS scheme is extended to calculate the numerical convective flux. Take the interaction between a helium bubble and a shockwave as an example. Our algorithm is applied to compute the flow field with different numbers of discrete points (33,044 and 66,089) and partitions (2 and 4). The results show that our algorithm achieves an efficiency of over 80%, and captures the interaction between shockwaves and the bubble accurately. Hence, our parallel algorithm is suitable for solving problems with largescale motions of deformation boundaries. The research results shed new light on the calculation speed for similar problems.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"58 1","pages":"79-92"},"PeriodicalIF":1.1,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78081687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.6688/JISE.202101_37(1).0007
Li-Zhang Xu, Q. Xie
The current production scheduling models cannot effectively enable the real-time interaction between information space and physical space. To dynamically schedule twin digital job-shop, this paper attempts to realize the dynamic scheduling of digital twin job-shop (DTJ) based on edge computing. First, the architecture of the DTJ was established by adding the digital twin between the business management layer and the operation execution layer of the traditional job-shop. On this basis, the DTJ was fully modelled, and the manufacturing process was monitored, analyzed and managed remoted by edge computing. To realize dynamic scheduling, a DTJ scheduling model was established through data mining. The model consists of two parts: a data collection model and a multi-scheduling knowledge model. Finally, the proposed DTJ scheduling model was verified through simulation on an actual job-shop. The research results shed new light on the optimization of manufacturing process in various types of job-shops.
{"title":"Dynamic Production Scheduling of Digital Twin Job-Shop Based on Edge Computing","authors":"Li-Zhang Xu, Q. Xie","doi":"10.6688/JISE.202101_37(1).0007","DOIUrl":"https://doi.org/10.6688/JISE.202101_37(1).0007","url":null,"abstract":"The current production scheduling models cannot effectively enable the real-time interaction between information space and physical space. To dynamically schedule twin digital job-shop, this paper attempts to realize the dynamic scheduling of digital twin job-shop (DTJ) based on edge computing. First, the architecture of the DTJ was established by adding the digital twin between the business management layer and the operation execution layer of the traditional job-shop. On this basis, the DTJ was fully modelled, and the manufacturing process was monitored, analyzed and managed remoted by edge computing. To realize dynamic scheduling, a DTJ scheduling model was established through data mining. The model consists of two parts: a data collection model and a multi-scheduling knowledge model. Finally, the proposed DTJ scheduling model was verified through simulation on an actual job-shop. The research results shed new light on the optimization of manufacturing process in various types of job-shops.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"6 1","pages":"93-105"},"PeriodicalIF":1.1,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82511399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.6688/jise.202109_37(5).0008
Pratik A. Barot, H. Jethva
In the time of COVID-19, medical facilities struggling to fight against the pandemic. Most of the countries face a tough time fighting against this virus outbreak. Even developed countries are struggling to deal with this virus outbreak. Common problem countries face is a lack of medical staff and medical equipment. Machine learning has the potential to play an important role in a different area of medical facilities. With the help of the machine learning model, an effective diagnostic tool can be built which helps in the time of scarcity of medical staff. However medical data is imbalanced and this skew nature of data prevent machine learning algorithm from achieving high accuracy. To deal with this problem of imbalanced data, we proposed a modified decision tree algorithm that uses a minority sensitive Gini index called Mgini. In an imbalanced dataset of COVID-19, it is important to focus on the reduction of overall misclassification cost instead of trying improvement in accuracy value. Mgini is useful splitting criteria when the misclassification cost of the minority sample is huge as compared to the majority class. The use of this proposed new Gini index as a splitting criterion in the decision tree reduces the misclassification cost. Mgini based decision tree has higher accuracy and low misclassification cost as compare to the traditional Gini index based CART algorithm. Our proposed cost-sensitive approach improves imbalanced data classification without the use of data level sampling techniques.
{"title":"Mgini - Improved Decision Tree using Minority Class Sensitive Splitting Criterion for Imbalanced Data of Covid-19","authors":"Pratik A. Barot, H. Jethva","doi":"10.6688/jise.202109_37(5).0008","DOIUrl":"https://doi.org/10.6688/jise.202109_37(5).0008","url":null,"abstract":"In the time of COVID-19, medical facilities struggling to fight against the pandemic. Most of the countries face a tough time fighting against this virus outbreak. Even developed countries are struggling to deal with this virus outbreak. Common problem countries face is a lack of medical staff and medical equipment. Machine learning has the potential to play an important role in a different area of medical facilities. With the help of the machine learning model, an effective diagnostic tool can be built which helps in the time of scarcity of medical staff. However medical data is imbalanced and this skew nature of data prevent machine learning algorithm from achieving high accuracy. To deal with this problem of imbalanced data, we proposed a modified decision tree algorithm that uses a minority sensitive Gini index called Mgini. In an imbalanced dataset of COVID-19, it is important to focus on the reduction of overall misclassification cost instead of trying improvement in accuracy value. Mgini is useful splitting criteria when the misclassification cost of the minority sample is huge as compared to the majority class. The use of this proposed new Gini index as a splitting criterion in the decision tree reduces the misclassification cost. Mgini based decision tree has higher accuracy and low misclassification cost as compare to the traditional Gini index based CART algorithm. Our proposed cost-sensitive approach improves imbalanced data classification without the use of data level sampling techniques.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"22 1","pages":"1097-1108"},"PeriodicalIF":1.1,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81769048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.6688/JISE.20210537(3).0007
Tz-Chia Tseng, Teng-Fu Shih, C. Fuh
Our proposed method is capable of authenticating the input image is from real user or spoofing attack, including paper photograph, digital photograph, and video, using only the Red, Green, Blue (RGB) frontal camera of common smart phone, without the help of depth camera or infrared thermal sensor. We first capture live faces in each frame of input video streams by single shot multi-box detector then feed into our designed convolution neural network after certain data augmentation and finally obtain a well-trained spoof face classifier. Finally, we compared to Parkin and Grinchuk’s results, using dataset CASIASURF[1], and compare the result of vgg16, InceptionNet, ResNet, DenseNet and MobileNet in CASIA-SURFT dataset.
{"title":"Anti-Spoofing of Live Face Authentication on Smartphone","authors":"Tz-Chia Tseng, Teng-Fu Shih, C. Fuh","doi":"10.6688/JISE.20210537(3).0007","DOIUrl":"https://doi.org/10.6688/JISE.20210537(3).0007","url":null,"abstract":"Our proposed method is capable of authenticating the input image is from real user or spoofing attack, including paper photograph, digital photograph, and video, using only the Red, Green, Blue (RGB) frontal camera of common smart phone, without the help of depth camera or infrared thermal sensor. We first capture live faces in each frame of input video streams by single shot multi-box detector then feed into our designed convolution neural network after certain data augmentation and finally obtain a well-trained spoof face classifier. Finally, we compared to Parkin and Grinchuk’s results, using dataset CASIASURF[1], and compare the result of vgg16, InceptionNet, ResNet, DenseNet and MobileNet in CASIA-SURFT dataset.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"17 1","pages":"605-616"},"PeriodicalIF":1.1,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88460251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-01DOI: 10.6688/JISE.202101_37(1).0015
J. Kuo, Hao-Ting Chung, Ping-Feng Wang, Baiying Lei
The deferral of graduation rate in Taiwan's universities is estimated 16%, which will affect the scheduling of school resources. Therefore, if we can expect to take notice of students' academic performance and provide guidance to students who cannot pass the threshold as expected, the waste of school resources can effectively be reduced. In this research, the recent years' student data and course results are used as training data to construct student performance prediction models. The K-Means algorithm was used to classify all courses from the freshman to the senior. The related courses will be grouped in the same cluster, which will more likely to find similar features and improve the accuracy of the prediction. Then, this study constructs independent neural networks for each course according to the different academic year. Each model will be pre-trained by using Denoising Auto-encoder. After pre-training, the corresponding structure and weights are taken as the initial value of the neural network model. Each neural network is treated as a base predictor. All predictors will be integrated into an Ensemble predictor according to different years' weights to predict the current student's course performance. As the students finish the course at the end of each semester, the prediction model will continue track and update to enhance model accuracy through online learning.
{"title":"Building Student Course Performance Prediction Model Based on Deep Learning","authors":"J. Kuo, Hao-Ting Chung, Ping-Feng Wang, Baiying Lei","doi":"10.6688/JISE.202101_37(1).0015","DOIUrl":"https://doi.org/10.6688/JISE.202101_37(1).0015","url":null,"abstract":"The deferral of graduation rate in Taiwan's universities is estimated 16%, which will affect the scheduling of school resources. Therefore, if we can expect to take notice of students' academic performance and provide guidance to students who cannot pass the threshold as expected, the waste of school resources can effectively be reduced. In this research, the recent years' student data and course results are used as training data to construct student performance prediction models. The K-Means algorithm was used to classify all courses from the freshman to the senior. The related courses will be grouped in the same cluster, which will more likely to find similar features and improve the accuracy of the prediction. Then, this study constructs independent neural networks for each course according to the different academic year. Each model will be pre-trained by using Denoising Auto-encoder. After pre-training, the corresponding structure and weights are taken as the initial value of the neural network model. Each neural network is treated as a base predictor. All predictors will be integrated into an Ensemble predictor according to different years' weights to predict the current student's course performance. As the students finish the course at the end of each semester, the prediction model will continue track and update to enhance model accuracy through online learning.","PeriodicalId":50177,"journal":{"name":"Journal of Information Science and Engineering","volume":"238 1","pages":"243-257"},"PeriodicalIF":1.1,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73263857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","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：