Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982383
Lavin J. Halawa, A. Wibowo, F. Ernawan
Detection and prevention of criminal incidents using CCTV are currently increasing trend, for example, car and motorcycle parking lot. However, not continuous people monitoring and careless of events produce useless CCTV function for the prevention of criminal incidents. In this paper, face recognition is used for the recognition of vehicle owners in parking lots that are CCTV installed. The Faster-RCNN method is used for face detection and also for face recognition. Inception V2 architecture is utilized due to has a high accuracy among Convolutional Neural Network architecture. The best learning rate and epoch parameters for the Faster R-CNN model are optimized to improve face recognition on CCTV. In this research, the dataset consists of 6 people images with 50 faces images for each people, which used as training data, testing data, and validation data.
{"title":"Face Recognition Using Faster R-CNN with Inception-V2 Architecture for CCTV Camera","authors":"Lavin J. Halawa, A. Wibowo, F. Ernawan","doi":"10.1109/ICICoS48119.2019.8982383","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982383","url":null,"abstract":"Detection and prevention of criminal incidents using CCTV are currently increasing trend, for example, car and motorcycle parking lot. However, not continuous people monitoring and careless of events produce useless CCTV function for the prevention of criminal incidents. In this paper, face recognition is used for the recognition of vehicle owners in parking lots that are CCTV installed. The Faster-RCNN method is used for face detection and also for face recognition. Inception V2 architecture is utilized due to has a high accuracy among Convolutional Neural Network architecture. The best learning rate and epoch parameters for the Faster R-CNN model are optimized to improve face recognition on CCTV. In this research, the dataset consists of 6 people images with 50 faces images for each people, which used as training data, testing data, and validation data.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132852460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982394
Noor Ifada, M. K. Sophan, Irvan Syachrudin, Selgy Zahranida Sugiharto
This paper investigates several schemes to combine the user demographic information and item attribute data that respectively beneficial to solve the data sparsity and cold-start problems in recommendation systems. We propose four schemes that are varied based on how the combination of the two data can be constructed. To test and evaluate the concept, we implement the schemes on a probabilistic-attribute method adapted to suit our attribute model. Compared to the benchmark methods, experiment results show that our approach is superior in solving the data sparsity and cold-start problems. In general, the scheme that combines the item attribute data with a partial user demographic information performs better than the other variations of the combined-attribute scheme. This finding confirms that combining both the user demographic information, though not all of them, and the item attribute can efficiently solve the data sparsity and cold-start problems.
{"title":"An Efficient Scheme to Combine the User Demographics and Item Attribute for Solving Data Sparsity and Cold-start Problems","authors":"Noor Ifada, M. K. Sophan, Irvan Syachrudin, Selgy Zahranida Sugiharto","doi":"10.1109/ICICoS48119.2019.8982394","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982394","url":null,"abstract":"This paper investigates several schemes to combine the user demographic information and item attribute data that respectively beneficial to solve the data sparsity and cold-start problems in recommendation systems. We propose four schemes that are varied based on how the combination of the two data can be constructed. To test and evaluate the concept, we implement the schemes on a probabilistic-attribute method adapted to suit our attribute model. Compared to the benchmark methods, experiment results show that our approach is superior in solving the data sparsity and cold-start problems. In general, the scheme that combines the item attribute data with a partial user demographic information performs better than the other variations of the combined-attribute scheme. This finding confirms that combining both the user demographic information, though not all of them, and the item attribute can efficiently solve the data sparsity and cold-start problems.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133715150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982400
Afifah Aprilia Ayuningtyas, R. Kusumaningrum
The history of Indonesia which is quite long causes difficulty for our people in obtaining information about the history of Indonesia. In order to obtain information, people still need to seek from many books or documents on the history of Indonesia. Such a way is considered less efficient, thus a question answering system is considered necessary so that the information can be obtained quickly and efficiently. Questions on the topic of history have a tendency on the factoid question type so the type of question in this research is factoid. This research uses the Dynamic Memory Networks (DMN) model to obtain answers to the given questions. The parameter of the tested DMN model is learning rate, iteration, and episodes. This study uses 0.0005; 0.005; 0.05 as the value of learning rate, 1563; 3125; 6250 as the value of the number of iteration, and 3, 4, 5 as the value of the number of episodes. The dataset used in this research is 500 questions with a context in the form of single sentences and 500 questions with a context in the form of compound sentences which are taken from Wikipedia. The highest accuracy results are obtained by using the learning rate value of 0.005, iteration of 6250, and episodes of 5 on the dataset with the context in the form of single sentences amounted to 56% whereas the dataset with the context in the form of compound sentences amounted to 38.6%.
{"title":"The Question Answering System of Indonesia's History Using Dynamic Memory Networks (DMN) Model","authors":"Afifah Aprilia Ayuningtyas, R. Kusumaningrum","doi":"10.1109/ICICoS48119.2019.8982400","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982400","url":null,"abstract":"The history of Indonesia which is quite long causes difficulty for our people in obtaining information about the history of Indonesia. In order to obtain information, people still need to seek from many books or documents on the history of Indonesia. Such a way is considered less efficient, thus a question answering system is considered necessary so that the information can be obtained quickly and efficiently. Questions on the topic of history have a tendency on the factoid question type so the type of question in this research is factoid. This research uses the Dynamic Memory Networks (DMN) model to obtain answers to the given questions. The parameter of the tested DMN model is learning rate, iteration, and episodes. This study uses 0.0005; 0.005; 0.05 as the value of learning rate, 1563; 3125; 6250 as the value of the number of iteration, and 3, 4, 5 as the value of the number of episodes. The dataset used in this research is 500 questions with a context in the form of single sentences and 500 questions with a context in the form of compound sentences which are taken from Wikipedia. The highest accuracy results are obtained by using the learning rate value of 0.005, iteration of 6250, and episodes of 5 on the dataset with the context in the form of single sentences amounted to 56% whereas the dataset with the context in the form of compound sentences amounted to 38.6%.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133826915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982494
Vicky Zilvan, A. Ramdan, Endang Suryawati, R. B. S. Kusumo, Dikdik Krisnandi, H. Pardede
Early detection is critical for maintaining quantity and quality of farming commodity. Currently, detection of plant diseases still requires human expertise and/or need microscopic identification such as spectroscopic technique and molecular biological. So, it would be very costly and time consuming, and hence unattainable for small-holder farmers. The rapid development of intelligent agriculture using machine learning has led the widespread use of computer or smart-phones to solve this problem. So, early detection of plant disease can be performed with minimal support from human experts and microscopic identification is no longer needed. However, conventional machine-learning techniques are limited in their ability to process raw data directly. So it require some efforts and domain expertise to design feature extractor to support it. Moreover, impulse noise such as salt-pepper noise may present on the images and it arises another challenge to provide a robust system. In this paper, we present denoising convolutional variational autoencoders as automatic unsupervised feature extractor and automatic denoiser to learn and to extract good features directly from the raw data. Here, we use the output of denoising convolutional variational auto encoders as inputs to fully connected networks classifiers for automatic detection of plant diseases. Our experiments show the average accuracies of our method is better than denoising variational autoencoders which is built using fully deep connected networks architectures. We also found that our proposed method is more robust against noisy test data.
{"title":"Denoising Convolutional Variational Autoencoders-Based Feature Learning for Automatic Detection of Plant Diseases","authors":"Vicky Zilvan, A. Ramdan, Endang Suryawati, R. B. S. Kusumo, Dikdik Krisnandi, H. Pardede","doi":"10.1109/ICICoS48119.2019.8982494","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982494","url":null,"abstract":"Early detection is critical for maintaining quantity and quality of farming commodity. Currently, detection of plant diseases still requires human expertise and/or need microscopic identification such as spectroscopic technique and molecular biological. So, it would be very costly and time consuming, and hence unattainable for small-holder farmers. The rapid development of intelligent agriculture using machine learning has led the widespread use of computer or smart-phones to solve this problem. So, early detection of plant disease can be performed with minimal support from human experts and microscopic identification is no longer needed. However, conventional machine-learning techniques are limited in their ability to process raw data directly. So it require some efforts and domain expertise to design feature extractor to support it. Moreover, impulse noise such as salt-pepper noise may present on the images and it arises another challenge to provide a robust system. In this paper, we present denoising convolutional variational autoencoders as automatic unsupervised feature extractor and automatic denoiser to learn and to extract good features directly from the raw data. Here, we use the output of denoising convolutional variational auto encoders as inputs to fully connected networks classifiers for automatic detection of plant diseases. Our experiments show the average accuracies of our method is better than denoising variational autoencoders which is built using fully deep connected networks architectures. We also found that our proposed method is more robust against noisy test data.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131328023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982384
Phyllalintang Nafasa, I. Waspada, N. Bahtiar, A. Wibowo
Moodle is one of the widely used Learning Management Systems in the field of education. Moodle stores all online learning activities to the database in the form of event log. These event logs can be used to improve the quality of learning through process analysis. One of the fields of science that can be used to discover the process model based on event log is Process Mining. The problem arise when an instructor willing to use the Moodle event log data to do a Process Mining activities. There are some preprocessing issues need to be done to the Moodle event log data as prerequisite to continue with Process Mining algorithm. As the solution, Moodle need to be integrated with the Process Mining. In this study an application was developed to integrate the Moodle event log data with the activities of Process Mining, especially to facilitate the preprocessing tools. The alpha miner algorithm was used here as the process model discovery algorithm. As the result, we successfully develop the application to discover process model from Moodle log event data. Instructors can use some functional features of the application to meet their need in process mining analysis. Experiments using real and artificial case studies have been conducted and it is proven that the implementation of the alpha miner algorithm can work correctly on the Moodle event log data.
{"title":"Implementation of Alpha Miner Algorithm in Process Mining Application Development for Online Learning Activities Based on MOODLE Event Log Data","authors":"Phyllalintang Nafasa, I. Waspada, N. Bahtiar, A. Wibowo","doi":"10.1109/ICICoS48119.2019.8982384","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982384","url":null,"abstract":"Moodle is one of the widely used Learning Management Systems in the field of education. Moodle stores all online learning activities to the database in the form of event log. These event logs can be used to improve the quality of learning through process analysis. One of the fields of science that can be used to discover the process model based on event log is Process Mining. The problem arise when an instructor willing to use the Moodle event log data to do a Process Mining activities. There are some preprocessing issues need to be done to the Moodle event log data as prerequisite to continue with Process Mining algorithm. As the solution, Moodle need to be integrated with the Process Mining. In this study an application was developed to integrate the Moodle event log data with the activities of Process Mining, especially to facilitate the preprocessing tools. The alpha miner algorithm was used here as the process model discovery algorithm. As the result, we successfully develop the application to discover process model from Moodle log event data. Instructors can use some functional features of the application to meet their need in process mining analysis. Experiments using real and artificial case studies have been conducted and it is proven that the implementation of the alpha miner algorithm can work correctly on the Moodle event log data.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"163 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116601644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982418
Rizqi Hadi Prawira, A. Wibowo, Ajif Yunizar Pratama Yusuf
Arrhythmia are disturbances in the heart where the heart beats slower or faster. Some types of Arrhythmia can became a serious problem and life-threatening. Early detection of Arrhythmia is very crucial to patients. Tools that can be used to determine heart condition is Electrocardiogram (ECG). Deep learning methods can be used to classify types of Arrhythmia from ECG images. Convolutional Neural Network is one of deep learning methods that is often used to classify images. CNN-based model such as VGG, ResNet, and MobileNet has gotten success in images classification. Those models are using lots of convolution layer, so those models are easily run into over fitting problem if those are used in small dataset. CNN model in this research needs parameter adjustments to solve over fitting problem. Parameter that were being adjusted were learning rate, dropout rate, and the number of convolution layer. The testing results on CNN model showed that the best learning rate and dropout rate which produced the best model to classify Arrhythmia were 0.0001, and 0.0075 respectively. The number of convolution layers which obtained the best accuracy was 4. Classification using CNN model for Arrhythmia with learning rate, dropout rate, and number of convolution layers were 0.0001, 0.0075, and 4 respectively resulted in the best model with 94.2 % accuracy value.
{"title":"Best Parameters Selection of Arrhythmia Classification Using Convolutional Neural Networks","authors":"Rizqi Hadi Prawira, A. Wibowo, Ajif Yunizar Pratama Yusuf","doi":"10.1109/ICICoS48119.2019.8982418","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982418","url":null,"abstract":"Arrhythmia are disturbances in the heart where the heart beats slower or faster. Some types of Arrhythmia can became a serious problem and life-threatening. Early detection of Arrhythmia is very crucial to patients. Tools that can be used to determine heart condition is Electrocardiogram (ECG). Deep learning methods can be used to classify types of Arrhythmia from ECG images. Convolutional Neural Network is one of deep learning methods that is often used to classify images. CNN-based model such as VGG, ResNet, and MobileNet has gotten success in images classification. Those models are using lots of convolution layer, so those models are easily run into over fitting problem if those are used in small dataset. CNN model in this research needs parameter adjustments to solve over fitting problem. Parameter that were being adjusted were learning rate, dropout rate, and the number of convolution layer. The testing results on CNN model showed that the best learning rate and dropout rate which produced the best model to classify Arrhythmia were 0.0001, and 0.0075 respectively. The number of convolution layers which obtained the best accuracy was 4. Classification using CNN model for Arrhythmia with learning rate, dropout rate, and number of convolution layers were 0.0001, 0.0075, and 4 respectively resulted in the best model with 94.2 % accuracy value.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"85 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123350095","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982516
Ronaldo Kristianto, Farida Dwi Handayani, A. Wibowo
Molecular biology-based tests are widely used to monitor various activities, such as molecular interaction dynamics, cell health, and in other health studies. At present molecular biology detection technology is widely available in city center laboratories, but this does not happen in small clinics and in remote areas. For this reason, a method called point of care (POC) was developed, which is a medical diagnostic test near a place of care that can provide fast results. Fluorescence is one method of labeling samples that are widely used in point of care activities. Recent research has detected fluorescence with quite good results, but the detection done is mostly based on RGB color space without regard to wavelength. In fact, wavelength is an important factor in fluorescence detection where using wavelength, the detection results can show the level of intensity of the light produced by the fluorescence sample. In this research, the curve fitting function is created which can convert the RGB value in an image or image to a wavelength value. From 3 fitting curves with RGB, HSV, and hue data, the function with the smallest mean squared error and the smallest root mean squared error will be selected. Next, using the best fitting curve function will read the wavelength value of a fluorescence sample photo. The results of this experiment show that the combination of the use of the fitting curve function obtained from HSV data and the fitting curve obtained from hue produces the most optimal error results, with a mean squared error (MSE) value of 367,373, compared to the MSE results of the RGB fitting curve with value 3908.1, HSV fitting curve with a value of 593.6, and hue fitting curve which is worth 1456.62.
{"title":"Selecting the Function of Color Space Conversion RGB / HSL to Wavelength for Fluorescence Intensity Measurement on Android Based Applications","authors":"Ronaldo Kristianto, Farida Dwi Handayani, A. Wibowo","doi":"10.1109/ICICoS48119.2019.8982516","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982516","url":null,"abstract":"Molecular biology-based tests are widely used to monitor various activities, such as molecular interaction dynamics, cell health, and in other health studies. At present molecular biology detection technology is widely available in city center laboratories, but this does not happen in small clinics and in remote areas. For this reason, a method called point of care (POC) was developed, which is a medical diagnostic test near a place of care that can provide fast results. Fluorescence is one method of labeling samples that are widely used in point of care activities. Recent research has detected fluorescence with quite good results, but the detection done is mostly based on RGB color space without regard to wavelength. In fact, wavelength is an important factor in fluorescence detection where using wavelength, the detection results can show the level of intensity of the light produced by the fluorescence sample. In this research, the curve fitting function is created which can convert the RGB value in an image or image to a wavelength value. From 3 fitting curves with RGB, HSV, and hue data, the function with the smallest mean squared error and the smallest root mean squared error will be selected. Next, using the best fitting curve function will read the wavelength value of a fluorescence sample photo. The results of this experiment show that the combination of the use of the fitting curve function obtained from HSV data and the fitting curve obtained from hue produces the most optimal error results, with a mean squared error (MSE) value of 367,373, compared to the MSE results of the RGB fitting curve with value 3908.1, HSV fitting curve with a value of 593.6, and hue fitting curve which is worth 1456.62.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"191 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121736259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-08-01DOI: 10.1109/icicos48119.2019.8982500
W. Jatmiko, Kurnianingsih
{"title":"Welcome Message from IEEE Indonesia Section","authors":"W. Jatmiko, Kurnianingsih","doi":"10.1109/icicos48119.2019.8982500","DOIUrl":"https://doi.org/10.1109/icicos48119.2019.8982500","url":null,"abstract":"","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123770601","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：