Pub Date : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689087
Zuqni Gina Puspita, L. Novamizanti, Ema Rachmawati, Maulin Nasari
Facial emotion is one of the nonverbal interactions in humans that occurs due to facial muscle changes caused by emotional state. For a decade, researchers have conducted research aimed at identifying emotional states. In the education field, students' emotional conditions and their motivation can influence the learning process both directly and indirectly. This paper proposes a facial expression classifier system using characteristic features of Fuzzy Local Binary Pattern (FLBP) and Weber Local Descriptor (WLD). Face detection is carried out in the preprocessing stage using the Viola-Jones algorithm, which cuts the detected faces and resizes them. The characteristic features used in the system are a combination of FLBP and WLD. Then, the classification method uses the Support Vector Machine (SVM). This study aims to facilitate the classification of types of facial expressions, where there are seven facial expressions: disgust, angry, neutral, sad, happy, fear, and surprise. The total data are 203 images, with 133 train data and 70 test data. The combined features of FLBP and WLD provide accuracy, precision, and recall of 92.86% and computation time of 6.19 seconds, respectively. The analysis of multiclass SVM parameters and the performance of each facial expression is also discussed in this paper. Multiclass One-Against-All (OAA) outperforms One-Against-One (OAO).
{"title":"Fuzzy Local Binary Pattern and Weber Local Descriptor for Facial Emotion Classification","authors":"Zuqni Gina Puspita, L. Novamizanti, Ema Rachmawati, Maulin Nasari","doi":"10.1109/ICICyTA53712.2021.9689087","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689087","url":null,"abstract":"Facial emotion is one of the nonverbal interactions in humans that occurs due to facial muscle changes caused by emotional state. For a decade, researchers have conducted research aimed at identifying emotional states. In the education field, students' emotional conditions and their motivation can influence the learning process both directly and indirectly. This paper proposes a facial expression classifier system using characteristic features of Fuzzy Local Binary Pattern (FLBP) and Weber Local Descriptor (WLD). Face detection is carried out in the preprocessing stage using the Viola-Jones algorithm, which cuts the detected faces and resizes them. The characteristic features used in the system are a combination of FLBP and WLD. Then, the classification method uses the Support Vector Machine (SVM). This study aims to facilitate the classification of types of facial expressions, where there are seven facial expressions: disgust, angry, neutral, sad, happy, fear, and surprise. The total data are 203 images, with 133 train data and 70 test data. The combined features of FLBP and WLD provide accuracy, precision, and recall of 92.86% and computation time of 6.19 seconds, respectively. The analysis of multiclass SVM parameters and the performance of each facial expression is also discussed in this paper. Multiclass One-Against-All (OAA) outperforms One-Against-One (OAO).","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127003903","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689130
Fauzan Hikmah Ramadhan, V. Suryani, Satria Mandala
Malware is a malicious program that executes destructive functions to destroy the resources in a computer system, gain some financial benefits, steal the privacy and confidentiality of data, and use computing resources to make a service unavailable in a computer system. One of the ways to prevent malware attacks is by detecting Portable Executable (PE) malware files using machine learning. However, not all machine learning algorithms have optimal performance in detecting a malware PE File because some have several weaknesses that result in low performance in detecting a malware PE File. However, these shortcomings can be reduced by combining two or more two different individual algorithms into one hybrid machine learning algorithm, so the advantages of some individual algorithms can cover the shortcomings of other individual algorithms. Therefore, this research proposes research on the performance of the hybrid machine learning algorithms in detecting malware PE File. The hybrid machine learning algorithms use the voting classifier method and LightGBM, XGBoost, and Logistic Regression as their base model. This research proves that the hybrid machine learning algorithm produces a higher recall value than the ensemble algorithm LightGBM. The hybrid machine learning algorithm produces the highest recall value with a recall value of 99.5026%, while the LightGBM algorithm only produces a recall value of 99.4480%. Furthermore, the recall value of another base model is 99.5004% for the XGBoost algorithm and 98.0539% for the Logistic Regression algorithm.
{"title":"Analysis Study of Malware Classification Portable Executable Using Hybrid Machine Learning","authors":"Fauzan Hikmah Ramadhan, V. Suryani, Satria Mandala","doi":"10.1109/ICICyTA53712.2021.9689130","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689130","url":null,"abstract":"Malware is a malicious program that executes destructive functions to destroy the resources in a computer system, gain some financial benefits, steal the privacy and confidentiality of data, and use computing resources to make a service unavailable in a computer system. One of the ways to prevent malware attacks is by detecting Portable Executable (PE) malware files using machine learning. However, not all machine learning algorithms have optimal performance in detecting a malware PE File because some have several weaknesses that result in low performance in detecting a malware PE File. However, these shortcomings can be reduced by combining two or more two different individual algorithms into one hybrid machine learning algorithm, so the advantages of some individual algorithms can cover the shortcomings of other individual algorithms. Therefore, this research proposes research on the performance of the hybrid machine learning algorithms in detecting malware PE File. The hybrid machine learning algorithms use the voting classifier method and LightGBM, XGBoost, and Logistic Regression as their base model. This research proves that the hybrid machine learning algorithm produces a higher recall value than the ensemble algorithm LightGBM. The hybrid machine learning algorithm produces the highest recall value with a recall value of 99.5026%, while the LightGBM algorithm only produces a recall value of 99.4480%. Furthermore, the recall value of another base model is 99.5004% for the XGBoost algorithm and 98.0539% for the Logistic Regression algorithm.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124372546","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689186
Phua Yeong Tsann, Yew Kwang Hooi, Mohd Fadzil bin Hassan, Matthew Teow Yok Wooi
Application of automatic text summarization is a popular Natural Language Processing task and often used in extracting lengthy content to produce short summary. This is a tedious yet time-consuming task. This study focuses on Malay news articles with the aim to select representative sentences for Malay news headline generation. The dataset used in the experiment is a collection of multi-genre Malay news published between year of 2017 and 2019 from Bernama.com. In this study, a leading sentence approach is applied in the TextRank with TF-IDF and Word2Vec as language models to perform salient sentence extraction. In the experiment, the top-ranking sentences extracted are based on the 15%, 20%, 25% and 30% of the original news content. The extracted contents are evaluation against the original news headline using ROUGE evaluation matric. The model shows that the inclusion of first sentence and first two sentences from the news are able to achieve significant improvement. This leading sentence approach is able to achieve improvement of the F1 score from 1.36 to 7.98. Besides that, the experiment also proofs that the ROUGE scores decrease as the percentage of extraction increase. Thus, the proposed method is fast and resource efficient as compared to other state-of-the-art Natural Language approach.
{"title":"Leading Sentence News TextRank","authors":"Phua Yeong Tsann, Yew Kwang Hooi, Mohd Fadzil bin Hassan, Matthew Teow Yok Wooi","doi":"10.1109/ICICyTA53712.2021.9689186","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689186","url":null,"abstract":"Application of automatic text summarization is a popular Natural Language Processing task and often used in extracting lengthy content to produce short summary. This is a tedious yet time-consuming task. This study focuses on Malay news articles with the aim to select representative sentences for Malay news headline generation. The dataset used in the experiment is a collection of multi-genre Malay news published between year of 2017 and 2019 from Bernama.com. In this study, a leading sentence approach is applied in the TextRank with TF-IDF and Word2Vec as language models to perform salient sentence extraction. In the experiment, the top-ranking sentences extracted are based on the 15%, 20%, 25% and 30% of the original news content. The extracted contents are evaluation against the original news headline using ROUGE evaluation matric. The model shows that the inclusion of first sentence and first two sentences from the news are able to achieve significant improvement. This leading sentence approach is able to achieve improvement of the F1 score from 1.36 to 7.98. Besides that, the experiment also proofs that the ROUGE scores decrease as the percentage of extraction increase. Thus, the proposed method is fast and resource efficient as compared to other state-of-the-art Natural Language approach.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132690559","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689169
Parama Pratummas, Chaiyaporn Khemapatpapan
Office syndrome is one of the important health issues worldwide. Sitting in one position for an extended period of time causes muscle fatigue. This study proposed static fatigue detection in office syndrome using surface electromyography (sEMG) and machine learning. The sEMG was recorded by EMG sensor board connected with NodeMCU V2 ESP8266 during sitting position with surface electrodes on the shoulder. The signals were extracted and preprocessed to obtain different features of datasets. Six machine learning models (Logistic Regression, Support Vector Machine, Naive Bayes, k-nearest Neighbors, Decision Tree, and Multi-layer Perceptron) with seven features (mean, integrated EMG, mean absolute value, mean absolute value1, mean absolute value2, simple square integral, and root mean square) of original datasets and featured-selected data were trained and tested, predicting an output class of fatigue or non-fatigue. Featured-selected data in this research were categorized to feature set I (mean, integrated EMG, mean absolute value, simple square integral, and root mean square) and feature set II (integrated EMG, mean absolute value2, and simple square integral). Consequently, multi-layer perceptron on feature set II has the best accuracy at 99.6690 percent with fit time of 18.322849 seconds. However, considered on the accuracy of 99.2482 percent and the fit time of 0.027955 seconds, decision tree could be an alternate machine learning model in this study.
{"title":"Static Fatigue Detection in Office Syndrome using sEMG and Machine Learning","authors":"Parama Pratummas, Chaiyaporn Khemapatpapan","doi":"10.1109/ICICyTA53712.2021.9689169","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689169","url":null,"abstract":"Office syndrome is one of the important health issues worldwide. Sitting in one position for an extended period of time causes muscle fatigue. This study proposed static fatigue detection in office syndrome using surface electromyography (sEMG) and machine learning. The sEMG was recorded by EMG sensor board connected with NodeMCU V2 ESP8266 during sitting position with surface electrodes on the shoulder. The signals were extracted and preprocessed to obtain different features of datasets. Six machine learning models (Logistic Regression, Support Vector Machine, Naive Bayes, k-nearest Neighbors, Decision Tree, and Multi-layer Perceptron) with seven features (mean, integrated EMG, mean absolute value, mean absolute value1, mean absolute value2, simple square integral, and root mean square) of original datasets and featured-selected data were trained and tested, predicting an output class of fatigue or non-fatigue. Featured-selected data in this research were categorized to feature set I (mean, integrated EMG, mean absolute value, simple square integral, and root mean square) and feature set II (integrated EMG, mean absolute value2, and simple square integral). Consequently, multi-layer perceptron on feature set II has the best accuracy at 99.6690 percent with fit time of 18.322849 seconds. However, considered on the accuracy of 99.2482 percent and the fit time of 0.027955 seconds, decision tree could be an alternate machine learning model in this study.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132894378","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689112
Y. Feng, T. Tang, Eric Tatt Wei Ho
Moderate level of stress is essential to drive an individual towards a specific goal. However, there is growing in prevalence of stress-related illnesses, cognitive and emotional disturbances in developing nations due to increasing task complexity (workload) and disturbances in daily life. Electrodermal activity (EDA) is a non-invasive peripheral index of the sympathetic nervous system that is widely used in psychophysiological studies. Typical EDA data undermined the phasic features that indicates skin conductance responses (SCR) towards stimuli. Here, we attempt deconvolution method to uncover the phasic activity and seek to answer if such features could help us unravel the interacting effects between affective distraction and workload stress. EDA findings showed that participants under the experimental group had heightened SCR when exposed to negative affective stimuli but reduced during cognitive tasks, as compared to neutral control. Although behavioral performance does not reveal significant group differences, negative affective group showed a significant lower SCR expressed by area under curve (AUC) of phasic EDA as compared to neutral control during the highest workload condition (3-back task). We postulate that significant lowered SCR and slight improved performance (accuracy) among negative affective group could indicate intense focus on the most challenging task. Our pilot study shows that phasic EDA is useful to indicate changes in internal states during high workload condition.
{"title":"Phasic Electrodermal Activity Indicates Changes in Workload and Affective States","authors":"Y. Feng, T. Tang, Eric Tatt Wei Ho","doi":"10.1109/ICICyTA53712.2021.9689112","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689112","url":null,"abstract":"Moderate level of stress is essential to drive an individual towards a specific goal. However, there is growing in prevalence of stress-related illnesses, cognitive and emotional disturbances in developing nations due to increasing task complexity (workload) and disturbances in daily life. Electrodermal activity (EDA) is a non-invasive peripheral index of the sympathetic nervous system that is widely used in psychophysiological studies. Typical EDA data undermined the phasic features that indicates skin conductance responses (SCR) towards stimuli. Here, we attempt deconvolution method to uncover the phasic activity and seek to answer if such features could help us unravel the interacting effects between affective distraction and workload stress. EDA findings showed that participants under the experimental group had heightened SCR when exposed to negative affective stimuli but reduced during cognitive tasks, as compared to neutral control. Although behavioral performance does not reveal significant group differences, negative affective group showed a significant lower SCR expressed by area under curve (AUC) of phasic EDA as compared to neutral control during the highest workload condition (3-back task). We postulate that significant lowered SCR and slight improved performance (accuracy) among negative affective group could indicate intense focus on the most challenging task. Our pilot study shows that phasic EDA is useful to indicate changes in internal states during high workload condition.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"349 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130060639","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689119
Wino Rama Putra, Satria Mandala, M. Pramudyo
Valvular Heart Disease (VHD) is a type of heart disease that is triggered by a failure or abnormality in one or more of the four heart valves which results in difficulty in circulating blood between the chambers or blood vessels of the heart. In recent years, many methods have been proposed to detect occurrence of VHD. With advances in technology, to detect these abnormalities can utilize telemedicine technology. The detection method in this paper analyzes the PCG signal (Phonocardiogram) from the patient. The performance value obtained from the detection process is strongly influenced by the algorithm at the feature extraction stage and the feature selection method. Therefore, the selection of the right feature extraction and feature selection method is important. Of the many literatures that propose detection of VHD with the application of feature extraction methods, the average performance obtained is still low. To solve the above problems, this research proposes the development of a feature extraction algorithm that supports the improvement of VHD detection accuracy. In addition, prototypes based on the proposed algorithms and methods were also developed. This research also analyzes the accuracy of the proposed prototype detection. The methods used in this research are 1. Literature study on VHD detection, 2. Development of feature extraction algorithms methods, 3. Performance testing and analysis. The performance test results show that the proposed algorithm has achieved an average accuracy of 99%, sensitivity of 100% and specificity of 97%.
{"title":"Study of Feature Extraction Methods to Detect Valvular Heart Disease (VHD) Using a Phonocardiogram","authors":"Wino Rama Putra, Satria Mandala, M. Pramudyo","doi":"10.1109/ICICyTA53712.2021.9689119","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689119","url":null,"abstract":"Valvular Heart Disease (VHD) is a type of heart disease that is triggered by a failure or abnormality in one or more of the four heart valves which results in difficulty in circulating blood between the chambers or blood vessels of the heart. In recent years, many methods have been proposed to detect occurrence of VHD. With advances in technology, to detect these abnormalities can utilize telemedicine technology. The detection method in this paper analyzes the PCG signal (Phonocardiogram) from the patient. The performance value obtained from the detection process is strongly influenced by the algorithm at the feature extraction stage and the feature selection method. Therefore, the selection of the right feature extraction and feature selection method is important. Of the many literatures that propose detection of VHD with the application of feature extraction methods, the average performance obtained is still low. To solve the above problems, this research proposes the development of a feature extraction algorithm that supports the improvement of VHD detection accuracy. In addition, prototypes based on the proposed algorithms and methods were also developed. This research also analyzes the accuracy of the proposed prototype detection. The methods used in this research are 1. Literature study on VHD detection, 2. Development of feature extraction algorithms methods, 3. Performance testing and analysis. The performance test results show that the proposed algorithm has achieved an average accuracy of 99%, sensitivity of 100% and specificity of 97%.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134243986","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689195
Muhammad Rayhan Ferdinand, Satria Mandala, Dita Oktaria
Vulnerability Scanning is one of the initial stages used in the practice of penetration testing (or pentesting), vulnerability scanning can be said to be a vital process because it can determine how the penetration testing process will be carried out later. The conventional method requires scanning to be done as a whole, which takes a long time and uses a large amount of resources. In this paper, the author proposes a method that applies the Gradient Boosting which is one of a few types from Boosting Algorithm to perform a vulnerability scan based on the port response of the target host. There are only 5 (five) types of ports that being used as a parameters, which all ports have been determined and considered from several books references. And from a several books references itself, it is stated that three of these five ports have a percentage of 65% the most frequent and vulnerable to exploitation activities, these three ports include TCP 22, TCP 80, TCP 443, whereas the two other ports is only an addition to increase exploitation rate percentage which also determined and considered from a book reference, the other two ports is UDP 53, and UDP 80. From the results of tests carried out in 15 times of testing using the CV (or Cross Validation) method, the model built by applying the Gradient Boosting Algorithm gets the results of accuracy, precision, and recall respectively by 98.810%, 98.903%, and 98.812% and with average error rate around 0.00260.
{"title":"Host Vulnerability Analysis Using Supervised Learning Based on Port Response","authors":"Muhammad Rayhan Ferdinand, Satria Mandala, Dita Oktaria","doi":"10.1109/ICICyTA53712.2021.9689195","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689195","url":null,"abstract":"Vulnerability Scanning is one of the initial stages used in the practice of penetration testing (or pentesting), vulnerability scanning can be said to be a vital process because it can determine how the penetration testing process will be carried out later. The conventional method requires scanning to be done as a whole, which takes a long time and uses a large amount of resources. In this paper, the author proposes a method that applies the Gradient Boosting which is one of a few types from Boosting Algorithm to perform a vulnerability scan based on the port response of the target host. There are only 5 (five) types of ports that being used as a parameters, which all ports have been determined and considered from several books references. And from a several books references itself, it is stated that three of these five ports have a percentage of 65% the most frequent and vulnerable to exploitation activities, these three ports include TCP 22, TCP 80, TCP 443, whereas the two other ports is only an addition to increase exploitation rate percentage which also determined and considered from a book reference, the other two ports is UDP 53, and UDP 80. From the results of tests carried out in 15 times of testing using the CV (or Cross Validation) method, the model built by applying the Gradient Boosting Algorithm gets the results of accuracy, precision, and recall respectively by 98.810%, 98.903%, and 98.812% and with average error rate around 0.00260.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125118357","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689127
Sakina Asna Dewi, H. Nuha, S. Mugitama, Rahmat Yasirandi
Clay moisture expresses the amount of water in the material. Clay that is dry or too moist yield failure in manufacturing earthenware. Therefore, we developed an Internet of Things (IoT) device that can detect the level of moisture in the clay. The device consists of clay moisture sensor, Liquid Crystal Display (LCD), Arduino Nano, and Relay Module. The condition of the clay can be seen on the LCD which is installed and connected to the tool. To evaluate the developed system, we conducted an experiment to observe the humidity of two different clay materials where one of them is mixed with additional water for 10 hours. The device is shown to be able to display the difference of the materials. The device is also able to determine the dry or wet status of the material. Once the material is detected to be dry, the device will pour water to the material. The developed device is able to aid craftsmen to maintain the quality of the clay for their crafts.
{"title":"Internet of Things Device for Clay Moisture Measurement","authors":"Sakina Asna Dewi, H. Nuha, S. Mugitama, Rahmat Yasirandi","doi":"10.1109/ICICyTA53712.2021.9689127","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689127","url":null,"abstract":"Clay moisture expresses the amount of water in the material. Clay that is dry or too moist yield failure in manufacturing earthenware. Therefore, we developed an Internet of Things (IoT) device that can detect the level of moisture in the clay. The device consists of clay moisture sensor, Liquid Crystal Display (LCD), Arduino Nano, and Relay Module. The condition of the clay can be seen on the LCD which is installed and connected to the tool. To evaluate the developed system, we conducted an experiment to observe the humidity of two different clay materials where one of them is mixed with additional water for 10 hours. The device is shown to be able to display the difference of the materials. The device is also able to determine the dry or wet status of the material. Once the material is detected to be dry, the device will pour water to the material. The developed device is able to aid craftsmen to maintain the quality of the clay for their crafts.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124151459","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689209
A. Sadiq, N. Yahya, T. Tang
The resting-state functional magnetic resonance imaging (rs-fMRI) is a non-invasive neuroimaging modality to measure brain activity and helps in the diagnosis of various brain-related disorders. Given the 1/f power spectrum characteristic of brain dynamics, where the energy value is higher at a low frequency than high frequency, it is established that low-frequency oscillations (LFO) provide a better representation of the spontaneous neuronal activity of the brain. In this research, a combination of the amplitude of low-frequency fluctuation (ALFF) and fractional ALFF (fALFF) from the resting-state blood oxygen level-dependent (BOLD) signal in the classic band i.e., 0.01-0.1 Hz is used for the classification of Alzheimer's disease (AD) from normal controls (NC). A total of 60 subjects participated in this study consisting of 30 AD patients and 30 NC from Alzheimer's disease neuroimaging initiative (ADNI). The feature selection is performed using minimum-redundancy maximum-relevance (mRMR) and ReliefF algorithm due to the large dimension of rs-fMRI data to be fed to the machine learning (ML) classifier. The proposed AD classification method employing the fusion of ALFF and fALFF obtained the highest classification accuracy of 96.36%, indicating the good potential of the proposed method for the diagnosis of AD, as well as other neurological conditions.
{"title":"Classification of Alzheimer's Disease using Low Frequency Fluctuation of rs-fMRI Signals","authors":"A. Sadiq, N. Yahya, T. Tang","doi":"10.1109/ICICyTA53712.2021.9689209","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689209","url":null,"abstract":"The resting-state functional magnetic resonance imaging (rs-fMRI) is a non-invasive neuroimaging modality to measure brain activity and helps in the diagnosis of various brain-related disorders. Given the 1/f power spectrum characteristic of brain dynamics, where the energy value is higher at a low frequency than high frequency, it is established that low-frequency oscillations (LFO) provide a better representation of the spontaneous neuronal activity of the brain. In this research, a combination of the amplitude of low-frequency fluctuation (ALFF) and fractional ALFF (fALFF) from the resting-state blood oxygen level-dependent (BOLD) signal in the classic band i.e., 0.01-0.1 Hz is used for the classification of Alzheimer's disease (AD) from normal controls (NC). A total of 60 subjects participated in this study consisting of 30 AD patients and 30 NC from Alzheimer's disease neuroimaging initiative (ADNI). The feature selection is performed using minimum-redundancy maximum-relevance (mRMR) and ReliefF algorithm due to the large dimension of rs-fMRI data to be fed to the machine learning (ML) classifier. The proposed AD classification method employing the fusion of ALFF and fALFF obtained the highest classification accuracy of 96.36%, indicating the good potential of the proposed method for the diagnosis of AD, as well as other neurological conditions.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126401687","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 : 2021-12-01DOI: 10.1109/ICICyTA53712.2021.9689160
Muhammad Raafi'u Firmansyah, Risanuri Hidayat, Agus Bejo
The speaker identification system is built by two main blocks; the first part is used to extract features from the input, while the second part is to classify the results from the features in the first part. Selection of the method to perform feature extraction is very important to obtain the optimal feature set. Mel-frequency cepstral coefficients (MFCC) is a feature extraction method that is used to convert the speaker's voice into coefficients as input for the classification process. There are several processes in MFCC, one of which is windowing. Windowing aims to reduce the discontinuous effect on the signal after the framing process. It is therefore important to use optimal windowing techniques so that the features of each sound are not wasted. This article highlights the use of several window functions such as hanning, hamming, bartlett, blackman, kaiser, and gaussian. The classification process proposed in this study is Artificial neural network (ANN). The data used amounted to 800 data from 16 speakers who were recorded directly. The data recorded for identification was the sound from the digits zero to nine (0-9) by each speaker. K-fold cross-validation was used as an evaluation of the classification model created to determine the combination with the best accuracy. The results shows that the use of 13 MFCC features with windowing hamming and gaussian with standard deviation values 72 obtains the best results. Both obtained an accuracy of 95%. This paper helps readers to gain insight in the field of speaker identification.
{"title":"Comparison of Windowing Function on Feature Extraction Using MFCC for Speaker Identification","authors":"Muhammad Raafi'u Firmansyah, Risanuri Hidayat, Agus Bejo","doi":"10.1109/ICICyTA53712.2021.9689160","DOIUrl":"https://doi.org/10.1109/ICICyTA53712.2021.9689160","url":null,"abstract":"The speaker identification system is built by two main blocks; the first part is used to extract features from the input, while the second part is to classify the results from the features in the first part. Selection of the method to perform feature extraction is very important to obtain the optimal feature set. Mel-frequency cepstral coefficients (MFCC) is a feature extraction method that is used to convert the speaker's voice into coefficients as input for the classification process. There are several processes in MFCC, one of which is windowing. Windowing aims to reduce the discontinuous effect on the signal after the framing process. It is therefore important to use optimal windowing techniques so that the features of each sound are not wasted. This article highlights the use of several window functions such as hanning, hamming, bartlett, blackman, kaiser, and gaussian. The classification process proposed in this study is Artificial neural network (ANN). The data used amounted to 800 data from 16 speakers who were recorded directly. The data recorded for identification was the sound from the digits zero to nine (0-9) by each speaker. K-fold cross-validation was used as an evaluation of the classification model created to determine the combination with the best accuracy. The results shows that the use of 13 MFCC features with windowing hamming and gaussian with standard deviation values 72 obtains the best results. Both obtained an accuracy of 95%. This paper helps readers to gain insight in the field of speaker identification.","PeriodicalId":448148,"journal":{"name":"2021 International Conference on Intelligent Cybernetics Technology & Applications (ICICyTA)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131245182","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：