Pub Date : 2019-10-01DOI: 10.1109/ICICoS48119.2019.8982426
M. U. Siregar, Sayekti Abriani
Verification of a rule-based expert system ensures that the knowledge base of the expert system is logically correct and consistent. Application of verification into a rule-based expert system is one approach to integrate software engineering methodology and knowledge base system. The expert system, which we has built, is a rule-based system developed by using forward chaining method and Dempster-Shafer theory of belief functions or evidence. We use Z language as the modelling language for this expert system and SAL model checker as the verification tool. To be able to use SAL model checker, Z2SAL will translate the Z specification, which models the system. In this paper, we present some parts of our Z specification that represent some parts of our rule-based expert system. We also present some parts of our SAL specification and theorems that we added to this SAL specification. At the last, we present the usage of SAL model checker over these theorems. Based on these model-checking processes, we argue that the results are expected. This means that each of theorems can be model checked and the outputs of those model checking are the same as the outputs that we obtain from manual investigation; either it is VALID or INVALID. Other interpretation of the model check's results is some parts of our rule-based expert system have been verified.
{"title":"Verification of a Rule-Based Expert System by Using SAL Model Checker","authors":"M. U. Siregar, Sayekti Abriani","doi":"10.1109/ICICoS48119.2019.8982426","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982426","url":null,"abstract":"Verification of a rule-based expert system ensures that the knowledge base of the expert system is logically correct and consistent. Application of verification into a rule-based expert system is one approach to integrate software engineering methodology and knowledge base system. The expert system, which we has built, is a rule-based system developed by using forward chaining method and Dempster-Shafer theory of belief functions or evidence. We use Z language as the modelling language for this expert system and SAL model checker as the verification tool. To be able to use SAL model checker, Z2SAL will translate the Z specification, which models the system. In this paper, we present some parts of our Z specification that represent some parts of our rule-based expert system. We also present some parts of our SAL specification and theorems that we added to this SAL specification. At the last, we present the usage of SAL model checker over these theorems. Based on these model-checking processes, we argue that the results are expected. This means that each of theorems can be model checked and the outputs of those model checking are the same as the outputs that we obtain from manual investigation; either it is VALID or INVALID. Other interpretation of the model check's results is some parts of our rule-based expert system have been verified.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"11 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":"128538991","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/ICICoS48119.2019.8982415
W. Hadikurniawati, Edy Winarno, D. Santoso, Purwatiningtyas
Determination of the selection of food crops on a suitable land planted based on the characteristics of the land is very important for decision makers. A proposed model that aggregated weight of parameters and determined the best alternative using an AHP and TOPSIS mixed method. Priority weights for parameters are calculated using the AHP method and then making a sequence alternative uses the TOPSIS method. The appliance of the planned model can help users in deciding the most suitable food crops to be planted on certain land according to the characteristics of the land. Based on calculations using the AHP and TOPSIS mix methods the highest priority results obtained from the alternative. The highest priority alternative to the consideration of 11 parameters is green beans. Ranking in this application depends on the choice of preference type and determination of parameter thresholds. Proposed method can solve the problem of determining food crops that are suitable for planting in dry land.
{"title":"A Mixed Method using AHP-TOPSIS for Dryland Agriculture Crops Selection Problem","authors":"W. Hadikurniawati, Edy Winarno, D. Santoso, Purwatiningtyas","doi":"10.1109/ICICoS48119.2019.8982415","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982415","url":null,"abstract":"Determination of the selection of food crops on a suitable land planted based on the characteristics of the land is very important for decision makers. A proposed model that aggregated weight of parameters and determined the best alternative using an AHP and TOPSIS mixed method. Priority weights for parameters are calculated using the AHP method and then making a sequence alternative uses the TOPSIS method. The appliance of the planned model can help users in deciding the most suitable food crops to be planted on certain land according to the characteristics of the land. Based on calculations using the AHP and TOPSIS mix methods the highest priority results obtained from the alternative. The highest priority alternative to the consideration of 11 parameters is green beans. Ranking in this application depends on the choice of preference type and determination of parameter thresholds. Proposed method can solve the problem of determining food crops that are suitable for planting in dry land.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"1 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":"130759852","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/ICICoS48119.2019.8982502
Sholiq Sholiq, Pandu Satrio Hutomo, A. D. Wulandari, A. P. Subriadi, Anisah Herdiyanti, Eko Wahyu Tyas Darmaningrat
This research continues previous research by testing the owner estimate cost (OEC) model in Android-based application development projects. Previously, the model of OEC had been tested with nine general software project data which produced acceptable accuracy. In this study, the same model was tested with 5 android-based software projects to find out the consistency of the model with test data that was different from before. To test the accuracy of the results using the magnitude of relative error (MRE) for each software's owner estimate cost and mean of MRE (MMRE). The test results with 5 android-based applications show that this model is also consistent and has an acceptable level of accuracy.
{"title":"Testing of Owner Estimate Cost Model with Android-based Application","authors":"Sholiq Sholiq, Pandu Satrio Hutomo, A. D. Wulandari, A. P. Subriadi, Anisah Herdiyanti, Eko Wahyu Tyas Darmaningrat","doi":"10.1109/ICICoS48119.2019.8982502","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982502","url":null,"abstract":"This research continues previous research by testing the owner estimate cost (OEC) model in Android-based application development projects. Previously, the model of OEC had been tested with nine general software project data which produced acceptable accuracy. In this study, the same model was tested with 5 android-based software projects to find out the consistency of the model with test data that was different from before. To test the accuracy of the results using the magnitude of relative error (MRE) for each software's owner estimate cost and mean of MRE (MMRE). The test results with 5 android-based applications show that this model is also consistent and has an acceptable level of accuracy.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"1 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":"123171960","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}
Finding a parking space is a tedious and time-consuming task in a metropolitan city. Due to this problem, many researchers proposed an automatic parking lot occupancy detection system using a camera with a deep learning method to provide useful information in the smart city system. Since object detection for the parking lot is performed in real-time by utilizing CPU and GPUs while parking detection is working 24 hours a day and 365 days a year, therefore power saving is important to reduce the electricity cost. However, the energy-aware is not considered in most related works. In this paper, we proposed an energy-saving algorithm for parking lot availability detection using YOLO running on the TX2 machine. We experiment using small parking lot prototype and remote control cars. In the experiment, we compare our algorithm with the direct application of original YOLO for parking lot detection, the results show that it reduces power by 97 percent when there is no moving object in the parking lot area and 71 percent when there are moving objects in the parking lot area.
{"title":"Energy Aware Parking Lot Availability Detection Using YOLO on TX2","authors":"Yohan Marvel Anggawijaya, Tien-Hsiung Weng, Rosita Herawati","doi":"10.1109/ICICoS48119.2019.8982448","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982448","url":null,"abstract":"Finding a parking space is a tedious and time-consuming task in a metropolitan city. Due to this problem, many researchers proposed an automatic parking lot occupancy detection system using a camera with a deep learning method to provide useful information in the smart city system. Since object detection for the parking lot is performed in real-time by utilizing CPU and GPUs while parking detection is working 24 hours a day and 365 days a year, therefore power saving is important to reduce the electricity cost. However, the energy-aware is not considered in most related works. In this paper, we proposed an energy-saving algorithm for parking lot availability detection using YOLO running on the TX2 machine. We experiment using small parking lot prototype and remote control cars. In the experiment, we compare our algorithm with the direct application of original YOLO for parking lot detection, the results show that it reduces power by 97 percent when there is no moving object in the parking lot area and 71 percent when there are moving objects in the parking lot area.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"72 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":"122417376","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/ICICoS48119.2019.8982431
Hafiz Marham, R. Saputra
This study was about analysis acceptance of Mobile-Online Games (M-OG) and proposed a model which is the result of integration between Unified Theory of Acceptance and Usage of Technology (UTAUT) and Game Design. Analysis of technology acceptance concerning to Mobile Online-Games (M-OG) was done by assessing the influence of the enjoyment factor that caused to users when playing games. The study was conducted for a month, from July 15 to August 20, 2019 in Java and Sumatera. In this study also collected 215 data from respondents who participated. To test the relationship between latent variables and indicators also hypotheses designed, this studied was used Partial Least Square - Structural Equation Modeling (PLS-SEM). The results showed the perceived enjoyment variable had a significant influence on user behavior in the continuance intention of playing M-OG. In addition, the performance expectancy variable also has a significant influence on user behavior to continuance in playing M-OG. Variables that also very significantly influence the user's enjoyment in playing M-OG are challenge, novelty, and effort expectancy. In this study also found variables that do not have a significant influence on the continuance of the user playing M-OG, these variables are social influence and facilitating conditions. The design aesthetic variable was also found to have no significant influence on the enjoyment that users get when playing M-OG.
{"title":"User Continuance in Playing Mobile Online Games Analyzed by Using UTAUT and Game Design","authors":"Hafiz Marham, R. Saputra","doi":"10.1109/ICICoS48119.2019.8982431","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982431","url":null,"abstract":"This study was about analysis acceptance of Mobile-Online Games (M-OG) and proposed a model which is the result of integration between Unified Theory of Acceptance and Usage of Technology (UTAUT) and Game Design. Analysis of technology acceptance concerning to Mobile Online-Games (M-OG) was done by assessing the influence of the enjoyment factor that caused to users when playing games. The study was conducted for a month, from July 15 to August 20, 2019 in Java and Sumatera. In this study also collected 215 data from respondents who participated. To test the relationship between latent variables and indicators also hypotheses designed, this studied was used Partial Least Square - Structural Equation Modeling (PLS-SEM). The results showed the perceived enjoyment variable had a significant influence on user behavior in the continuance intention of playing M-OG. In addition, the performance expectancy variable also has a significant influence on user behavior to continuance in playing M-OG. Variables that also very significantly influence the user's enjoyment in playing M-OG are challenge, novelty, and effort expectancy. In this study also found variables that do not have a significant influence on the continuance of the user playing M-OG, these variables are social influence and facilitating conditions. The design aesthetic variable was also found to have no significant influence on the enjoyment that users get when playing M-OG.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"1 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":"131281548","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/ICICoS48119.2019.8982446
Aisha Adetia, Peny Rishartati, Sari Agustin Wulandari, D. I. Sensuse, Sofian Lusa, P. Prima, R. C. Handayani
The capacity that needs to be improved in a woman is one of her main tasks as a mother who educates her children. The Professional Mother Institute (IIP) is a virtual community that empowers women to become professional mothers. IIP has expanded the community to various cities and countries, and has also won many awards. But in the last batch there has been decline significantly in the numbers of members along with applying code of conduct in an effort for IIP members can always participate actively in knowledge sharing. Beside that, as far as we know, no one has discussed the factors that influence knowledge sharing in a virtual community that empowers mothers. Therefore, researchers are encouraged to explore what factors influence knowledge sharing at IIP which uses the Social Networking Sites (SNS) as one of them whatsapp. This research can contribute to IIP by providing input on how IIP makes members keep active in sharing knowledge and keep on increasing. Other than that, this certainly can be a learning for other virtual communities. The method of collecting data in this study used an online questionnaire that was distributed to IIP members of the class “Bunda Sayang”. This data collection received a response of 115 valid respondents. This data then processed with PLS SEM. The results of this study are perceived reciprocal benefit factors, Perceived enjoyment, Perceived status and Outcome expectation have a positive effect on knowledge sharing in whatsapp group IIP members. Future research can raise other factors or see the impact of knowledge sharing by women in virtual communities, especially those related to community empowerment.
{"title":"Factors Influence Knowledge Sharing Through Social Networking Site Case Study: Virtual Community Institut Ibu Profesional (IIP)","authors":"Aisha Adetia, Peny Rishartati, Sari Agustin Wulandari, D. I. Sensuse, Sofian Lusa, P. Prima, R. C. Handayani","doi":"10.1109/ICICoS48119.2019.8982446","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982446","url":null,"abstract":"The capacity that needs to be improved in a woman is one of her main tasks as a mother who educates her children. The Professional Mother Institute (IIP) is a virtual community that empowers women to become professional mothers. IIP has expanded the community to various cities and countries, and has also won many awards. But in the last batch there has been decline significantly in the numbers of members along with applying code of conduct in an effort for IIP members can always participate actively in knowledge sharing. Beside that, as far as we know, no one has discussed the factors that influence knowledge sharing in a virtual community that empowers mothers. Therefore, researchers are encouraged to explore what factors influence knowledge sharing at IIP which uses the Social Networking Sites (SNS) as one of them whatsapp. This research can contribute to IIP by providing input on how IIP makes members keep active in sharing knowledge and keep on increasing. Other than that, this certainly can be a learning for other virtual communities. The method of collecting data in this study used an online questionnaire that was distributed to IIP members of the class “Bunda Sayang”. This data collection received a response of 115 valid respondents. This data then processed with PLS SEM. The results of this study are perceived reciprocal benefit factors, Perceived enjoyment, Perceived status and Outcome expectation have a positive effect on knowledge sharing in whatsapp group IIP members. Future research can raise other factors or see the impact of knowledge sharing by women in virtual communities, especially those related to community empowerment.","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":"133247616","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/ICICoS48119.2019.8982519
Fika Hastarita Rachman, Riyanarto Samo, C. Fatichah
Arousal and Valence value represent of song emotions. Arousal is an emotional dimension of musically energy level, while Valence is an emotional dimension of the comfortable level of the listener. Label emotion of Thayer using Arousal and Valence dimension. This research proposed a rule base method for detecting song emotion using arousal and valence values, however many studies do not use this data. The datasets are audio and lyric features of the song structural segment chorus. Preprocessing of Audio and lyric data are uses Correlation Feature Selection (CFS) and preprocessing text. Audio feature extraction is using MIRToolbox. Stylistic and psycholinguistic are used for lyrics feature extraction. Rule based method is used to detect the emotions of the whole song by using the predictive feature of the arousal and valence values. The arousal and valence prediction values are representing withmatrices of frequencyfor audio and lyrics. From the analysis of testing data, it shows that the audio feature more represents the value of Valence while the lyrics feature more represents the Arousal value. There are seven (7) rule base models that used in this research, the best accuracy is 0.798.
{"title":"Song Emotion Detection Based on Arousal-Valence from Audio and Lyrics Using Rule Based Method","authors":"Fika Hastarita Rachman, Riyanarto Samo, C. Fatichah","doi":"10.1109/ICICoS48119.2019.8982519","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982519","url":null,"abstract":"Arousal and Valence value represent of song emotions. Arousal is an emotional dimension of musically energy level, while Valence is an emotional dimension of the comfortable level of the listener. Label emotion of Thayer using Arousal and Valence dimension. This research proposed a rule base method for detecting song emotion using arousal and valence values, however many studies do not use this data. The datasets are audio and lyric features of the song structural segment chorus. Preprocessing of Audio and lyric data are uses Correlation Feature Selection (CFS) and preprocessing text. Audio feature extraction is using MIRToolbox. Stylistic and psycholinguistic are used for lyrics feature extraction. Rule based method is used to detect the emotions of the whole song by using the predictive feature of the arousal and valence values. The arousal and valence prediction values are representing withmatrices of frequencyfor audio and lyrics. From the analysis of testing data, it shows that the audio feature more represents the value of Valence while the lyrics feature more represents the Arousal value. There are seven (7) rule base models that used in this research, the best accuracy is 0.798.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"1 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":"130339749","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/ICICoS48119.2019.8982409
H. Al-Ash, Mutia Fadhila Putri, P. Mursanto, A. Bustamam
The news is information about a recently changed situation or a recent event. Serving as popular media information the internet has the power spread the news not only real news but fake news as well. We propose an ensemble learning approach on Indonesian fake news in order to separate fake news from the real one and to tackle imbalanced data problem which we face on the given dataset. Our experiment result shows that random forest classifier as the ensemble classifier which obtained 0.98 f1-score is superior to multinomial naive bayes and support vector machine as non-ensemble classifiers which achieve 0.43 and 0.74 f1-score respectively across 660 evaluation documents. We also compare our result against other research that using the same data and our approach achieved better results.
{"title":"Ensemble Learning Approach on Indonesian Fake News Classification","authors":"H. Al-Ash, Mutia Fadhila Putri, P. Mursanto, A. Bustamam","doi":"10.1109/ICICoS48119.2019.8982409","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982409","url":null,"abstract":"The news is information about a recently changed situation or a recent event. Serving as popular media information the internet has the power spread the news not only real news but fake news as well. We propose an ensemble learning approach on Indonesian fake news in order to separate fake news from the real one and to tackle imbalanced data problem which we face on the given dataset. Our experiment result shows that random forest classifier as the ensemble classifier which obtained 0.98 f1-score is superior to multinomial naive bayes and support vector machine as non-ensemble classifiers which achieve 0.43 and 0.74 f1-score respectively across 660 evaluation documents. We also compare our result against other research that using the same data and our approach achieved better results.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"10 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":"130268526","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/ICICoS48119.2019.8982455
Mawanda Almuhayar, Henry Horng Shing Lu, Nur Iriawan
Deep learning development nowadays has attracted a lot of attention because of its effectiveness and good performance. The performance of deep learning in medical images analysis already can compete with medical image experts. However, there are experts that still believe deep learning only efficient for the big datasets, because of deep learning performance in small datasets still not satisfying enough. In this study, it is aimed to build a deep learning model for image classification that can achieve high accuracy using chest X-ray images with a relatively small dataset. We classify chest X-ray into a binary classification which is a normal image and image with abnormalities. We built and experimented our model using the public dataset of Shenzen Hospital dataset. We also use a different type of input based on different images preprocessing so that the model can perform accurate classification. Based on the result, pre-trained CheXNet with a newly trained fully connected network on the cropped dataset can achieve the accuracy 0.8761, the sensitivity 0.8909, and the specificity 0.8621. The performance of the model also influenced by the certain region inside the images, such as other regions outside the lung region and black colored region outside the body region.
{"title":"Classification of Abnormality in Chest X-Ray Images by Transfer Learning of CheXNet","authors":"Mawanda Almuhayar, Henry Horng Shing Lu, Nur Iriawan","doi":"10.1109/ICICoS48119.2019.8982455","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982455","url":null,"abstract":"Deep learning development nowadays has attracted a lot of attention because of its effectiveness and good performance. The performance of deep learning in medical images analysis already can compete with medical image experts. However, there are experts that still believe deep learning only efficient for the big datasets, because of deep learning performance in small datasets still not satisfying enough. In this study, it is aimed to build a deep learning model for image classification that can achieve high accuracy using chest X-ray images with a relatively small dataset. We classify chest X-ray into a binary classification which is a normal image and image with abnormalities. We built and experimented our model using the public dataset of Shenzen Hospital dataset. We also use a different type of input based on different images preprocessing so that the model can perform accurate classification. Based on the result, pre-trained CheXNet with a newly trained fully connected network on the cropped dataset can achieve the accuracy 0.8761, the sensitivity 0.8909, and the specificity 0.8621. The performance of the model also influenced by the certain region inside the images, such as other regions outside the lung region and black colored region outside the body region.","PeriodicalId":105407,"journal":{"name":"2019 3rd International Conference on Informatics and Computational Sciences (ICICoS)","volume":"78 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":"116365051","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/ICICoS48119.2019.8982470
S. R. Juwita, S. Endah
Music has a variety of genres, namely pop, rock, jazz, and so on. Indonesia has its own music that other countries do not have, including campursari, dangdut, and keroncong music. The three types of music have musical instruments that are almost similar, which makes it difficult for listeners to distinguish the genre of music, especially the younger generation, so we need a tool called classification. This study uses a mel-spectogram and the Convolutional Neural Network (CNN) method to classify Indonesian music. The CNN parameters and architecture tested in this study were batch normalization, ReLU activation, dropout, activation of sigmoid and softmax output, epoch value, learning rate value, and dense layer value. The entire parameter is tested using input with two different data sharing methods, namely stratified split and k-fold cross validation. The highest accuracy of 82% was obtained by using the stratified split data distribution method and using batch normalization parameters, ReLU activation, activation of outputs sigmoid and softmax, 30 epoch values, 0.05 learning rate values, and 200 layer dense values. The model with the highest accuracy value is used as the basis for classifying Indonesian music into campursari, dangdut, or keroncong classes
{"title":"Classification of Indonesian Music Using the Convolutional Neural Network Method","authors":"S. R. Juwita, S. Endah","doi":"10.1109/ICICoS48119.2019.8982470","DOIUrl":"https://doi.org/10.1109/ICICoS48119.2019.8982470","url":null,"abstract":"Music has a variety of genres, namely pop, rock, jazz, and so on. Indonesia has its own music that other countries do not have, including campursari, dangdut, and keroncong music. The three types of music have musical instruments that are almost similar, which makes it difficult for listeners to distinguish the genre of music, especially the younger generation, so we need a tool called classification. This study uses a mel-spectogram and the Convolutional Neural Network (CNN) method to classify Indonesian music. The CNN parameters and architecture tested in this study were batch normalization, ReLU activation, dropout, activation of sigmoid and softmax output, epoch value, learning rate value, and dense layer value. The entire parameter is tested using input with two different data sharing methods, namely stratified split and k-fold cross validation. The highest accuracy of 82% was obtained by using the stratified split data distribution method and using batch normalization parameters, ReLU activation, activation of outputs sigmoid and softmax, 30 epoch values, 0.05 learning rate values, and 200 layer dense values. The model with the highest accuracy value is used as the basis for classifying Indonesian music into campursari, dangdut, or keroncong classes","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":"128973269","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：