Pub Date : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034619
D. Ratnawati, Marjono, Widodo, S. Anam
The classification of the active compound based on their function is important to be done because most of them are unknown their function. The structure of the active compound can be represented by SMILES code that unique, compact and complete. Preprocessing SMILES code is a crucial task before SMILES codes are classified. In this research, preprocessing is extracting SMILES code into several features. The features must represent patterns or information from SMILES codes because the proper features of the SMILES codes will increase the accuracy of classification results. This paper uses features from SMILES codes directed by an expert and based on the previous research. Features will be normalized and are classified by an efficient and good classification method, Extreme Learning Machine (ELM). The experiment results show that first, adding features will increase the average of the accuracy of the system until 10.9% on dataset 1-3-4 (nerve-bacterial-cancer). Second, ELM is superior to SVM and KMNB in terms of both accuracy and processing time.
{"title":"Features Selection for Classification of SMILES Codes Based on Their Function","authors":"D. Ratnawati, Marjono, Widodo, S. Anam","doi":"10.1109/ISRITI48646.2019.9034619","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034619","url":null,"abstract":"The classification of the active compound based on their function is important to be done because most of them are unknown their function. The structure of the active compound can be represented by SMILES code that unique, compact and complete. Preprocessing SMILES code is a crucial task before SMILES codes are classified. In this research, preprocessing is extracting SMILES code into several features. The features must represent patterns or information from SMILES codes because the proper features of the SMILES codes will increase the accuracy of classification results. This paper uses features from SMILES codes directed by an expert and based on the previous research. Features will be normalized and are classified by an efficient and good classification method, Extreme Learning Machine (ELM). The experiment results show that first, adding features will increase the average of the accuracy of the system until 10.9% on dataset 1-3-4 (nerve-bacterial-cancer). Second, ELM is superior to SVM and KMNB in terms of both accuracy and processing time.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126634757","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034615
Aidil Fajar Zulfahri, Catur Edi Widodo, R. Gernowo
Understanding student perceptions and levels of satisfaction with the services provided are one way for colleges to increase their competitive advantages. Most of the colleges have established a quality control system to evaluate services that have been provided to students. However, the service evaluation questionnaire only measures the level of performance on each service attribute provided. While the colleges have limited resources, so it would be better if prioritizing improvements to service attributes that are rated important from the student's perspective. Importance-performance analysis (IPA) can be used to resolve this problem. IPA provides valid information on which products or services should be prioritized for immediate improvement. The results of this evaluation will provide a clear sight of which service attributes are already good, and which service attributes are still not good. Implementing IPA for measuring student satisfaction levels provide an overview of service attributes that are very important based on the students' perspective. So the college can improve the quality of services that are assessed most important to increase student satisfaction in the future.
{"title":"Implementing Importance-Performance Analysis (IPA) for Measuring Students Satisfaction Levels","authors":"Aidil Fajar Zulfahri, Catur Edi Widodo, R. Gernowo","doi":"10.1109/ISRITI48646.2019.9034615","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034615","url":null,"abstract":"Understanding student perceptions and levels of satisfaction with the services provided are one way for colleges to increase their competitive advantages. Most of the colleges have established a quality control system to evaluate services that have been provided to students. However, the service evaluation questionnaire only measures the level of performance on each service attribute provided. While the colleges have limited resources, so it would be better if prioritizing improvements to service attributes that are rated important from the student's perspective. Importance-performance analysis (IPA) can be used to resolve this problem. IPA provides valid information on which products or services should be prioritized for immediate improvement. The results of this evaluation will provide a clear sight of which service attributes are already good, and which service attributes are still not good. Implementing IPA for measuring student satisfaction levels provide an overview of service attributes that are very important based on the students' perspective. So the college can improve the quality of services that are assessed most important to increase student satisfaction in the future.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124893273","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034622
Mutiara Auliya Khadija, S. Widyawan, Ir. Lukito Edi Nugroho
The internet technology has grown rapidly, which may cause problems in computer network systems. Intrusion Detection System (IDS) has been introduced for detecting attacks in that system. Using artificial intelligence, Intrusion Detection System will able to recognize anomalies or signatures of attacks. For some years, research has focused on the data mining method for detecting the anomaly to increase the accuracy of classification results. To obtain high accuracy, required several stages of data preparation and feature selection. It because the feature selection are not correlated to the dataset and not accordance with requirement of the classification process. In this research, we perform anomaly detection on IDS using combination of DBSCAN, Principle Component Analysis (PCA) and Ranker with classification method. For evaluation, we employ Kyoto 2006, NSL-KDD 99 and KDD Cup 99 as datasets. It is found that this preprocessing step increases the accuracy, when it is applied to Naïve Bayes, Random Forest and k-NN methods. Specifically, the highest increase of accuracy is achieved by Naïve Bayes Classification method with KDD Cup 1999 which gains 6.11%.
互联网技术发展迅速，这可能会给计算机网络系统带来问题。引入入侵检测系统(IDS)来检测系统中的攻击。利用人工智能，入侵检测系统将能够识别异常或攻击的特征。为了提高分类结果的准确性，近年来研究的重点是利用数据挖掘方法检测异常。为了获得较高的精度，需要经过几个阶段的数据准备和特征选择。这是因为特征选择与数据集不相关，不符合分类过程的要求。在本研究中，我们使用DBSCAN、主成分分析(PCA)和Ranker结合分类方法对IDS进行异常检测。为了进行评估，我们使用京都2006、NSL-KDD 99和KDD Cup 99作为数据集。当将此预处理步骤应用于Naïve贝叶斯、随机森林和k-NN方法时，发现该预处理步骤提高了精度。其中，使用KDD Cup 1999的Naïve贝叶斯分类方法准确率提高最高，提高了6.11%。
{"title":"Detecting Network Intrusion by Combining DBSCAN, Principle Component Analysis and Ranker","authors":"Mutiara Auliya Khadija, S. Widyawan, Ir. Lukito Edi Nugroho","doi":"10.1109/ISRITI48646.2019.9034622","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034622","url":null,"abstract":"The internet technology has grown rapidly, which may cause problems in computer network systems. Intrusion Detection System (IDS) has been introduced for detecting attacks in that system. Using artificial intelligence, Intrusion Detection System will able to recognize anomalies or signatures of attacks. For some years, research has focused on the data mining method for detecting the anomaly to increase the accuracy of classification results. To obtain high accuracy, required several stages of data preparation and feature selection. It because the feature selection are not correlated to the dataset and not accordance with requirement of the classification process. In this research, we perform anomaly detection on IDS using combination of DBSCAN, Principle Component Analysis (PCA) and Ranker with classification method. For evaluation, we employ Kyoto 2006, NSL-KDD 99 and KDD Cup 99 as datasets. It is found that this preprocessing step increases the accuracy, when it is applied to Naïve Bayes, Random Forest and k-NN methods. Specifically, the highest increase of accuracy is achieved by Naïve Bayes Classification method with KDD Cup 1999 which gains 6.11%.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124440059","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034605
Naufal Praditya, A. H. Saputro
Machine learning has been a big help to perform specific tasks by learning the data and improving the results. The way this system works was based on pattern recognition and computational algorithms. Some classification algorithms go through a process called feature selection or dimensionality reduction. This process was used to minimize the number of features used. In this study, the process was observed through a hyperspectral image to identify beeswax on Rome Beauty apples and to define the essential variables on the wavelengths. The hyperspectral image was acquired on a wavelength ranging from 400 to 1000 nm. The spatial and spectral data of the image can be obtained through this technique. Thus the reflectance profile from the object was used to classify the nonwaxed apple and the waxed apple based on the variable importance. Compared to the accuracy of the support vector machine model, the accuracy of the decision tree model shows a better outcome with 81.25% correct predictions from 48 testing data. In the decision tree model, there are 13 essential variables on 13 features (wavelength) that was used by the classifier to get the best result.
{"title":"Hyperspectral Band Selection based on Decision Tree Algorithm in Beeswax Identification on Rome Beauty Apple","authors":"Naufal Praditya, A. H. Saputro","doi":"10.1109/ISRITI48646.2019.9034605","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034605","url":null,"abstract":"Machine learning has been a big help to perform specific tasks by learning the data and improving the results. The way this system works was based on pattern recognition and computational algorithms. Some classification algorithms go through a process called feature selection or dimensionality reduction. This process was used to minimize the number of features used. In this study, the process was observed through a hyperspectral image to identify beeswax on Rome Beauty apples and to define the essential variables on the wavelengths. The hyperspectral image was acquired on a wavelength ranging from 400 to 1000 nm. The spatial and spectral data of the image can be obtained through this technique. Thus the reflectance profile from the object was used to classify the nonwaxed apple and the waxed apple based on the variable importance. Compared to the accuracy of the support vector machine model, the accuracy of the decision tree model shows a better outcome with 81.25% correct predictions from 48 testing data. In the decision tree model, there are 13 essential variables on 13 features (wavelength) that was used by the classifier to get the best result.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122287475","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034629
Andre Rabiula, S. Sulistyo, R. Ferdiana
The main challenges of e-commerce is to attract and maintain a healthy relationship with users, because users simply interact with software. Development of E-commerce in Indonesia within the last four years rapidly increased up to 500%. The growth of e-commerce has made testings on e-commerce important. Search engine is one of the most important features in e-commerce. This study conducts testing on e-commerce’s search engines using metamorphic testing. This study proposes MR_Sorting and MR_SubsetEquality. The proposed metamorphic relation is a part of a feature provided by e-commerce. The testing is implemented on three major e-commerce websites used in Indonesia, which is Lazada, Tokopedia, and Blibli. Based on the proposed metamorphic relation, conducted testing discovered an inconsistency problem which could affect customer satisfaction in using e-commerce websites. It is assumed the problem occur on source data in which the feature’s implementation wasn’t applied on the original source data requested when a user conduct searches on search engines and a product’s low description convention on e-commerces. Proposed metamorphic relation in this study can be used on other e-commerce websites. It is expected upon future studies to conduct similar studies on more e-commerce websites so the maturity of the proposed metamorphic improves and to conduct deeper studies on discovered inconsistencies in influencing user satisfaction and the declining quality of e-commerce websites.
{"title":"Applying Metamorphic Testing to E-commerce Website Search Engines","authors":"Andre Rabiula, S. Sulistyo, R. Ferdiana","doi":"10.1109/ISRITI48646.2019.9034629","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034629","url":null,"abstract":"The main challenges of e-commerce is to attract and maintain a healthy relationship with users, because users simply interact with software. Development of E-commerce in Indonesia within the last four years rapidly increased up to 500%. The growth of e-commerce has made testings on e-commerce important. Search engine is one of the most important features in e-commerce. This study conducts testing on e-commerce’s search engines using metamorphic testing. This study proposes MR_Sorting and MR_SubsetEquality. The proposed metamorphic relation is a part of a feature provided by e-commerce. The testing is implemented on three major e-commerce websites used in Indonesia, which is Lazada, Tokopedia, and Blibli. Based on the proposed metamorphic relation, conducted testing discovered an inconsistency problem which could affect customer satisfaction in using e-commerce websites. It is assumed the problem occur on source data in which the feature’s implementation wasn’t applied on the original source data requested when a user conduct searches on search engines and a product’s low description convention on e-commerces. Proposed metamorphic relation in this study can be used on other e-commerce websites. It is expected upon future studies to conduct similar studies on more e-commerce websites so the maturity of the proposed metamorphic improves and to conduct deeper studies on discovered inconsistencies in influencing user satisfaction and the declining quality of e-commerce websites.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"14 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120884155","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034588
LN Hamaningrum, A. Subagyo
The development of wireless technology will reach the fifth generation (5G). If this condition occurs, then there will be many objects that can interconnect and exchange data. Embedded system is one area that will play many roles. One of them is the use of Near Field Communication (NFC). NFC is a short-range communication tool that has been used in many applications. NFC is one of the payment methods that will continue to be used by consumers. The development of this use must be accompanied by increasing trust. In practice, NFC has three modes: peer to peer, read / write and card emulation mode. In the state of card emulation mode, NFC mounted on a smartphone starts to bring up the Host Card Emulation which is launched on Android Kitkat 4.4. This condition allows the possibilies of interruptions or attacks. Then it needs to be examined the system that runs whether there is a gap or not. This study analyzed this with the response time parameters and the distance of the device. Card emulation conditions on NFC-enabled mobile with HCE show that communication can be done in combination with distance and the results show that distance does not affect response time. As long as it is still in the reader range, NFC-enabled mobile can still be read.
{"title":"Response Time Analysis to Support Business Process of the NFC Communication in Smartphone","authors":"LN Hamaningrum, A. Subagyo","doi":"10.1109/ISRITI48646.2019.9034588","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034588","url":null,"abstract":"The development of wireless technology will reach the fifth generation (5G). If this condition occurs, then there will be many objects that can interconnect and exchange data. Embedded system is one area that will play many roles. One of them is the use of Near Field Communication (NFC). NFC is a short-range communication tool that has been used in many applications. NFC is one of the payment methods that will continue to be used by consumers. The development of this use must be accompanied by increasing trust. In practice, NFC has three modes: peer to peer, read / write and card emulation mode. In the state of card emulation mode, NFC mounted on a smartphone starts to bring up the Host Card Emulation which is launched on Android Kitkat 4.4. This condition allows the possibilies of interruptions or attacks. Then it needs to be examined the system that runs whether there is a gap or not. This study analyzed this with the response time parameters and the distance of the device. Card emulation conditions on NFC-enabled mobile with HCE show that communication can be done in combination with distance and the results show that distance does not affect response time. As long as it is still in the reader range, NFC-enabled mobile can still be read.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"6 30","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120930195","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 : 2019-12-01DOI: 10.1109/isriti48646.2019.9034586
{"title":"ISRITI 2019 Paper Titles","authors":"","doi":"10.1109/isriti48646.2019.9034586","DOIUrl":"https://doi.org/10.1109/isriti48646.2019.9034586","url":null,"abstract":"","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129650085","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034617
Willy Jordan, Agus Bejo, A. G. Persada
At this time, there are many products based on embedded systems, each of these products requires an embedded processor. This results in high demand for embedded processors. Therefore, we need a tool that is able to design an embedded processor completely with its software development tool. One of the embedded processor that is developed is GAMA32 processor. GAMA32 processor is a 32-bit embedded processor designed using System C Model. Its processor architecture is General Purpose Processor (GPP). GAMA32 processor has its own software development tool, but it has no compiler. The most common used compiler is open source compiler like GNU Compiler Collection (GCC) and Low Level Virtual Machine (LLVM). However, this two open source compilers are not compatible with GAMA32 processor because their complexity. In this research, a simple and flexible compiler will be developed, especially the Abstract Syntax Tree (AST) generator part using Python library rPLY and the AST’s output is in the format of JavaScript Object Notation (JSON). For testing the result of AST generator, 5 statement patterns are used namely if, if-else, while, do-while, and for patterns with initial value, and data type declaration as well. The test results show that the correct AST can be produced from the patterns used.
{"title":"The Development of Lexer and Parser as Parts of Compiler for GAMA32 Processor’s Instruction-set using Python","authors":"Willy Jordan, Agus Bejo, A. G. Persada","doi":"10.1109/ISRITI48646.2019.9034617","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034617","url":null,"abstract":"At this time, there are many products based on embedded systems, each of these products requires an embedded processor. This results in high demand for embedded processors. Therefore, we need a tool that is able to design an embedded processor completely with its software development tool. One of the embedded processor that is developed is GAMA32 processor. GAMA32 processor is a 32-bit embedded processor designed using System C Model. Its processor architecture is General Purpose Processor (GPP). GAMA32 processor has its own software development tool, but it has no compiler. The most common used compiler is open source compiler like GNU Compiler Collection (GCC) and Low Level Virtual Machine (LLVM). However, this two open source compilers are not compatible with GAMA32 processor because their complexity. In this research, a simple and flexible compiler will be developed, especially the Abstract Syntax Tree (AST) generator part using Python library rPLY and the AST’s output is in the format of JavaScript Object Notation (JSON). For testing the result of AST generator, 5 statement patterns are used namely if, if-else, while, do-while, and for patterns with initial value, and data type declaration as well. The test results show that the correct AST can be produced from the patterns used.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132723367","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 : 2019-12-01DOI: 10.1109/isriti48646.2019.9034610
{"title":"ISRITI 2019 Committees","authors":"","doi":"10.1109/isriti48646.2019.9034610","DOIUrl":"https://doi.org/10.1109/isriti48646.2019.9034610","url":null,"abstract":"","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115076914","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 : 2019-12-01DOI: 10.1109/ISRITI48646.2019.9034590
Ananto Tri Sasongko, M. Ivan Fanany
Research in vehicle classification through various methods has become a popular field of study for decades. Mainly, vehicles are categorized based on the model, manufactured, logo, types, and dimensions, and the dataset for it is available publicly and relatively easy to get. However, based on our survey, vehicle classification based on the number of axles using deep learning has not been conducted, and the public dataset for it is not available yet. This paper aims to compose a vehicle classification based on type and number of axles then categorize it into five groups, namely, Group-I, Group-II, Group-III, Group-IV, and Group-V. This vehicle grouping refers to the Indonesia toll road tariff regulation. Nowadays, deep learning as one of the most advanced methods becomes the preferred technique to apply in image classifications due to its high performance, so do this study. Utilizing Convolution Neural Networks (CNN) as image segmentation and classification, Transfer Learning as a technique, Resnet architectures as base models, and fine-tuning as an enhancement, we can achieve accuracy about 99% for the specific vehicle classification in this study.
{"title":"Indonesia Toll Road Vehicle Classification Using Transfer Learning with Pre-trained Resnet Models","authors":"Ananto Tri Sasongko, M. Ivan Fanany","doi":"10.1109/ISRITI48646.2019.9034590","DOIUrl":"https://doi.org/10.1109/ISRITI48646.2019.9034590","url":null,"abstract":"Research in vehicle classification through various methods has become a popular field of study for decades. Mainly, vehicles are categorized based on the model, manufactured, logo, types, and dimensions, and the dataset for it is available publicly and relatively easy to get. However, based on our survey, vehicle classification based on the number of axles using deep learning has not been conducted, and the public dataset for it is not available yet. This paper aims to compose a vehicle classification based on type and number of axles then categorize it into five groups, namely, Group-I, Group-II, Group-III, Group-IV, and Group-V. This vehicle grouping refers to the Indonesia toll road tariff regulation. Nowadays, deep learning as one of the most advanced methods becomes the preferred technique to apply in image classifications due to its high performance, so do this study. Utilizing Convolution Neural Networks (CNN) as image segmentation and classification, Transfer Learning as a technique, Resnet architectures as base models, and fine-tuning as an enhancement, we can achieve accuracy about 99% for the specific vehicle classification in this study.","PeriodicalId":367363,"journal":{"name":"2019 International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131868758","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：