Information Systems Development as an information provider, both for the users and outsiders, becomes very important at this time. Through a good data management, it can minimize the need for human labor that is currently being replaced by computers. This paper will discuss a design for a blood transfusion information system because there are several issues found in the Blood Transfusion Service Unit (UPTD), some of which are donor data recording at the Blood Transfusion Service Unit which still use manual recording with a book, donors data stored in the book is not well organized, and there are blood donors who do not obey the blood donor regulation, reminder for blood donor level is 70 days and the screening result. This design can do managements which are needed when people want to donor their blood. The database in this system was designed using S-Designor, while the model design was designed using Rational Rose.
{"title":"Blood Transfusion Information System Design for Blood Transfusion Services Unit","authors":"Yulius Harjoseputro","doi":"10.22146/IJITEE.43702","DOIUrl":"https://doi.org/10.22146/IJITEE.43702","url":null,"abstract":"Information Systems Development as an information provider, both for the users and outsiders, becomes very important at this time. Through a good data management, it can minimize the need for human labor that is currently being replaced by computers. This paper will discuss a design for a blood transfusion information system because there are several issues found in the Blood Transfusion Service Unit (UPTD), some of which are donor data recording at the Blood Transfusion Service Unit which still use manual recording with a book, donors data stored in the book is not well organized, and there are blood donors who do not obey the blood donor regulation, reminder for blood donor level is 70 days and the screening result. This design can do managements which are needed when people want to donor their blood. The database in this system was designed using S-Designor, while the model design was designed using Rational Rose.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114736293","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}
Naufal Azmi Verdikha, T. B. Adji, A. E. Permanasari
A text classification system is needed to address the problem of hate speech in social media. However, texts of hate speech are very hard to find in social media. This will make the distribution of training data to be unbalanced (imbalanced data). Classification with imbalanced data will make a poor performance. There are several methods to solve the problem of classification with imbalanced data. One of them is undersampling with Instance Hardness Threshold (IHT) method. IHT method balances the dataset by eliminating data that are frequently misclassified. To find those data, IHT requires an estimator, which is a classifier. This research aims to compare estimators of IHT method to solve imbalanced data problem in hate speech classification using TF-IDF weighting method. This research uses the class ratio of dataset after undersampling, time of the undersampling process, and Index of Balanced Accuracy (IBA) evaluation to determine the best IHT method. The results of this research show that IHT method using the Logistic Regression (IHT(LR)) has the fastest undersampling process (1.91 s), perfectly balance dataset with the class ratio is 1:1, and has the best of IBA evaluation in all estimation process. This result makes IHT(LR) be the best method to solve the imbalanced data problem in hate speech classification.
{"title":"Study of Undersampling Method: Instance Hardness Threshold with Various Estimators for Hate Speech Classification","authors":"Naufal Azmi Verdikha, T. B. Adji, A. E. Permanasari","doi":"10.22146/IJITEE.42152","DOIUrl":"https://doi.org/10.22146/IJITEE.42152","url":null,"abstract":"A text classification system is needed to address the problem of hate speech in social media. However, texts of hate speech are very hard to find in social media. This will make the distribution of training data to be unbalanced (imbalanced data). Classification with imbalanced data will make a poor performance. There are several methods to solve the problem of classification with imbalanced data. One of them is undersampling with Instance Hardness Threshold (IHT) method. IHT method balances the dataset by eliminating data that are frequently misclassified. To find those data, IHT requires an estimator, which is a classifier. This research aims to compare estimators of IHT method to solve imbalanced data problem in hate speech classification using TF-IDF weighting method. This research uses the class ratio of dataset after undersampling, time of the undersampling process, and Index of Balanced Accuracy (IBA) evaluation to determine the best IHT method. The results of this research show that IHT method using the Logistic Regression (IHT(LR)) has the fastest undersampling process (1.91 s), perfectly balance dataset with the class ratio is 1:1, and has the best of IBA evaluation in all estimation process. This result makes IHT(LR) be the best method to solve the imbalanced data problem in hate speech classification.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133412201","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}
Institute of Business (IOB) is committed to achieve its goals, i.e. becomes a place for economics and business as well as computer science developers, and prepare a ready-to-use Human Resources (HR), especially for Timor Leste. At present, IOB does not have an alignment between business processes and information systems owned. Therefore, this paper proposes an architectural design that bridges the alignment. The methods used to build the framework are include Enterprise Architecture Planning (EAP), SWOT, Value Chain, and Mc Farlan Grid. The built frameworks are focused on the needs of the application architecture. The resulted portfolio has 45 applications for various divisions of IOB. In addition, the SWOT analysis shows that IOB's internal and external factors are in the second quadrant. Thus, IOB's position is relatively strong, though it is facing a big challenge. The recommended strategy is the Strengths-Threats (ST) strategy that uses the company’s strength to overcome the threats. This strategy includes improving facilities, adding faculties and departments, developing academic information systems, improving the quality of learning, and improving the human resources quality.
{"title":"The Design of Application Architecture of the Institute of Business Based on Enterprise Architecture Planning","authors":"Muhammad Yusuf Morais, Habibullah Akbar","doi":"10.22146/IJITEE.42149","DOIUrl":"https://doi.org/10.22146/IJITEE.42149","url":null,"abstract":"Institute of Business (IOB) is committed to achieve its goals, i.e. becomes a place for economics and business as well as computer science developers, and prepare a ready-to-use Human Resources (HR), especially for Timor Leste. At present, IOB does not have an alignment between business processes and information systems owned. Therefore, this paper proposes an architectural design that bridges the alignment. The methods used to build the framework are include Enterprise Architecture Planning (EAP), SWOT, Value Chain, and Mc Farlan Grid. The built frameworks are focused on the needs of the application architecture. The resulted portfolio has 45 applications for various divisions of IOB. In addition, the SWOT analysis shows that IOB's internal and external factors are in the second quadrant. Thus, IOB's position is relatively strong, though it is facing a big challenge. The recommended strategy is the Strengths-Threats (ST) strategy that uses the company’s strength to overcome the threats. This strategy includes improving facilities, adding faculties and departments, developing academic information systems, improving the quality of learning, and improving the human resources quality.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126899537","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}
The purpose of this research is to design the application of digital image processing system to identify the image of medicinal plants of Gorontalo region using artificial neural network method using back propagation. This research used a digital image processing method with segmentation and extraction techniques. Segmentation process was carried out using thresholding method. Furthermore, a process of characteristic extraction from medicinal plants drawings was carried out using feature and color feature extractions to obtain the value of metric, eccentricity, hue, saturation and value. these five values were used as parameters for input neurons and one output neuron which denoted the class of the medicinal plants image. Data of this research consisted of 91 images which had been divided into two types, training data and test data. The training data consisted of 80 images and the test data consisted of eleven images. A network architecture was obtained from the training result and it provided the highest accuracy level (100%) and least number of iteration with a number of 50 neurons on hidden layer and 143 epochs. The testing result showed a lower accuracy of 54.54%.
{"title":"Gorontalo Medicinal Plants Image Identification System Using Artificial Neural Network with Back Propagation","authors":"M. Latief, R. Yusuf","doi":"10.22146/IJITEE.42154","DOIUrl":"https://doi.org/10.22146/IJITEE.42154","url":null,"abstract":"The purpose of this research is to design the application of digital image processing system to identify the image of medicinal plants of Gorontalo region using artificial neural network method using back propagation. This research used a digital image processing method with segmentation and extraction techniques. Segmentation process was carried out using thresholding method. Furthermore, a process of characteristic extraction from medicinal plants drawings was carried out using feature and color feature extractions to obtain the value of metric, eccentricity, hue, saturation and value. these five values were used as parameters for input neurons and one output neuron which denoted the class of the medicinal plants image. Data of this research consisted of 91 images which had been divided into two types, training data and test data. The training data consisted of 80 images and the test data consisted of eleven images. A network architecture was obtained from the training result and it provided the highest accuracy level (100%) and least number of iteration with a number of 50 neurons on hidden layer and 143 epochs. The testing result showed a lower accuracy of 54.54%.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133941295","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}
Electricity demand increase as function of population and economic activity growth. To meet the demand growth, one kind of approaches to expand electrical system is to calculate the need of generating unit and the result will be used to determine the needs of transmission line. In this research, a model was developed with focused on transmission line expansion based on Mix Integer Linear Programming method. The objective function was to minimize overall investment cost for transmission and operating cost of all generating units. The developed model was implemented in 6-bus Garver’s test system. Distributed generation implementation impact is also studied in this study in term of network configuration and overall expansion cost. The results show that distributed generation implementation will differ the network configuration and reduce the overall system cost, with overall system cost with and without distributed generation implementation was $106.4 million and $103.18 million respectively.
{"title":"Role Analysis of Distributed Generation Towards Transmission Expansion Planning Using MILP","authors":"Gessa Firman Febrian, S. P. Hadi, S. Sarjiya","doi":"10.22146/IJITEE.42551","DOIUrl":"https://doi.org/10.22146/IJITEE.42551","url":null,"abstract":"Electricity demand increase as function of population and economic activity growth. To meet the demand growth, one kind of approaches to expand electrical system is to calculate the need of generating unit and the result will be used to determine the needs of transmission line. In this research, a model was developed with focused on transmission line expansion based on Mix Integer Linear Programming method. The objective function was to minimize overall investment cost for transmission and operating cost of all generating units. The developed model was implemented in 6-bus Garver’s test system. Distributed generation implementation impact is also studied in this study in term of network configuration and overall expansion cost. The results show that distributed generation implementation will differ the network configuration and reduce the overall system cost, with overall system cost with and without distributed generation implementation was $106.4 million and $103.18 million respectively.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128419909","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}
One of the causes of smog as well as climate damage, particularly in Palangka Raya, Center Kalimantan, are peat forest fires. There are a lot of losses inflicted by the smog including the increasing number of people who suffer respiratory infection (ARI) due to polluted air and any other related aspects. Peat fires are problematic to overcome because the locations of fires are difficult to be accessed. This paper focuses on building the system to predict the distribution of peat forest fire hotspots by utilizing satellite imagery. In designing the system for predicting the fire hotspots distribution, wavelet orthogonal was used as the initial processing of mapping the distribution of peat forest fire hotspots. Meanwhile, backpropagation method was used to identify the fire hotspot distribution patterns of peat forest fire in this system. From the result of the data tested which had been done for predicting the peat forest fire hotspots, the decomposition image obtained using Haar wavelet had the highest percentage of accuracy to recognize the fire hotspots, which is 90%. The recency of this system was its ability to predict the peat forest fire hotspots distribution which can be used as peat forest fires prevention, especially in Palangka Raya, Central Kalimantan.
{"title":"Prediction of Peat Forest Fires Using Wavelet and Backpropagation","authors":"Novera Kristianti, A. Santoso, P. Pranowo","doi":"10.22146/ijitee.42156","DOIUrl":"https://doi.org/10.22146/ijitee.42156","url":null,"abstract":"One of the causes of smog as well as climate damage, particularly in Palangka Raya, Center Kalimantan, are peat forest fires. There are a lot of losses inflicted by the smog including the increasing number of people who suffer respiratory infection (ARI) due to polluted air and any other related aspects. Peat fires are problematic to overcome because the locations of fires are difficult to be accessed. This paper focuses on building the system to predict the distribution of peat forest fire hotspots by utilizing satellite imagery. In designing the system for predicting the fire hotspots distribution, wavelet orthogonal was used as the initial processing of mapping the distribution of peat forest fire hotspots. Meanwhile, backpropagation method was used to identify the fire hotspot distribution patterns of peat forest fire in this system. From the result of the data tested which had been done for predicting the peat forest fire hotspots, the decomposition image obtained using Haar wavelet had the highest percentage of accuracy to recognize the fire hotspots, which is 90%. The recency of this system was its ability to predict the peat forest fire hotspots distribution which can be used as peat forest fires prevention, especially in Palangka Raya, Central Kalimantan.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133068680","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}
Interleaved boost converter gives good conversion efficiency due to its zero-current switching capability when operating in discontinuous conduction mode while keeping its input-output ripple current low. However, operating this kind of converter at interleaved operation for all the time gives poor efficiency under light-load condition. In this paper, an automatic operation region selector switch based on detection of the continuous or discontinuous current mode is proposed. With this switch, during the light-load condition, only one converter is activated, while during full-load condition both converters will be activated. The simulation results using LTspice software show that the proposed boost converter has a better efficiency compared to the conventional boost converter with efficiency range of 84.6 % to 95.32 % under various load conditions.
{"title":"Operation Region Selector Circuit to Obtain Maximum Efficiency of 250 W Boost Converter","authors":"R. R. O. Sasue, E. Firmansyah, S. Suharyanto","doi":"10.22146/IJITEE.36649","DOIUrl":"https://doi.org/10.22146/IJITEE.36649","url":null,"abstract":"Interleaved boost converter gives good conversion efficiency due to its zero-current switching capability when operating in discontinuous conduction mode while keeping its input-output ripple current low. However, operating this kind of converter at interleaved operation for all the time gives poor efficiency under light-load condition. In this paper, an automatic operation region selector switch based on detection of the continuous or discontinuous current mode is proposed. With this switch, during the light-load condition, only one converter is activated, while during full-load condition both converters will be activated. The simulation results using LTspice software show that the proposed boost converter has a better efficiency compared to the conventional boost converter with efficiency range of 84.6 % to 95.32 % under various load conditions.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130889645","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}
Today, insulation system involves more than one insulation material, whether serial, parallel, or both. In insulation system, the simplest form is composite which consists of two layers of the same material. The benefit of using layered insulation is that the dielectric strength is higher than one layer with the same thickness. There are two methods to test bamboo isolator, i.e., tests of the breakdown voltage of bamboo wall and bamboo fiber. The tests used two methods of composite layers (acrylic and resin) and without composite layer (normal condition). The tests were intended to observe the breakdown voltage on each incidence in each test sample and discover the dielectric strength. The test result showed that the breakdown voltage of bamboo wall was bigger than bamboo fiber with or without composite layer. The highest dielectric strengths of acrylic layer in petung bamboo wall were 2.07 kV/mm and acrylic layer in bamboo fiber was 0.95 kV/mm in layered method. Meanwhile, in ampel bamboo wall it was 1.35 kV/mm and petung bamboo fiber it was 0.43 kV/mm in normal condition (without layer).
{"title":"The Breakdown Voltage Characteristics of Bamboo Insulation under Composite Layer Based","authors":"L. Hakim, T. Haryono, S. Suharyanto","doi":"10.22146/IJITEE.36648","DOIUrl":"https://doi.org/10.22146/IJITEE.36648","url":null,"abstract":"Today, insulation system involves more than one insulation material, whether serial, parallel, or both. In insulation system, the simplest form is composite which consists of two layers of the same material. The benefit of using layered insulation is that the dielectric strength is higher than one layer with the same thickness. There are two methods to test bamboo isolator, i.e., tests of the breakdown voltage of bamboo wall and bamboo fiber. The tests used two methods of composite layers (acrylic and resin) and without composite layer (normal condition). The tests were intended to observe the breakdown voltage on each incidence in each test sample and discover the dielectric strength. The test result showed that the breakdown voltage of bamboo wall was bigger than bamboo fiber with or without composite layer. The highest dielectric strengths of acrylic layer in petung bamboo wall were 2.07 kV/mm and acrylic layer in bamboo fiber was 0.95 kV/mm in layered method. Meanwhile, in ampel bamboo wall it was 1.35 kV/mm and petung bamboo fiber it was 0.43 kV/mm in normal condition (without layer).","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124617673","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}
D. Muriyatmoko, A. Abdullah, Zaenury Adhiim Musyafa
The goal of the Android mobile application for Arabic learning is to help students who are not alumni of Muslim boarding school in everyday interaction using the Arabic language at Universitas Darussalam Gontor and to function as a medium to introduce Arabic of Pondok Modern Darussalam Gontor version to beginners. The contents come from the book of the first edition of chapter 1 to 12 written by KH. Imam Zarkasyi and KH. Imam Syubani. These books are studied by santri of KMI PMDG of class one. This application applies grammar translation and audiolingual method to learn Arabic for beginners and is developed using ADDIE (analysis, design, development, implementation, and evaluation) approach model. The test was done in six ways: its compatibility with any android device at least 4.5 inchi with the type of android mimimum version of 4.0.4 (Ice Cream Sandwich), an expert material of Arabic language learning give rate score 95%, an expert of media learning give rate score 88.9%, 7 of 11 lecturers of Arabic language give an average score of 81.4%, 6 of 30 students from non-alumni muslim boarding school at this campus give 91.6% and 373 general users via playstore from February 1, 2017, to November 18, 2017, rated 4.9. The result of the overall evaluation of all the tests on this apps shows that this application is relevant to beginner users because of its ease. However, it cannot replace the role of teacher, class, and situation of real Muslim boarding school.
{"title":"Durus Al-Lughah Gontory: Interactive Arabic Mobile Learning for Beginners","authors":"D. Muriyatmoko, A. Abdullah, Zaenury Adhiim Musyafa","doi":"10.22146/IJITEE.36641","DOIUrl":"https://doi.org/10.22146/IJITEE.36641","url":null,"abstract":"The goal of the Android mobile application for Arabic learning is to help students who are not alumni of Muslim boarding school in everyday interaction using the Arabic language at Universitas Darussalam Gontor and to function as a medium to introduce Arabic of Pondok Modern Darussalam Gontor version to beginners. The contents come from the book of the first edition of chapter 1 to 12 written by KH. Imam Zarkasyi and KH. Imam Syubani. These books are studied by santri of KMI PMDG of class one. This application applies grammar translation and audiolingual method to learn Arabic for beginners and is developed using ADDIE (analysis, design, development, implementation, and evaluation) approach model. The test was done in six ways: its compatibility with any android device at least 4.5 inchi with the type of android mimimum version of 4.0.4 (Ice Cream Sandwich), an expert material of Arabic language learning give rate score 95%, an expert of media learning give rate score 88.9%, 7 of 11 lecturers of Arabic language give an average score of 81.4%, 6 of 30 students from non-alumni muslim boarding school at this campus give 91.6% and 373 general users via playstore from February 1, 2017, to November 18, 2017, rated 4.9. The result of the overall evaluation of all the tests on this apps shows that this application is relevant to beginner users because of its ease. However, it cannot replace the role of teacher, class, and situation of real Muslim boarding school.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117164674","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}
Moch. Ari Nasichuddin, T. B. Adji, Widyawan Widyawan
The approach using Deep Learning method provides great results in various field implementations, especially in the field of Sentiment Analysis. One of Deep Learning methods is CNN which has the ability to provide great accuracy in some previous research. However, there are some parts of the training process which can be improved to upgrade the accuracy level and the training time. In this paper, we try to improve the accuracy and processing time of sentiment analysis using CNN model. By tuning the filter size, frameworks, and pre-training, the results show that the use of smaller filter size and pre-training word2vec provide greater accuracy than some previous studies.
{"title":"Performance Improvement Using CNN for Sentiment Analysis","authors":"Moch. Ari Nasichuddin, T. B. Adji, Widyawan Widyawan","doi":"10.22146/IJITEE.36642","DOIUrl":"https://doi.org/10.22146/IJITEE.36642","url":null,"abstract":"The approach using Deep Learning method provides great results in various field implementations, especially in the field of Sentiment Analysis. One of Deep Learning methods is CNN which has the ability to provide great accuracy in some previous research. However, there are some parts of the training process which can be improved to upgrade the accuracy level and the training time. In this paper, we try to improve the accuracy and processing time of sentiment analysis using CNN model. By tuning the filter size, frameworks, and pre-training, the results show that the use of smaller filter size and pre-training word2vec provide greater accuracy than some previous studies.","PeriodicalId":292390,"journal":{"name":"IJITEE (International Journal of Information Technology and Electrical Engineering)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115171403","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：