Pub Date : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285509
Luthfia Rahman, N. A. Setiawan, A. E. Permanasari
Data mining began to be applied in various fields, one of them on educational data. By exploring information or knowledge in a data allows an institution to improve the learning process and the quality of the institution. This research proposes feature selection techniques in improving Student's Academic Performance classification accuracy. The algorithm used is Naive Bayes, Decision Tree, and Artificial Neural Network, which will be applied to the features selection; wrapper and information gain. The application of feature selection is intended to obtain a higher accuracy value. When compared to the embedded method in previous studies, the feature selection on this experiment has a lower accuracy rate.
{"title":"Feature selection methods in improving accuracy of classifying students' academic performance","authors":"Luthfia Rahman, N. A. Setiawan, A. E. Permanasari","doi":"10.1109/ICITISEE.2017.8285509","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285509","url":null,"abstract":"Data mining began to be applied in various fields, one of them on educational data. By exploring information or knowledge in a data allows an institution to improve the learning process and the quality of the institution. This research proposes feature selection techniques in improving Student's Academic Performance classification accuracy. The algorithm used is Naive Bayes, Decision Tree, and Artificial Neural Network, which will be applied to the features selection; wrapper and information gain. The application of feature selection is intended to obtain a higher accuracy value. When compared to the embedded method in previous studies, the feature selection on this experiment has a lower accuracy rate.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"9 40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129830115","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285511
Tomi Afrizal, C. Surussavadee
This paper evaluates climate simulations at high resolution for Sumatra simulated using the Weather Research and Forecasting (WRF) model driven by climate simulations of the global climate model Community Earth System Model (CESM) version 1. CESM/WRF simulated climatological temperature and precipitation at 10-km resolution covering the 20-y period of 1980–1999 were evaluated using the University of East Anglia Climatic Research Unit, the University of Delaware Air Temperature and Precipitation, and the Global Precipitation Climate Center observation datasets, and the ERA-Interim reanalysis dataset. Results show that CESM/WRF performs much better than CESM does and can simulate local spatial patterns and values of precipitation and temperature for Sumatra well. CESM/WRF simulated climatological temperature and precipitation agree well with observations and reanalysis. CESM/WRF is useful for providing climate simulations at high resolution for Sumatra.
{"title":"Evaluation of CESM/WRF climate simulations at high resolution over sumatra","authors":"Tomi Afrizal, C. Surussavadee","doi":"10.1109/ICITISEE.2017.8285511","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285511","url":null,"abstract":"This paper evaluates climate simulations at high resolution for Sumatra simulated using the Weather Research and Forecasting (WRF) model driven by climate simulations of the global climate model Community Earth System Model (CESM) version 1. CESM/WRF simulated climatological temperature and precipitation at 10-km resolution covering the 20-y period of 1980–1999 were evaluated using the University of East Anglia Climatic Research Unit, the University of Delaware Air Temperature and Precipitation, and the Global Precipitation Climate Center observation datasets, and the ERA-Interim reanalysis dataset. Results show that CESM/WRF performs much better than CESM does and can simulate local spatial patterns and values of precipitation and temperature for Sumatra well. CESM/WRF simulated climatological temperature and precipitation agree well with observations and reanalysis. CESM/WRF is useful for providing climate simulations at high resolution for Sumatra.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125287768","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285496
Iswandi, Albert Kristian Danan Jaya, Eny Sukani Rahayu
This paper proposed and investigated the printed dipole antenna for indoor small cell base stations. The antenna was designed for working at three bands of frequency allocation for Long Term Evolution (LTE) i.e. 800 MHz, 1700 MHz, and 2300 MHz. The antenna was etched on the FR4/Epoxy of printed circuit board and consisted of three dipole elements for each frequency, with the lower frequency radiating element was folded to reduce the antenna size. The probe feeding method was applied and the input impedance was adjusted by the size of microstrip line from antenna to feeder. The designed antenna has been fabricated and a series of measurement was conducted to validate the design. The findings show the measured antenna characteristic is agree to simulated results and satisfy the requirements to be implemented for LTE base station antenna.
{"title":"Design of triple band printed dipole antenna for indoor small cell base station in lte systems","authors":"Iswandi, Albert Kristian Danan Jaya, Eny Sukani Rahayu","doi":"10.1109/ICITISEE.2017.8285496","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285496","url":null,"abstract":"This paper proposed and investigated the printed dipole antenna for indoor small cell base stations. The antenna was designed for working at three bands of frequency allocation for Long Term Evolution (LTE) i.e. 800 MHz, 1700 MHz, and 2300 MHz. The antenna was etched on the FR4/Epoxy of printed circuit board and consisted of three dipole elements for each frequency, with the lower frequency radiating element was folded to reduce the antenna size. The probe feeding method was applied and the input impedance was adjusted by the size of microstrip line from antenna to feeder. The designed antenna has been fabricated and a series of measurement was conducted to validate the design. The findings show the measured antenna characteristic is agree to simulated results and satisfy the requirements to be implemented for LTE base station antenna.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125381931","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285559
A. Emanuel, Andreas Widjaja
Indonesia is a great country with many cultural diversities, including traditional dances. With the increasing trend toward digitalization, the need to preserving these priceless treasures in digital format is unavoidable. The storing of dances in the form of video format seems inadequate since studying, comparing, and analyzing the video is cumbersome task. This paper discuses about the feasibility study of scripting the Indonesian traditional dances in XML format. Some studies related to these attempts are presented such as Indonesian traditional dances and current technological approach, the motion / gesture capture and recognition, and many more. The proposed approach of dance motion scripting and turn it into XML format seem promising.
{"title":"Feasibility study of scripting Indonesian traditional dance motion in XML format","authors":"A. Emanuel, Andreas Widjaja","doi":"10.1109/ICITISEE.2017.8285559","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285559","url":null,"abstract":"Indonesia is a great country with many cultural diversities, including traditional dances. With the increasing trend toward digitalization, the need to preserving these priceless treasures in digital format is unavoidable. The storing of dances in the form of video format seems inadequate since studying, comparing, and analyzing the video is cumbersome task. This paper discuses about the feasibility study of scripting the Indonesian traditional dances in XML format. Some studies related to these attempts are presented such as Indonesian traditional dances and current technological approach, the motion / gesture capture and recognition, and many more. The proposed approach of dance motion scripting and turn it into XML format seem promising.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121161291","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285560
S. Ismail, Babak Basharirad, S. Ismail
The study consists of 5 major sections which involves with the related works, methodology, result and finding, discussion and conclusion. In related works, this paper focus on Myer Briggs Type Indicator (MBTI) personality model, the issues and factors of indecisiveness and undecided students in choosing university program and the challenges of making decision which related to university environment, advising system and agents. The objectives of this study are to examine students' personality and attitudes in making decision of university program, to understand how student makes decision for the program and what the factors of his or her decision making are. The analysis process has performed and the frequencies pattern was produced by calculate the percentage of personality preferences with the factors and demographic correlates which then as a result of this paper findings.
本研究分为5个主要部分，包括相关工作、方法、结果和发现、讨论和结论。在相关工作中，本文重点研究了Myer Briggs Type Indicator (MBTI)人格模型、优柔优断和犹豫不决的学生在大学项目选择中存在的问题和因素以及与大学环境、咨询系统和代理人相关的决策挑战。本研究的目的是考察学生在大学专业决策中的个性和态度，了解学生是如何做出专业决策的，以及影响其决策的因素是什么。分析过程已经完成，频率模式是通过计算人格偏好与因素和人口相关因素的百分比产生的，然后作为本文发现的结果。
{"title":"Significant of MBTI personality model on decision making in university program selection","authors":"S. Ismail, Babak Basharirad, S. Ismail","doi":"10.1109/ICITISEE.2017.8285560","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285560","url":null,"abstract":"The study consists of 5 major sections which involves with the related works, methodology, result and finding, discussion and conclusion. In related works, this paper focus on Myer Briggs Type Indicator (MBTI) personality model, the issues and factors of indecisiveness and undecided students in choosing university program and the challenges of making decision which related to university environment, advising system and agents. The objectives of this study are to examine students' personality and attitudes in making decision of university program, to understand how student makes decision for the program and what the factors of his or her decision making are. The analysis process has performed and the frequencies pattern was produced by calculate the percentage of personality preferences with the factors and demographic correlates which then as a result of this paper findings.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128094636","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285541
A. Setiawan, E. Sediyono
The method provides a much simpler way to determine the regional area than the existing way. Our calculated regional area of Gili Air island, Salatiga city, Central Java province, DIY province and DKI Jakarta province are 179.46107 ha, 57.437 km2, 34667.79 km2, 3219.87 km2 and 670.895 km2, respectively. All of these calculated area are larger than the corresponding reference regional areas, the maximum deviation is less than 7%.
{"title":"A new determination of regional area by utilizing rectangular approach method and google maps","authors":"A. Setiawan, E. Sediyono","doi":"10.1109/ICITISEE.2017.8285541","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285541","url":null,"abstract":"The method provides a much simpler way to determine the regional area than the existing way. Our calculated regional area of Gili Air island, Salatiga city, Central Java province, DIY province and DKI Jakarta province are 179.46107 ha, 57.437 km2, 34667.79 km2, 3219.87 km2 and 670.895 km2, respectively. All of these calculated area are larger than the corresponding reference regional areas, the maximum deviation is less than 7%.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133919948","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285508
Andi Riyanto, Hendra Marcos, Zulia Karini, Kamal Miftahul Amin
Micro Small Medium Enterprises an important part of a country's economy. Various strategies are created through technology one of them is to determine the amount of production and inventory of goods in order to meet market needs. Fuzzy logic is a logic that is easy to understand and very flexible, it means able to adapt to the changes and certainty that accompanied the problem, so it can be implemented into an application that can optimize the inventory. The purpose of this research is to optimize the supply of goods in Pekanita by using fuzzy mamdani method on sales system, so there is no shortage of inventory and no excess goods. Pekanita is a distributor of sanitary napkin which is engaged in Micro Small Medium Enterprises.
{"title":"Fuzzy logic implementation to optimize multiple inventories on micro small medium enterprises using mamdani method (Case Study: Pekanita, Kroya, Cilacap)","authors":"Andi Riyanto, Hendra Marcos, Zulia Karini, Kamal Miftahul Amin","doi":"10.1109/ICITISEE.2017.8285508","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285508","url":null,"abstract":"Micro Small Medium Enterprises an important part of a country's economy. Various strategies are created through technology one of them is to determine the amount of production and inventory of goods in order to meet market needs. Fuzzy logic is a logic that is easy to understand and very flexible, it means able to adapt to the changes and certainty that accompanied the problem, so it can be implemented into an application that can optimize the inventory. The purpose of this research is to optimize the supply of goods in Pekanita by using fuzzy mamdani method on sales system, so there is no shortage of inventory and no excess goods. Pekanita is a distributor of sanitary napkin which is engaged in Micro Small Medium Enterprises.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115342453","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285567
Litasari, Budi Rahadjo
Well-known techniques for securing data are cryptography and steganography. Cryptography is an information security technique that turns information into another random and incomprehensible form, while steganography secures data by hiding information in storage media such as text, images, sound and video. In this research will be merged both security techniques, namely Stegocrypto. This stegocrypto uses ElGamal's elliptic curve for the coding process and utilizes noise as a steganographic medium. The elliptic curve used in this study has 2026 pairs of points. While the input of this system is an ASCII character, so input should be represented first before being processed. The process of inserting cipher with noise is done without using a marker. In this study, the stegocrypto system was successfully implemented. The system successfully encrypts and decrypts the input message. Decryption is done by reading all received stego-noise and will be read as cipher text when it is an elliptic group element. Then from the test is also obtained the influence of message length and k variables to the time required for the process of encryption and decryption.
{"title":"Design and implementation stegocrypto based on elgamal elliptic curve","authors":"Litasari, Budi Rahadjo","doi":"10.1109/ICITISEE.2017.8285567","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285567","url":null,"abstract":"Well-known techniques for securing data are cryptography and steganography. Cryptography is an information security technique that turns information into another random and incomprehensible form, while steganography secures data by hiding information in storage media such as text, images, sound and video. In this research will be merged both security techniques, namely Stegocrypto. This stegocrypto uses ElGamal's elliptic curve for the coding process and utilizes noise as a steganographic medium. The elliptic curve used in this study has 2026 pairs of points. While the input of this system is an ASCII character, so input should be represented first before being processed. The process of inserting cipher with noise is done without using a marker. In this study, the stegocrypto system was successfully implemented. The system successfully encrypts and decrypts the input message. Decryption is done by reading all received stego-noise and will be read as cipher text when it is an elliptic group element. Then from the test is also obtained the influence of message length and k variables to the time required for the process of encryption and decryption.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122837230","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285506
R. M. Awangga, Nuraini Siti Fathonah, Trisna Irmayadi Hasanudin
Stroke is one of the biggest causes of death in the world. Due to lack of supervision and family support to rehabilitate stroke patients, the number of deaths caused by stroke is increasing. Its because stroke is a neurological disorder that takes a long rehabilitation to get motor skill acquisition for each patient. For those whose goals are to walk again, having physiotherapists help is the way to do so. Unfortunately, there has not been a tool in physiotherapy to conduct medical records measuring travel and mileage of the patient walking outdoor. It is needed for further physiotherapists analysis and recommendation. In this study, we proposed a model called Colenak. It is proposed a Global Positioning System(GPS) tracking model to record the walking (physio)therapy of the stroke patients. This model combines the use of Advance Encryption Stanford (AES) Cipher Block Chaining (CBC) mode and Quick Response (QR) Code. Advance Encryption Standard (AES) CBC algorithm is applied to encrypt the URL. While Uniform Resource Locator (URL)s are known only to patients given QR Code, the QR code will be created based on patient identification and physiotherapy system encrypted along with the URL. By using GPS, the stroke patients who are practicing to walk can be tracked and the tracking result of each patient is expected to support the advance physiotherapy method within this model of physiotherapy.
{"title":"Colenak: GPS tracking model for post-stroke rehabilitation program using AES-CBC URL encryption and QR-Code","authors":"R. M. Awangga, Nuraini Siti Fathonah, Trisna Irmayadi Hasanudin","doi":"10.1109/ICITISEE.2017.8285506","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285506","url":null,"abstract":"Stroke is one of the biggest causes of death in the world. Due to lack of supervision and family support to rehabilitate stroke patients, the number of deaths caused by stroke is increasing. Its because stroke is a neurological disorder that takes a long rehabilitation to get motor skill acquisition for each patient. For those whose goals are to walk again, having physiotherapists help is the way to do so. Unfortunately, there has not been a tool in physiotherapy to conduct medical records measuring travel and mileage of the patient walking outdoor. It is needed for further physiotherapists analysis and recommendation. In this study, we proposed a model called Colenak. It is proposed a Global Positioning System(GPS) tracking model to record the walking (physio)therapy of the stroke patients. This model combines the use of Advance Encryption Stanford (AES) Cipher Block Chaining (CBC) mode and Quick Response (QR) Code. Advance Encryption Standard (AES) CBC algorithm is applied to encrypt the URL. While Uniform Resource Locator (URL)s are known only to patients given QR Code, the QR code will be created based on patient identification and physiotherapy system encrypted along with the URL. By using GPS, the stroke patients who are practicing to walk can be tracked and the tracking result of each patient is expected to support the advance physiotherapy method within this model of physiotherapy.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127025519","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 : 2017-11-01DOI: 10.1109/ICITISEE.2017.8285516
A. Prahara, Dewi Pramudi Ismi, A. I. Kistijantoro, M. L. Khodra
Clustering is a technique to cluster data into defined number of cluster. K-means clustering is the most well-known and widely used clustering algorithm. While data become large in terms of volume, the needs of high performance computing (HPC) to perform data clustering is raising. One of the solutions with compromised budget but high efficiency is to utilize highly parallel architecture of Graphics Processing Unit (GPU). In this research, k-means clustering algorithm is implemented on GPU and optimized by exploiting instruction level parallelism (ILP) at low occupancy. ILP on k-means clustering algorithm is achieved by running a number of independent instruction per thread i.e. when calculating distance or sum of data in each cluster. By loading more works into thread at lower occupancy, the higher utilization can be achieved. Experiment on clustering several data shows that the proposed method can speed up k-means clustering several times faster than other parallelized k-means clustering and k-means implementation on CPU.
{"title":"Parallelized k-means clustering by exploiting instruction level parallelism at low occupancy","authors":"A. Prahara, Dewi Pramudi Ismi, A. I. Kistijantoro, M. L. Khodra","doi":"10.1109/ICITISEE.2017.8285516","DOIUrl":"https://doi.org/10.1109/ICITISEE.2017.8285516","url":null,"abstract":"Clustering is a technique to cluster data into defined number of cluster. K-means clustering is the most well-known and widely used clustering algorithm. While data become large in terms of volume, the needs of high performance computing (HPC) to perform data clustering is raising. One of the solutions with compromised budget but high efficiency is to utilize highly parallel architecture of Graphics Processing Unit (GPU). In this research, k-means clustering algorithm is implemented on GPU and optimized by exploiting instruction level parallelism (ILP) at low occupancy. ILP on k-means clustering algorithm is achieved by running a number of independent instruction per thread i.e. when calculating distance or sum of data in each cluster. By loading more works into thread at lower occupancy, the higher utilization can be achieved. Experiment on clustering several data shows that the proposed method can speed up k-means clustering several times faster than other parallelized k-means clustering and k-means implementation on CPU.","PeriodicalId":130873,"journal":{"name":"2017 2nd International conferences on Information Technology, Information Systems and Electrical Engineering (ICITISEE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121794321","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：