Kusprasapta Mutijarsa
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Kusprasapta Mutijarsa has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Abdul Karim
Brawijaya University,Faculty of Economics and Business,Department of Management,Malang,INDONESIA
3 shared publications - Agustan Agustan
Research Center for Geospatial (PRG), National Research and Innovation Agency (BRIN),Jakarta,INDONESIA
3 shared publications Elisabeth Gunawan
Information System and Information Technology Bandung Institute of Technology (ITB),Bandung,INDONESIA
3 shared publicationsNadira Fawziyya Masnur
Bandung Institute of Technology (ITB),Information System and Information Technology,Bandung,INDONESIA
3 shared publicationsNurul Izza Afkharinah
Bandung Institute of Technology (ITB),Information System and Information Technology,Bandung,INDONESIA
3 shared publicationsSwasetyo Yulianto
Information System and Information Technology Bandung Institute of Technology (ITB),Bandung,INDONESIA
3 shared publications
Latest Publications
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Estimating the Distribution of Paddy Growth Stage in Karawang, West Java Based on Gradient Boosting Algorithm
Paddy is an important and strategic commodity to supply food needs in Indonesia. Karawang Regency in West Java Province is well known as rice producer due to its natural resources and farming system. This study aims to estimate the distribution of paddy growth stage in Karawang Regency based on the classification results using the Gradient Boosting algorithm. The classification was carried out to conduct training on monthly data so as to produce predictions of the distribution of paddy growth stage area. There are 8 classifications defined, including water bodies, early vegetative, late vegetative, generative, harvesting & land preparation, not paddy fields, others, also settlements and roads. As validation material, Area Frame Sampling (Kerangka Sampling Area or KSA) data of Karawang Regency 2020 was used and comparisons were made using the Rainy Season Paddy Cropping Calendar for the October 2020-March 2021 period of Karawang Regency as an evaluation of the results. The results of the accuracy test for estimation of the distribution of paddy growth stage using the Gradient Boosting algorithm show that this algorithm has a good level of suitability and confidence level between rows and columns in the process of estimating the paddy growth stage in Karawang Regency as evidenced by the overall accuracy dominant value is >75% and the kappa statistics dominant value is > 0.7.
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
Estimating the Distribution of Paddy Growth Stage in Karawang, West Java Based on Gradient Boosting Algorithm
Paddy is an important and strategic commodity to supply food needs in Indonesia. Karawang Regency in West Java Province is well known as rice producer due to its natural resources and farming system. This study aims to estimate the distribution of paddy growth stage in Karawang Regency based on the classification results using the Gradient Boosting algorithm. The classification was carried out to conduct training on monthly data so as to produce predictions of the distribution of paddy growth stage area. There are 8 classifications defined, including water bodies, early vegetative, late vegetative, generative, harvesting & land preparation, not paddy fields, others, also settlements and roads. As validation material, Area Frame Sampling (Kerangka Sampling Area or KSA) data of Karawang Regency 2020 was used and comparisons were made using the Rainy Season Paddy Cropping Calendar for the October 2020-March 2021 period of Karawang Regency as an evaluation of the results. The results of the accuracy test for estimation of the distribution of paddy growth stage using the Gradient Boosting algorithm show that this algorithm has a good level of suitability and confidence level between rows and columns in the process of estimating the paddy growth stage in Karawang Regency as evidenced by the overall accuracy dominant value is >75% and the kappa statistics dominant value is > 0.7.
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Comparison of CART Algorithm and Cropping Calendar in Estimating Paddy Growth Stage in Karawang Regency, West Java
Classification And Regression Trees (CART) is one of the classic and simple algorithm in predictive modeling machine learning. This study aims to compare the result of paddy growth stage estimates based on CART model of Sentinel-1A Synthetic Aperture Radar (SAR) data and Cropping Calendar (KATAM). The construction of the CART model utilises real data field from Area Frame Sampling (Kerangka Sampling Area or KSA) in Karawang Regency observed on 2020. The CART algorithm makes predictions using a tree structure or hierarchical structure. The CART algorithm focuses on finding a decision tree model that has a Gini impurities value = 0. The rules for classifying class based on the physical polarization spectrum which is represented by pixel digital number from Vertical-Vertical (VV), Vertical-Horizontal (VH), and VV/VH of SAR image properties. This study found that the initial planting time is different. The CART model estimates the initial planting time is on September, while the KATAM estimates on November-December.
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Comparison of CART Algorithm and Cropping Calendar in Estimating Paddy Growth Stage in Karawang Regency, West Java
Classification And Regression Trees (CART) is one of the classic and simple algorithm in predictive modeling machine learning. This study aims to compare the result of paddy growth stage estimates based on CART model of Sentinel-1A Synthetic Aperture Radar (SAR) data and Cropping Calendar (KATAM). The construction of the CART model utilises real data field from Area Frame Sampling (Kerangka Sampling Area or KSA) in Karawang Regency observed on 2020. The CART algorithm makes predictions using a tree structure or hierarchical structure. The CART algorithm focuses on finding a decision tree model that has a Gini impurities value = 0. The rules for classifying class based on the physical polarization spectrum which is represented by pixel digital number from Vertical-Vertical (VV), Vertical-Horizontal (VH), and VV/VH of SAR image properties. This study found that the initial planting time is different. The CART model estimates the initial planting time is on September, while the KATAM estimates on November-December.
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
The Assessment of Random Forest Algorithm in Identifying Paddy Growth Stage in Karawang, West Java
This study tried to estimate distribution and area of land cover by focusing on the area of the paddy growing phase using the Random Forest model on Sentinel 1A data and Area Frame Sample (Kerangka Sampling Area or KSA) observation data as reference data. Sentinel-1 data from Karawang Regency, West Java was taken monthly for a one-year period starting from January 2020 to December 2020. It was found that the results of extrapolation with the Random Forest algorithm have similar trends and patterns to the results of the KSA. Still, compared to the Cropping Calendar (KATAM) model of the Agricultural Research and Development Agency, the initial planting period of paddy in Karawang Regency, the results of the identification of the Random Forest algorithm appeared two months earlier.
- 24 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-7
The Assessment of Random Forest Algorithm in Identifying Paddy Growth Stage in Karawang, West Java
This study tried to estimate distribution and area of land cover by focusing on the area of the paddy growing phase using the Random Forest model on Sentinel 1A data and Area Frame Sample (Kerangka Sampling Area or KSA) observation data as reference data. Sentinel-1 data from Karawang Regency, West Java was taken monthly for a one-year period starting from January 2020 to December 2020. It was found that the results of extrapolation with the Random Forest algorithm have similar trends and patterns to the results of the KSA. Still, compared to the Cropping Calendar (KATAM) model of the Agricultural Research and Development Agency, the initial planting period of paddy in Karawang Regency, the results of the identification of the Random Forest algorithm appeared two months earlier.
- 12 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
An Engine Abstraction for Smart Engineering Platform
This paper describes an engine abstraction as the core of our engineering platform for multi domain systems. Engineering is called to answer the challenges of living in the 21st century, where traditional engineering is not well equipped for multi domain cases. In our Smart Engineering Platform (SEP), we have expanded engine dimensions, resulting in a design and implementation platform of hard-ware, soft-ware, liquid-ware, vapor-ware, and spatial-ware as core engines. While each dimension has unique peculiarities, our engine abstraction can be used to analyze existing or proposed engineering constructs, as well as to create or adapt a new engine. The paper discusses how the abstraction can be applied to a wide range of engineering, including signal processing systems, competitive systems, and immersive environment.
- 12 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
An Engine Abstraction for Smart Engineering Platform
This paper describes an engine abstraction as the core of our engineering platform for multi domain systems. Engineering is called to answer the challenges of living in the 21st century, where traditional engineering is not well equipped for multi domain cases. In our Smart Engineering Platform (SEP), we have expanded engine dimensions, resulting in a design and implementation platform of hard-ware, soft-ware, liquid-ware, vapor-ware, and spatial-ware as core engines. While each dimension has unique peculiarities, our engine abstraction can be used to analyze existing or proposed engineering constructs, as well as to create or adapt a new engine. The paper discusses how the abstraction can be applied to a wide range of engineering, including signal processing systems, competitive systems, and immersive environment.
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 566-571
Performance Improvement of Recommender Hybrid Techniques Using GRU for Rating Calculation
The recommender system has gained attention in the past few years, because it can provide benefits for both the owner of the company and the user to get relevant items quickly. Collaborative filtering is the most commonly used recommender system method because of several advantages. However, collaborative filtering still has several issues such as cold start, sparse data, gray sheep and dynamic taste. Many studies have tried to accomplish these issues with various approaches. In general, the approach used is hybrid technique to cover the weaknesses of each other. One study has tried to accomplish these issues using a 7 block hybrid technique approach. However, the study has problem when dealing with reviews with varying length of sentences (from very short to very long) for calculating ratings, hence it can reduce the overall quality of the recommender system. This paper aims to propose another approach in calculating rating using gated recurrent unit (GRU) and it will be combined with other blocks of hybrid technique. In addition, new blocks will be added, namely demographic filtering. Based on the results obtained, the proposed method provides better results compared to method of previous study.
- 1 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 566-571
Performance Improvement of Recommender Hybrid Techniques Using GRU for Rating Calculation
The recommender system has gained attention in the past few years, because it can provide benefits for both the owner of the company and the user to get relevant items quickly. Collaborative filtering is the most commonly used recommender system method because of several advantages. However, collaborative filtering still has several issues such as cold start, sparse data, gray sheep and dynamic taste. Many studies have tried to accomplish these issues with various approaches. In general, the approach used is hybrid technique to cover the weaknesses of each other. One study has tried to accomplish these issues using a 7 block hybrid technique approach. However, the study has problem when dealing with reviews with varying length of sentences (from very short to very long) for calculating ratings, hence it can reduce the overall quality of the recommender system. This paper aims to propose another approach in calculating rating using gated recurrent unit (GRU) and it will be combined with other blocks of hybrid technique. In addition, new blocks will be added, namely demographic filtering. Based on the results obtained, the proposed method provides better results compared to method of previous study.
- 1 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Query Support for Data Processing and Analysis on Ethereum Blockchain
Blockchain technology has gained immense popularity because many researchers believe that it could solve numerous problems and could be applied in various fields of study. Unfortunately, behind its potentials, blockchain also possessed many challenges and limitations. The highlighted problem is the usability aspect of blockchain technology examined from developer and user perspective. This paper tried to address this problem by proposing query functionalities, with the help of query layer system, to facilitate the developer and the user to access blockchain data easily. There are three main query functionalities that will be discussed in this paper: (1) finding blockchain data based on multiple search parameters (retrieval query), (2) providing simple statistical analysis from a collection of blockchain data (aggregate query) and (3) sorting blockchain data according to its blockchain component (ranking query). For the implementation stage, Ethereum is used as platform to provide blockchain network, MongoDB is used as cloud storage service and REST API is used as web services. For the testing stage, throughput and time response are used to evaluate the performance of the developed query functionalities in the query layer system. The results are: (1) the throughput of query layer system is lower than Ethereum service for blockchain data retrieval and (2) the time response of query layer system is affected by the number of thread and the amount of data stored in cloud storage.
- 1 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Query Support for Data Processing and Analysis on Ethereum Blockchain
Blockchain technology has gained immense popularity because many researchers believe that it could solve numerous problems and could be applied in various fields of study. Unfortunately, behind its potentials, blockchain also possessed many challenges and limitations. The highlighted problem is the usability aspect of blockchain technology examined from developer and user perspective. This paper tried to address this problem by proposing query functionalities, with the help of query layer system, to facilitate the developer and the user to access blockchain data easily. There are three main query functionalities that will be discussed in this paper: (1) finding blockchain data based on multiple search parameters (retrieval query), (2) providing simple statistical analysis from a collection of blockchain data (aggregate query) and (3) sorting blockchain data according to its blockchain component (ranking query). For the implementation stage, Ethereum is used as platform to provide blockchain network, MongoDB is used as cloud storage service and REST API is used as web services. For the testing stage, throughput and time response are used to evaluate the performance of the developed query functionalities in the query layer system. The results are: (1) the throughput of query layer system is lower than Ethereum service for blockchain data retrieval and (2) the time response of query layer system is affected by the number of thread and the amount of data stored in cloud storage.
- 9 March 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 157-162
Simulation of marketplace customer satisfaction analysis based on machine learning algorithms
Twitter can be a source of public opinion data and sentiment. Such data can be used efficiently for marketing or social studies. In this research addresses this issue by measuring net sentiment based on customer satisfaction through customer's sentiment analysis from Twitter data. Sample model is built and extracted from more than 3.000 raw Twitter messages data from March to April 2016 of top marketplace in Indonesia. We compared several algorithms, and the classification schemes. The sentiments are classified and compared using five different algorithms classification. There are, K-Nearest Neighbor, Logistic Regression, Naïve Bayes, Random Forest, and Support Vector Machine. The five machine learning can be applied to the Indonesian-language sentiment analysis. Preprocessing on the stages of tokenization, parsing, and stop word deletion of word frequency counting. Frequency of the word of the document used weighted by TF-IDF. The Random Forest, Support Vector Machine, and Logistic Regression generate better accuracy and stable compared with the Naïve Bayes and K-Nearest Neighbor. The results showed Support Vector Machine has accuracy 81.82% with 1000 sampling dataset and 85.4% with 2000 sampling dataset. This shows that the more the number of training data will improve the accuracy of the system. The Net Sentiment score for marketplace in Indonesia is 73%. This results also showed that customer satisfaction has average Net Promoter Score (NPS) 3.3%.
- 9 March 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 157-162
Simulation of marketplace customer satisfaction analysis based on machine learning algorithms
Twitter can be a source of public opinion data and sentiment. Such data can be used efficiently for marketing or social studies. In this research addresses this issue by measuring net sentiment based on customer satisfaction through customer's sentiment analysis from Twitter data. Sample model is built and extracted from more than 3.000 raw Twitter messages data from March to April 2016 of top marketplace in Indonesia. We compared several algorithms, and the classification schemes. The sentiments are classified and compared using five different algorithms classification. There are, K-Nearest Neighbor, Logistic Regression, Naïve Bayes, Random Forest, and Support Vector Machine. The five machine learning can be applied to the Indonesian-language sentiment analysis. Preprocessing on the stages of tokenization, parsing, and stop word deletion of word frequency counting. Frequency of the word of the document used weighted by TF-IDF. The Random Forest, Support Vector Machine, and Logistic Regression generate better accuracy and stable compared with the Naïve Bayes and K-Nearest Neighbor. The results showed Support Vector Machine has accuracy 81.82% with 1000 sampling dataset and 85.4% with 2000 sampling dataset. This shows that the more the number of training data will improve the accuracy of the system. The Net Sentiment score for marketplace in Indonesia is 73%. This results also showed that customer satisfaction has average Net Promoter Score (NPS) 3.3%.
- 1 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 640-645
Development of multimedia system for supporting education in rural areas
In Indonesia, the condition of education still tends to be unbalanced. The quality of education in urban areas is generally far better than the quality of education in rural areas. Education processes in rural areas face some problems like limited teaching resources, slight teaching tools, and monotonous teaching method. Therefore, an idea is emerged, a thinking to create a system in order to improve the quality of education in rural areas. Virtual Classbox 4.1 is a distance learning system developed to improve the quality of education in rural areas with video conferencing technology. Virtual Classbox 4.1 is designed based on Design Science in Information System Research. In order to produce a great system, a scientific guidance must be used. Remarkable system need astounding design, like a good house needs a strong foundation. Because of that, Design Science in Information System Research, which is very well known in information system design research world, is used.
- 1 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 640-645
Development of multimedia system for supporting education in rural areas
In Indonesia, the condition of education still tends to be unbalanced. The quality of education in urban areas is generally far better than the quality of education in rural areas. Education processes in rural areas face some problems like limited teaching resources, slight teaching tools, and monotonous teaching method. Therefore, an idea is emerged, a thinking to create a system in order to improve the quality of education in rural areas. Virtual Classbox 4.1 is a distance learning system developed to improve the quality of education in rural areas with video conferencing technology. Virtual Classbox 4.1 is designed based on Design Science in Information System Research. In order to produce a great system, a scientific guidance must be used. Remarkable system need astounding design, like a good house needs a strong foundation. Because of that, Design Science in Information System Research, which is very well known in information system design research world, is used.
- 1 August 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 650-655
Humanoid robot locomotion system with balancing feedback using leg and arm strategy and stepping strategy
Humanoid robot must has an ability to move with dynamic motion. Robot is required to walk even get any interference of other robots or other disturbance, and keep its balance. In general, robot motion system is built by 2 kinematics systems, inverse kinematics and forward kinematics. Inverse kinematics is more often used for robot motion because it is more dynamic compare to forward kinematics. Robot motion also has to have a balancing system. In this capstone design, feedback system for robot motion balancing is developed using 2 strategy, leg and arm strategy, and stepping strategy. Leg and arm strategy is a technique to obtain balancing by giving direct feedback to actuator based on Robot Center of Mass (CoM). Stepping is a technique to maintain balance when robot is walking. Stepping is done by making robot motion trajectory to make one of robot's legs take a charge as robot support. Robot motion balancing algorithm has been implemented in robot platform Baldhart v1.0. Robot has ability to keep its balance with maximum disturbance 25° in 4 directions. Robot has ability to move in a plane with different height for maximum 0.6 cm.
- 1 August 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 650-655
Humanoid robot locomotion system with balancing feedback using leg and arm strategy and stepping strategy
Humanoid robot must has an ability to move with dynamic motion. Robot is required to walk even get any interference of other robots or other disturbance, and keep its balance. In general, robot motion system is built by 2 kinematics systems, inverse kinematics and forward kinematics. Inverse kinematics is more often used for robot motion because it is more dynamic compare to forward kinematics. Robot motion also has to have a balancing system. In this capstone design, feedback system for robot motion balancing is developed using 2 strategy, leg and arm strategy, and stepping strategy. Leg and arm strategy is a technique to obtain balancing by giving direct feedback to actuator based on Robot Center of Mass (CoM). Stepping is a technique to maintain balance when robot is walking. Stepping is done by making robot motion trajectory to make one of robot's legs take a charge as robot support. Robot motion balancing algorithm has been implemented in robot platform Baldhart v1.0. Robot has ability to keep its balance with maximum disturbance 25° in 4 directions. Robot has ability to move in a plane with different height for maximum 0.6 cm.