Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Not a member yet
1071 research outputs found
Sort by
The Image Extraction Using the HSV Method to Determine the Maturity Level of Palm Oil Fruit with the k-nearest Neighbor Algorithm
No full textThe oil palm is one of the monocot oil-producing plants in Indonesia. Sorting errors in oil palm fruit are caused by a sorter error when distinguishing the color of ripe and immature oil palm fruit. In addition to inefficient time, the area of oil palm plantations is also a factor that causes the sorter to make mistakes in sorting. This study aims to produce a system that can classify the maturity of oil palms based on feature extraction of characteristics of the hue, saturation and value (HSV) color features. The HSV method is used to produce color characteristics from the image of the oil palm fruit. Classification of oil palm fruit maturity is classified using the K-Nearest Neighbor (KNN) algorithm with a dataset of 400 oil palm fruit image data with a data sharing ratio of 70% training data and 30% test data. 280 image data were used as training data, divided into 140 image data of ripe oil palm fruit, 140 image data of immature oil palm fruit and 120 image data of oil palm used as test data which is divided into 60 image data of ripe oil palm and 45 unripe palm oil. Based on the result of tests that have been carried out using a confusion matrix with varied k values, namely, 5 and 7, the average precision is 94.16%.
 
Implementation of Naïve bayes Method for Predictor Prevalence Level for Malnutrition Toddlers in Magelang City
Get PDFNutritional status is an important factor in assessing the growth and development rate of babies and toddlers. Cases of malnutrition are increasing, especially in magelang city. Because nutritional problems (Malnutrition) can affect the health of toddlers. Therefore, this study aims to predict the level of prevalence of malnutrition with the Naïve Bayes method. This research uses an observational design, a single center study at the Magelang City Office, using the Naïve bayes method which is used as an application of time series data, and is most widely used for prediction, especially in data sets that have many categorical or nominal type attributes. The Naïve bayes method is used to predict such cases of malnutrition. The results of this study show that the Naïve Bayes method succeeded in predicting the magnitude of cases of malnourished toddlers in Magelang City with an accuracy percentage of 75% due to the very minimal amount of training data, and the areas that have the most malnutrition are in three areas, namely Magersari, North Tidar and Panjang.
Nutritional status is an important factor in assessing the growth and development rate of babies and toddlers. Cases of malnutrition are increasing, especially in magelang city. Because nutritional problems (Malnutrition) can affect the health of toddlers. Therefore, this study aims to predict the level of prevalence of malnutrition with the Naïve Bayes method. This research uses an observational design, a single center study at the Magelang City Office, using the Naïve bayes method which is used as an application of time series data, and is most widely used for prediction, especially in data sets that have many categorical or nominal type attributes. The Naïve bayes method is used to predict such cases of malnutrition. The results of this study show that the Naïve Bayes method succeeded in predicting the magnitude of cases of malnourished toddlers in Magelang City with an accuracy percentage of 75% due to the very minimal amount of training data, and the areas that have the most malnutrition are in three areas, namely Magersari, North Tidar and Panjang
TAARA Method for Processing on the Network Forensics in the Event of an ARP Spoofing Attack
No full textAccording to reports in 2021 by Kaspersky, requests for investigations into suspicious network activity, such as ARP Spoofing, which can result in sophisticated attacks, reached up to 22%. Several difficulties with examining network systems have been overcome thanks to network forensic investigations. This study aims to perform a network forensic analysis of ARP spoofing attacks using Wireshark forensic tools and Network Miner with a sniffer design process to capture traffic on the router side. In order to gather reliable evidence, this study employs the TAARA method as a network forensic investigation process. Based on the research conducted, it can be demonstrated that an attack took place from eight PCAP files. The information that was gathered, such as the IP address and MAC address of the attacker, the IP address and MAC address of the target, and the date and time of the attack are examples of evidence information that was gathered. This study also shows that network forensic operations can use the Wireshark forensic tool to obtain more detailed data.
According to reports in 2021 by Kaspersky, requests for investigations into suspicious network activity, such as ARP Spoofing, which can result in sophisticated attacks, reached up to 22%. Several difficulties with examining network systems have been overcome thanks to network forensic investigations. This study aims to perform a network forensic analysis of ARP spoofing attacks using Wireshark forensic tools and Network Miner with a sniffer design process to capture traffic on the router side. In order to gather reliable evidence, this study employs the TAARA method as a network forensic investigation process. Based on the research conducted, it can be demonstrated that an attack took place from eight PCAP files. The information that was gathered, such as the IP address and MAC address of the attacker, the IP address and MAC address of the target, and the date and time of the attack are examples of evidence information that was gathered. This study also shows that network forensic operations can use the Wireshark forensic tool to obtain more detailed data
Face Recognition of Indonesia’s Top Government Officials Using Deep Convolutional Neural Network
Get PDFFacial recognition is a part of Computer Vision that is used to get facial coordinates from an image. Many algorithms have been developed to support Facial Detection such as Cascade Face Detection using Haar-Like features and AdaBoost to classify its Cascade and Convolutional Neural Network (CNN). Face recognition in this study uses the Deep Convolutional Neural Network (DCNN) method, and the output of this method is the measurement value of the face. In the model training process, Triplet Loss from Triplet Network Deep Metric Learning is used to get good face grouping results. The value of this face measurement will then be measured using the Euclidean distance calculation to determine the similarity of the input face from the dataset. This Research is using 6 images of Government officers in Indonesia to determine the accuracy of the model when there is a new picture of these officers inputted into the training machine. The result provides a 94% accuracy level with a variety of face positions and levels of brightness.Facial recognition is a part of Computer Vision that is used to get facial coordinates from an image. Many algorithms have been developed to support Facial Detection such as Cascade Face Detection using Haar-Like features and AdaBoost to classify its Cascade and Convolutional Neural Network (CNN). Face recognition in this study uses the Deep Convolutional Neural Network (DCNN) method, and the output of this method is the measurement value of the face. In the model training process, Triplet Loss from Triplet Network Deep Metric Learning is used to get good face grouping results. The value of this face measurement will then be measured using the Euclidean distance calculation to determine the similarity of the input face from the dataset. This Research is using 6 images of Government officers in Indonesia to determine the accuracy of the model when there is a new picture of these officers inputted into the training machine. The result provides a 94% accuracy level with a variety of face positions and levels of brightness
Classification of Fruits Based on Shape and Color using Combined Nearest Mean Classifiers
Get PDFFruit classification is an important task in many agriculture industry. The fruit classification system can be used to identify the types and prices of fruit. Manual classification of fruit is not efficient for large amount of fruits. The advancement of information technology has made possible fruit classification be done by a machine. This research aims to propose a fruit classification methodology based on shape and color. To reduce the effect of lighting variability a color normalization is carried out prior to feature extraction. The color features used in this research are mean and standard deviation. The shape features are area, perimeter, and compactness. The classification of an unknown fruit is carried out using the nearest mean classifier. The method developed in this research is tested using 12 classes of fruits where each class is represented by a number of samples. The experimental results show that the method proposed in this research provides an accuracy of 95.83% for two samples per class and 100% for three samples per class. Experiment on small training samples has been conducted to evaluate the performance of the proposed combined nearest mean classifiers and results obtained showed that the technique was able to provide good accuracy.Fruit classification is an important task in many agriculture industry. The fruit classification system can be used to identify the types and prices of fruit. Manual classification of fruit is not efficient for large amount of fruits. The advancement of information technology has made possible fruit classification be done by a machine. This research aims to propose a fruit classification methodology based on shape and color. To reduce the effect of lighting variability a color normalization is carried out prior to feature extraction. The color features used in this research are mean and standard deviation. The shape features are area, perimeter, and compactness. The classification of an unknown fruit is carried out using the nearest mean classifier. The method developed in this research is tested using 12 classes of fruits where each class is represented by a number of samples. The experimental results show that the method proposed in this research provides an accuracy of 95.83% for two samples per class and 100% for three samples per class. Experiment on small training samples has been conducted to evaluate the performance of the proposed combined nearest mean classifiers and results obtained showed that the technique was able to provide good accuracy
Covid-19 Fake News Detection on Twitter Based on Author Credibility Using Information Gain and KNN Methods
Get PDFTwitter is one of the social media that is used as a tool to share various kinds of information about various kinds of things that are of concern to social media users. One of the information shared is information about COVID-19, which is known that the COVID-19 pandemic is currently spreading throughout the world at a very alarming rate. COVID-19 is an infectious disease caused by SARS-COV-2. The World Health Organization (WHO) claims that the spread of COVID-19 is supported by the spread of false/fake news. So to find out the truth of the news, a COVID-19 fake news detector is needed so that users don't fall for the hoaxes circulating. This study aims to classify COVID-19 news on Twitter based on author credibility. Credibility in question is a person's perception of the validity of information and is a multidimensional concept that is used as a means of receiving information to assess the source of communication. The method used in this research is Information Gain and KNN. KNN (K-Nearest Neighbor) is a supervised learning algorithm that works by classifying a set of data based on classified training data. Information Gain is used to ranking the most influential attributes, and KNN is used to classify data based on learning data taken from the nearest neighbors. The research consists of 6 main stages, namely data collection (crawling data), data preprocessing, feature extraction, feature selection, data split into training data and testing data, KNN stage, and data evaluation stage. The research carried out succeeded in obtaining an accuracy value of 91%, a correlation value between credibility and hoax of 0.115, and a p-value <0.005.
Twitter is one of the social media that is used as a tool to share various kinds of information about various kinds of things that are of concern to social media users. One of the information shared is information about COVID-19, which is known that the COVID-19 pandemic is currently spreading throughout the world at a very alarming rate. COVID-19 is an infectious disease caused by SARS-COV-2. The World Health Organization (WHO) claims that the spread of COVID-19 is supported by the spread of false/fake news. So to find out the truth of the news, a COVID-19 fake news detector is needed so that users don't fall for the hoaxes circulating. This study aims to classify COVID-19 news on Twitter based on author credibility. Credibility in question is a person's perception of the validity of information and is a multidimensional concept that is used as a means of receiving information to assess the source of communication. The method used in this research is Information Gain and KNN. KNN (K-Nearest Neighbor) is a supervised learning algorithm that works by classifying a set of data based on classified training data. Information Gain is used to ranking the most influential attributes, and KNN is used to classify data based on learning data taken from the nearest neighbors. The research consists of 6 main stages, namely data collection (crawling data), data preprocessing, feature extraction, feature selection, data split into training data and testing data, KNN stage, and data evaluation stage. The research carried out succeeded in obtaining an accuracy value of 91%, a correlation value between credibility and hoax of 0.115, and a p-value <0.005
A Security Architecture for Mobile Computing-Based IoT
No full textThe Internet of Things (IoT) is a complex technology with various applications that have become a vital part of our daily life. The number of connected devices and the Internet of Things continues to increase. In terms of bandwidth, service availability, security controls, cyberattacks, and privacy problems, transporting the huge data created by these IoT devices to the cloud offers concern and challenges comprising intermittent connection, service unavailability, data loss at rest, in use, and in motion, unhardened device and server, unpatched device and server, and exploitation vulnerabilities. Mobile computing (MC) is a strategic solution to tackle these difficulties by offering flexible data processing and storage to end users, increase security controls and with customized IoT devices for varied geographic locations. This paper provides a complete description of the components of the IoT architecture. After that, it elaborates on similar security threats and potential cyberattacks in the context of mobile computing-based IoT and suggests solutions. In conclusion, we provide a secure mobile computing architecture design for IoT applications.
The Internet of Things (IoT) is a complex technology with various applications that have become a vital part of our everyday life. The amount of Internet of Things and linked devices continues to rise. In terms of bandwidth, service availability, security controls, cyber-attacks and privacy problems, transporting the huge data created by these IoT devices to the cloud offers concern and challenges comprising of intermittent connection, service unavailability, data loss at rest, in use and in motion, unhardened device and server, unpatched device and server and exploitation vulnerabilities. Mobile computing (MC) is a strategic solution to tackle these difficulties by offering flexible data processing and storage to end users, increase security controls and with customized IoT devices for varied geographic locations. This paper is providing a complete description of the IoT architecture's components. After that, it elaborates on similar security threats and potential cyber-attacks in the context of mobile computing-based IoT and suggested solutions. In conclusion, we provide a secure mobile-computing-based architecture design for IoT applications
Implementation of n-gram Methodology to Analyze Sentiment Reviews for Indonesian Chips Purchases in Shopee E-Marketplace
Get PDFChips are a well-known product among Small and Medium Enterprises (SMEs). In order to enhance the quality of chips as an SME product, sentiment analysis is a crucial step. In this research, sentiment analysis of chip purchases on the Shopee E-marketplace was conducted using the Natural Language Processing (NLP) method, utilizing the N-Gram Model and Term Frequent-Inverse Document Frequency (TF-IDF) as feature extraction techniques, and the Support Vector Machine (SVM) algorithm for sentiment classification. The objective of this research is to identify the most suitable feature extraction model and optimal SVM kernel type from the options of Linear, Polynomial degree, Gaussian RBF, and Sigmoid kernels. Results from the experiments indicate that the TF-IDF and unigram feature extraction techniques offer the best performance for SVM classification when utilizing the Linear kernel. By labeling the dataset, it was observed that using a lexicon-based approach for sentiment classification resulted in 84.31% of the total reviews being positive. The words "price", "cheap" and "quality" in unigram have the highest weights above 0.040. In the unigram model, linear kernel accuracy and precision performance values are 88.4% and 87.3%. At the same time, the recall performance values is 88.4%. The results of the F1-Score assessment matrix from Unigram were 86.9%, Bigram was 78.5% and Trigram was 77.4%. Ultimately, the unigram model combined with a linear kernel in the SVM algorithm demonstrates strong potential for application in the development of various systems focused on detecting user reviews in the Indonesian language on the Shopee E-Marketplace. Chips are a well-known product among Small and Medium Enterprises (SMEs). In order to enhance the quality of chips as an SME product, sentiment analysis is a crucial step. In this research, sentiment analysis of chip purchases on the Shopee E-marketplace was conducted using the Natural Language Processing (NLP) method, utilizing the N-Gram Model and Term Frequent-Inverse Document Frequency (TF-IDF) as feature extraction techniques, and the Support Vector Machine (SVM) algorithm for sentiment classification. The objective of this research is to identify the most suitable feature extraction model and optimal SVM kernel type from the options of Linear, Polynomial degree, Gaussian RBF, and Sigmoid kernels. Results from the experiments indicate that the TF-IDF and unigram feature extraction techniques offer the best performance for SVM classification when utilizing the Linear kernel. By labeling the dataset, it was observed that using a lexicon-based approach for sentiment classification resulted in 84.31% of the total reviews being positive. The words "price", "cheap" and "quality" in unigram have the highest weights above 0.040. In the unigram model, linear kernel accuracy and precision performance values are 88.4% and 87.3%. At the same time, the recall performance values is 88.4%. The results of the F1-Score assessment matrix from Unigram were 86.9%, Bigram was 78.5% and Trigram was 77.4%. Ultimately, the unigram model combined with a linear kernel in the SVM algorithm demonstrates strong potential for application in the development of various systems focused on detecting user reviews in the Indonesian language on the Shopee E-Marketplace. 
Analysis of User Readiness Using the TRI Model for Smart School Implementation in the City of Pekanbaru
Get PDFCurrently, Smart Schools have been widely applied in several schools, within the scope of education and services as they are being encouraged to support Smart City. Smart Schools is a school concept utilizing information technology used in the teaching and learning process in the class and school administration. One of the schools in Pekanbaru City that will implement smart schools is Junior High School 17 Pekanbaru. The aspect of building smart schools themselves is not only adequate infrastructure such as servers, labor, and integrated systems but also the readiness on the part of schools and students in implementing Smart Schools in the future. Therefore, to find out the level of readiness of prospective users of the Smart Schools concept, the technology readiness index (TRI) method with four personality variables; optimism, innovativeness, discomfort, and insecurity was used. The purpose of this research was to find out the readiness index of prospective users in the implementation of Smart Schools and see what factors need to be improved from the readiness of prospective users. The results show that teachers and students are ready to apply new technologies in an effort to implement smart schools at Junior High School 17 Pekanbaru. This can be seen from the results obtained, namely the optimism and innovation variables received medium to high ratings. for the discomfort and insecurity to be completely low. However, the student guardians are still unsure because all variables get medium scores. From these results it was stated that Junior High School 17 Pekanbaru was ready to apply new technology for implementing smart schools. In addition, this research can also serve as a guideline for other junior high schools in analyzing new technology users, so that the applied technology can run well.
Currently, Smart Schools have been widely applied in several schools, within the scope of education and services as they are being encouraged to support Smart City. Smart Schools is a school concept utilizing information technology used in the teaching and learning process in the class and school administration. One of the schools in Pekanbaru City that will implement smart schools is Junior High School 17 Pekanbaru. The aspect of building smart schools themselves is not only adequate infrastructure such as servers, labor, and integrated systems but also the readiness on the part of schools and students in implementing Smart Schools in the future. Therefore, to find out the level of readiness of prospective users of the Smart Schools concept, the technology readiness index (TRI) method with four personality variables; optimism, innovativeness, discomfort, and insecurity was used. The purpose of this research was to find out the readiness index of prospective users in the implementation of Smart Schools and see what factors need to be improved from the readiness of prospective users. The results show that teachers and students are ready to apply new technologies in an effort to implement smart schools at Junior High School 17 Pekanbaru. This can be seen from the results obtained, namely the optimism and innovation variables received medium to high ratings. for the discomfort and insecurity to be completely low. However, the student guardians are still unsure because all variables get medium scores. From these results it was stated that Junior High School 17 Pekanbaru was ready to apply new technology for implementing smart schools. In addition, this research can also serve as a guideline for other junior high schools in analyzing new technology users, so that the applied technology can run well
Modified Q-Learning Algorithm for Mobile Robot Real-Time Path Planning using Reduced States
No full textPath planning is an essential algorithm in any autonomous mobile robot, including agricultural robots. One of the reinforcement learning methods that can be used for mobile robot path planning is the Q-Learning algorithm. However, the conventional Q-learning method explores all possible robot states in order to find the most optimum path. Thus, this method requires extensive computational cost especially when there are considerable grids to be computed. This study modified the original Q-Learning algorithm by removing the impassable area, so that these areas are not considered as grids to be computed. This modified Q-Learning method was simulated as path finding algorithm for autonomous mobile robot operated at the Agribusiness and Technology Park (ATP), IPB University. Two simulations were conducted to compare the original Q-Learning method and the modified Q-Learning method. The simulation results showed that the state reductions in the modified Q-Learning method can lower the computation cost to 50.71% from the computation cost of the original Q-Learning method, that is, an average computation time of 25.74s as compared to 50.75s, respectively. Both methods produce similar number of states as the robot’s optimal path, i.e. 56 states, based on the reward obtained by the robot while selecting the path. However, the modified Q-Learning algorithm is capable of finding the path to the destination point with a minimum learning rate parameter value of 0.2 when the discount factor value is 0.9.Path planning is an essential algorithm in any autonomous mobile robot, including agricultural robots. One of the reinforcement learning methods that can be used for mobile robot path planning is the Q-Learning algorithm. However, the conventional Q-learning method explores all possible robot states in order to find the most optimum path. Thus, this method requires extensive computational cost especially when there are considerable grids to be computed. This study modified the original Q-Learning algorithm by removing the impassable area, so that these areas are not considered as grids to be computed. This modified Q-Learning method was simulated as path finding algorithm for autonomous mobile robot operated at the Agribusiness and Technology Park (ATP), IPB University. Two simulations were conducted to compare the original Q-Learning method and the modified Q-Learning method. The simulation results showed that the state reductions in the modified Q-Learning method can lower the computation cost to 50.71% from the computation cost of the original Q-Learning method, that is, an average computation time of 25.74s as compared to 50.75s, respectively. Both methods produce similar number of states as the robot’s optimal path, i.e. 56 states, based on the reward obtained by the robot while selecting the path. However, the modified Q-Learning algorithm is capable of finding the path to the destination point with a minimum learning rate parameter value of 0.2 when the discount factor value is 0.9