Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Not a member yet
1071 research outputs found
Sort by
k-Nearest Neighbor and Feature Extraction on Detection of Pest and Diseases of Cocoa
Get PDFKnowledge and utilization of digital images are growing rapidly not only in the fields of medicine and industry but also in the field of agriculture. This knowledge can apply it to a computer-based program that is used to detect agricultural products more effectively and efficiently. this research aims to build a system to detect the types of pests and diseases of cocoa pods because in general, an inspection of pests and diseases of cocoa pods is still manual based on the visual analysis of the color of the pods visually by the human eye which has limitations, which requires more energy to sort, the level of human consistency. In terms of assessing the symptoms of pests and fruit diseases, it is not guaranteed, because humans can experience fatigue, and humans also assess symptoms of pests and fruit diseases, sometimes it is subjective. This study utilizes digital image processing techniques to extract the color features of digital images of cocoa pods, the method used to extract the color features of Hue, Saturation, Value (HSV), and the classification algorithm used by K-Nearest Neighbor. The data used as many as 150 images divided into 70% training data and 30% testing data. Based on the results of trials using k values of 5,7,11 and 13 in the holdout method, the best accuracy is 84.44% with a value of k = 5. And in the k-5 cross-validation test, the best accuracy is also found at k = 5 with a value accuracy of 99.33%.
Knowledge and utilization of digital images are growing rapidly not only in the fields of medicine and industry but also in the field of agriculture. This knowledge can apply it to a computer-based program that is used to detect agricultural products more effectively and efficiently. this research aims to build a system to detect the types of pests and diseases of cocoa pods because in general, an inspection of pests and diseases of cocoa pods is still manual based on the visual analysis of the color of the pods visually by the human eye which has limitations, which requires more energy to sort, the level of human consistency. In terms of assessing the symptoms of pests and fruit diseases, it is not guaranteed, because humans can experience fatigue, and humans also assess symptoms of pests and fruit diseases, sometimes it is subjective. This study utilizes digital image processing techniques to extract the color features of digital images of cocoa pods, the method used to extract the color features of Hue, Saturation, Value (HSV), and the classification algorithm used by K-Nearest Neighbor. The data used as many as 150 images divided into 70% training data and 30% testing data. Based on the results of trials using k values of 5,7,11 and 13 in the holdout method, the best accuracy is 84.44% with a value of k = 5. And in the k-5 cross-validation test, the best accuracy is also found at k = 5 with a value accuracy of 99.33%
Application of Neural Network Variations for Determining the Best Architecture for Data Prediction
Get PDFThis study focuses on the application and comparison of the epoch, time, performance/MSE training, and performance/MSE testing of variations of the Backpropagation algorithm. The main problem in this study is that the Backpropagation algorithm tends to be slow to reach convergence in obtaining optimum accuracy, requires extensive training data, and the optimization used is less efficient and has performance/MSE which can still be improved to produce better performance/MSE in this research—data prediction process. Determination of the best model for data prediction is seen from the performance/MSE test. This data prediction uses five variations of the Backpropagation algorithm: standard Backpropagation, Resistant Backpropagation, Conjugate Gradient, Fletcher Reeves, and Powell Beale. The research stage begins with processing the avocado production dataset in Indonesia by province from 2016 to 2021. The dataset is first normalized to a value between 0 to 1. The test in this study was carried out using Matlab 2011a. The dataset is divided into two, namely training data and test data. This research's benefit is producing the best model of the Backpropagation algorithm in predicting data with five methods in the Backpropagation algorithm. The test results show that the Resilient Backpropagation method is the best model with a test performance of 0.00543829, training epochs of 1000, training time of 12 seconds, and training performance of 0.00012667.This study focuses on the application and comparison of the epoch, time, performance/MSE training, and performance/MSE testing of variations of the Backpropagation algorithm. The main problem in this study is that the Backpropagation algorithm tends to be slow to reach convergence in obtaining optimum accuracy, requires extensive training data, and the optimization used is less efficient and has performance/MSE which can still be improved to produce better performance/MSE in this research—data prediction process. Determination of the best model for data prediction is seen from the performance/MSE test. This data prediction uses five variations of the Backpropagation algorithm: standard Backpropagation, Resistant Backpropagation, Conjugate Gradient, Fletcher Reeves, and Powell Beale. The research stage begins with processing the avocado production dataset in Indonesia by province from 2016 to 2021. The dataset is first normalized to a value between 0 to 1. The test in this study was carried out using Matlab 2011a. The dataset is divided into two, namely training data and test data. This research's benefit is producing the best model of the Backpropagation algorithm in predicting data with five methods in the Backpropagation algorithm. The test results show that the Resilient Backpropagation method is the best model with a test performance of 0.00543829, training epochs of 1000, training time of 12 seconds, and training performance of 0.00012667.
 
Identify the Color and Shape of Eggplant Using Back Propagation Method
Get PDFCurrently, artificial neural networks are being developed as a tool that can help with human tasks. The main purpose of this study is to identify the structure of an eggplant, and to distinguish the type of eggplant. This study empirically tested the shape and color of several eggplants using the back propagation neural network learning method. The data is obtained from an image that will be entered into the program. The data used in the identification process are two photos containing two types of eggplant, the first eggplant is green and round and the next eggplant is purple and oval. The results of the identification process using this backpropagation from the tests that have been carried out previously, the highest calculation results obtained with the best results using a learning rate of 0.7 and epoch iterations of 500 and producing an accuracy of 73.33%.Currently, artificial neural networks are being developed as a tool that can help human tasks. The main purpose of this study is to identify the structure of an eggplant, in order to distinguish the type of eggplant. This study empirically tested the shape and color of several eggplants using the back propagation neural network learning method. The data is obtained from an image that will be entered into the program. The data used in the identification process are two photos containing two types of eggplant, the first eggplant is green and round and the next eggplant is purple and oval. The results of the identification process using this back propagation from the tests that have been carried out previously, the highest calculation results obtained with the best results using a learning rate of 0.7 and epoch iterations of 500 and producing an accuracy of 73.33%
House Prices Segmentation Using Gaussian Mixture Model-Based Clustering
Get PDFHouse is a place for humans to live and the main necessity for humans. For years, the need for houses is increasing and varied so it affects the selling price of the house. Therefore, more research is needed to learn about the selling price of houses. This research is only focusing on house price segmentation in DKI Jakarta using the Gaussian Mixture Model-Based Clustering Method with the Expectation-Maximization algorithm. The goal of this research is to make a house price segmentation model so that we can obtain useful information for the potential buyer. Clustering with GMM utilizes the log-likelihood function to optimize the GMM parameters. The result of this research is housed in DKI Jakarta and can be segmented into 3 different clusters. The first cluster is for the low-profile houses. The second cluster is for the mid-profile houses. The third cluster is for high-profile houses. The silhouette score that was produced by the clustering method is 0.60866 meaning that this score is quite good because it’s close to a value of 1.House is a place for humans to live and a main necessity for humans. For years, the need for houses is increasing and varied so that it affects the selling price of the house. Therefore, more research is needed to learn about the selling price of houses. This research is only focusing on house price segmentation in DKI Jakarta using the Gaussian Mixture Model-Based Clustering Method with the Expectation-Maximization algorithm. The goal of this research is to make a house price segmentation model so that we can obtain useful information for the potential buyer. Clustering with GMM utilize the log-likelihood function to optimize the GMM parameters. The result of this research is houses in DKI Jakarta can be segmented into 3 different clusters. The first cluster is for the low-profile houses. The second cluster is for the mid-profile houses. The third cluster is for the high-profile houses. The silhouette score that was produced by the clustering method is 0.60866 meaning that this score is quite good because it’s close to a value of 1
Information Technology Governance Awareness: A Proposed Formula for Assessment
No full textThis article aims to provide a proposed formula that can be used to measure the level of success in the practice of Information Technology Governance. To obtain this formulation, in-depth surveys and interviews involving several experts are needed. The calculation results show that organization G has an awareness value of 93 (good) with a maturity value of 3.13. On the other hand, organization E has an awareness value of 70 (medium) with a maturity value of 2.60. This proposed formula can be used as an alternative way to determine the level of success of an organization in the practice of Information Technology Governance by knowing the level of awareness. So far, to determine the level of success in implementing IT Governance practices in an organization, the method used is to calculate the maturity level that refers to COBIT best practices, which only focus on objects but do not focus on subjects (stakeholders) in the organization.
This article aims to provide a proposed formula that can be used to measure the level of success in the practice of Information Technology Governance. To obtain this formulation, in-depth surveys and interviews involving several experts are needed. The calculation results show that organization G has an awareness value of 93 (good) with a maturity value of 3.13. On the other hand, organization E has an awareness value of 70 (medium) with a maturity value of 2.60. This proposed formula can be used as an alternative way to determine the level of success of an organization in the practice of Information Technology Governance by knowing the level of awareness. So far, to determine the level of success in implementing IT Governance practices in an organization, the method used is to calculate the maturity level that refers to COBIT best practices, which only focus on objects but do not focus on subjects (stakeholders) in the organization
Hate Speech Classification on Twitter Using Support Vector Machine
No full textNowadays social media has become a place for peoples to express their opinions, there are many ways that can be done to express both positive and negative opinions. Hate speech is one of the problems that we find quite a lot in cyberspace, that things can be detrimental to many parties. Twitter as one of social media, can be used as a source of analysis about people's behavior in cyberspace. Many of our society that unconsciously act of hate speech on social media, therefore this study finds out how people's behavior patterns in cyberspace and the main issue of hate speech on a particular topic and time period by classify it into five classes, namely ethnicity, religion, race, inter-groups and neutral using Support Vector Machine. In this study also compares three kernel that common to use and the result is the system can classify hate speech by using RBF kernel and got the highest result with 93% accuracy on 700 data train and 300 data test.Saat ini media sosial telah menjadi sarana bagi masyarakat untuk menyampaikan pendapat yang mereka miliki, terdapat banyak cara yang dapat dilakukan untuk menyampaikan pendapat tersebut dari yang bersifat positif maupun negatif. Ujaran kebencian merupakan salah satu masalah yang cukup banyak kita temukan di dunia maya, perilaku tersebut dapat merugikan pihak tertentu. Twitter sebagai salah satu platform yang dapat digunakan untuk menyampaikan opini bisa dijadikan sebagai sumber analisis tentang perilaku masyarakat di dunia maya. Banyak dari masyarakat kita yang tidak sadar bahwa mereka telah melakukan tindakan ujaran kebencian di dunia maya, maka dari itu penelitian ini dibuat untuk mengetahui bagaimana pola perilaku masyarakat di dunia maya dan hal yang menjadi isu utama dalam masalah ujaran kebencian pada topik dan kurun waktu tertentu dengan membaginya menjadi lima kelas yaitu suku, agama, ras, antar golongan dan netral menggunakan Support Vector Machine. Dalam penelitian ini juga dilakukan pengujian dengan membandingkan tiga buah kernel yang umum dipakai, dan hasilnya menunjukan bahwa sistem dapat melakukan klasifikasi ujaran kebencian dengan penggunaan kernel RBF yang menghasilkan nilai akurasi paling tinggi yaitu 93% dengan komposisi data latih sebanyak 700 data dan data uji sebanyak 300 data
Analisis Kesiapan Masyarakat Pada Penerapan Smart City di Sosial Media Menggunakan SNA
No full textSmart People, which means smart city residents or people, not only refers to one's education but also the quality of social interactions that are formed. This Social Network Analysis (SNA) emphasizes the relationship between actors / users rather than the attributes of these actors. This analysis aims to see whether the people of Pekanbaru are ready to face changes to a Smart City. Pekanbaru is a civil city that will build a Smart City, with a concept that adopts 6 pillars, one of which is Smart People. There are 720,000 Twitter users in Pekanbaru City, while the people who actively interact are only 227 users or around 0.031%. Meanwhile, a city that can be said to be ready should be around 60-80% of active users who provide opinions or comments to the government of Pekanbaru City. From this research, it can be concluded that the people of Pekanbaru City are not ready to face Smart City Madani as seen from the interaction of the community on social media Twitter.Smart People yang berarti penduduk atau masyarakat Kota yang cerdas, tidak hanya mengacu pada edukasi seseorang tapi juga kualitas interaksi sosial yang terbentuk. Social Network Analysis (SNA) ini lebih menekankan pada hubungan antar aktor/user dari pada atribut actor tersebut. Analisa ini bertujuan untuk melihat apakah masyarakat pekanbaru siap dalam menghadapi perubahan menuju Smart City.Pekanbaru merupakan Kota madani yang akan membangun Smart City, dengan konsep yang diusung mengadopsi 6 pilar, salah satunya adalah Smart People. Terdapat 720.000 user pengguna twitter di Kota Pekanbaru sedangkan masyarakat yang aktif berinteraksi hanya terdapat 227 user atau sekitar 0.031%. Sedangkan Kota yang bisa dikatakan siap harus sekitar 60-80% dari user aktif yang memberikan opini atau komentar pada pemerintah Kota Pekanbaru. Dari penelitian ini di dapat bahwa masyarakat Kota Pekanbaru Belum Siap dalam menghadapai Smart City Madani dilihat dari interaksi masyarakat dimedia sosial Twitter
Classification of Malaria Complication Using CART (Classification and Regression Tree) and Naïve Bayes
Get PDFMalaria is a disease caused by the Plasmodium parasite that transmitted by female Anopheles mosquitoes. Malaria can become a dangerous disease if late have the medical treatment. The late medical treatment happened because of misdiagnosis and lack of medical staff, especially in the countryside. This problem can cause severe malaria that has complications. This study creates a system prediction to classify the severe malaria disease using Classification and Regression Tree (CART) method and the probability of malaria complication using Naïve Bayes method. The first step of this study is classifying the patients that have symptom are infected severe malaria or not based on the model that has been built. The next step, if the patient classified severe malaria then the data predicted if there any probability of complication by the malaria. There are 8 possibilities of complication malaria which are convulsion, hypoglycemia, hyperpyrexia, and the combinations of these four. The first step will evaluate by using F-score, precision and recall while the second step will evaluate by using accuracy. The highest result F-score, precision and recall are 0.551, 0.471 and 0.717. The highest accuracy 81.2% which predicted the complication is Hypoglycemia.Malaria is a disease caused by the Plasmodium parasite that transmitted by female Anopheles mosquitoes. Malaria can become a dangerous disease if late have the medical treatment. The late medical treatment happened because of misdiagnosis and lack of medical staff, especially in the countryside. This problem can cause severe malaria that has complications. This study creates a system prediction to classify the severe malaria disease using Classification and Regression Tree (CART) method and the probability of malaria complication using Naïve Bayes method. The first step of this study is classifying the patients that have symptom are infected severe malaria or not based on the model that has been built. The next step, if the patient classified severe malaria then the data predicted if there any probability of complication by the malaria. There are 8 possibilities of complication malaria which are convulsion, hypoglycemia, hyperpyrexia, and the combinations of these four. The first step will evaluate by using F-score, precision and recall while the second step will evaluate by using accuracy. The highest result F-score, precision and recall are 0.551, 0.471 and 0.717. The highest accuracy 81.2% which predicted the complication is Hypoglycemia
Analisis Perbandingan Algoritma Optimasi pada Random Forest untuk Klasifikasi Data Bank Marketing
No full textThe world of banking requires a marketer to be able to reduce the risk of borrowing by keeping his customers from occurring non-performing loans. One way to reduce this risk is by using data mining techniques. Data mining provides a powerful technique for finding meaningful and useful information from large amounts of data by way of classification. The classification algorithm that can be used to handle imbalance problems can use the Random Forest (RF) algorithm. However, several references state that an optimization algorithm is needed to improve the classification results of the RF algorithm. Optimization of the RF algorithm can be done using Bagging and Genetic Algorithm (GA). This study aims to classify Bank Marketing data in the form of loan application receipts, which data is taken from the www.data.world site. Classification is carried out using the RF algorithm to obtain a predictive model for loan application acceptance with optimal accuracy. This study will also compare the use of optimization in the RF algorithm with Bagging and Genetic Algorithms. Based on the tests that have been done, the results show that the most optimal performance of the classification of Bank Marketing data is by using the RF algorithm with an accuracy of 88.30%, AUC (+) of 0.500 and AUC (-) of 0.000. The optimization of Bagging and Genetic Algorithm has not been able to improve the performance of the RF algorithm for classification of Bank Marketing data.
Dunia perbankan mengharuskan seorang marketing agar mampu mengurangi resiko peminjaman dengan cara menjaga nasabahnya agar tidak terjadi kredit bermasalah. Salah satu cara untuk mengurangi resiko tersebut adalah dengan menggunakan teknik data mining. Data mining menyediakan teknik yang kuat untuk menemukan informasi yang bermakna dan berguna dari sejumlah besar data dengan cara klasifikasi. Algoritma klasifikas yang dapat digunakan untuk menangani masalah imbalance dapat menggunakan algoritma Random Forest (RF). Namun demikian beberapa referensi menyebutkan bahwa diperlukan algoritma optimasi guna meningkatkan hasil klasifikasi dari algoritma RF. Optimasi algoritma RF dapat dilakukan dengan menggunakan Bagging dan Genetic Algorithm (GA). Penelitian ini bertujuan untuk mengklasifikasikan data Bank Marketing berupa penerimaan pengajuan pinjaman yang mana datanya diambil dari situs www.data.world. Klasifikasi dilakukan dengan menggunakan algoritma RF untuk memperoleh model prediksi penerimaan pengajuan pinjaman dengan akurasi yang optimal. Penelitian ini juga akan membandingkan penggunaan optimasi pada algoritma RF dengan Bagging dan Genetic Algorithm. Berdasarkan pengujian yang telah dilakukan diperoleh hasil bahwa performa paling optimal dari klasifikasi data Bank Marketing adalah dengan menggunakan algoritma RF dengan akurasi sebesar 88,30%, AUC (+) sebesar 0,500 dan AUC(-) sebesar 0,000. Adapun penggunaan optimasi Bagging dan Genetic Algorithm ternyata belum mampu meningkatkan performa dari algoritma RF untuk klasifikasi data Bank Marketing.
 
Optimasi Akurasi Metode Convolutional Neural Network untuk Identifikasi Jenis Sampah
Get PDFWaste is goods / materials that have no value in the scope of production, where in some cases the waste is disposed of carelessly and can damage the environment. The Indonesian government in 2019 recorded waste reaching 66-67 million tons, which is higher than the previous year, which was 64 million tons. Waste is differentiated based on its type, namely organic and anorganic waste. In the field of computer science, the process of sensing the type waste can be done using a camera and the Convolutional Neural Networks (CNN) method, which is a type of neural network that works by receiving input in the form of images. The input will be trained using CNN architecture so that it will produce output that can recognize the object being inputted. This study optimizes the use of the CNN method to obtain accurate results in identifying types of waste. Optimization is done by adding several hyperparameters to the CNN architecture. By adding hyperparameters, the accuracy value is 91.2%. Meanwhile, if the hyperparameter is not used, the accuracy value is only 67.6%. There are three hyperparameters used to increase the accuracy value of the model. They are dropout, padding, and stride. 20% increase in dropout to increase training overfit. Whereas padding and stride are used to speed up the model training process.Sampah merupakan barang/bahan yang tidak memiliki nilai dalam lingkup produksi, dimana dalam beberapa kasus sampah dibuang sembarangan dan dapat merusak lingkungan. Pemerintah Indonesia tahun 2019 mencatat sampah mencapai 66-67 juta ton, dimana lebih tinggi dibandingkan tahun sebelumnya yaitu 64 juta ton. Sampah dibedakan berdasarkan jenisnya yaitu sampah organik dan anorganik. Pada bidang ilmu komputer, proses penginderaan jenis dan bentuk sampah dapat dilakukan menggunakan kamera dan metode Convolutional Neural Networks (CNN) yang merupakan jenis neural network yang bekerja dengan cara menerima masukan berupa citra. Masukan tersebut akan di training menggunakan arsitekur CNN sehingga akan menghasilkan output yang dapat mengenali objek yang diinputkan. Pada penelitian ini dilakukan optimasi penggunaan metode CNN untuk mendapatkan hasil yang akurat dalam mengidentifikasi jenis sampah. Optimasi dilakukan dengan menambah beberapa hyperparameter pada arsitektur CNN. Dengan menambahkan hyperparameter diperoleh nilai akurasi yang tinggi yaitu 91,2%. Sedangkan apabila tidak menggunakan hyperparameter nilai akurasi hanya sebesar 67,6%. Terdapat tiga hyperparameter yang digunakan untuk menaikan nilai akurasi model yaitu dropout, padding, dan stride. Penambahan dropout sebesar 20% untuk meningkatkatkan overfitting saat pelatihan. Sedangkan padding dan stride digunakan untuk mempercepat proses pelatihan model