Jurnal Transformatika
Not a member yet
319 research outputs found
Sort by
Classification Covid-19 Based on X-Ray Using GLCM and ANN Backpropagation
No full textCoronavirus (Covid-19) is a disease that belongs to a large family of disorders that can cause mild to severe symptoms. The coronaviruses Middle East Respiratory Syndrome (MERS) and Serious Acute Respiratory Syndrome (SARS) are two types of coronaviruses that cause severe illness (SARS). According to WHO estimates as of December 17, 2021, Covid-19 has infected about 271,963,258 individuals, with a death rate of 5,331,019 cases. Hospitals only have a limited number of Covid-19 test kits because of the daily increase in cases. As a result, to prevent the spread of Covid-19 among persons, it is necessary to develop an automatic detection method as quickly as feasible, as well as other diagnosis options. The goal of this research was to employ GLCM to extract features and the Backpropagation Neural Network classification technique to automatically develop a Covid-19 diagnosis system by classifying the lungs into two groups: normal lungs and Covid-19 lungs. Pre-processing, segmentation, feature extraction, and classification were all used in this study\u27s lung classification method. The accuracy of the test findings was assessed to be 85%. The sensitivity value for the normal class was 92.5%, whereas the sensitivity value for the Covid-19 class was 77.5%. The specification value for the normal class was 22.5%, whereas the specification value for the covid-19 class was 7.5%. It may be deduced from the accuracy, sensitivity, and specification percentages that the developed system is capable of categorizing lungs using X-Ray lung pictures.Coronavirus (Covid-19) is a disease that belongs to a large family of disorders that can cause mild to severe symptoms. The coronaviruses Middle East Respiratory Syndrome (MERS) and Serious Acute Respiratory Syndrome (SARS) are two types of coronaviruses that cause severe illness (SARS). According to WHO estimates as of December 17, 2021, Covid-19 has infected about 271,963,258 individuals, with a death rate of 5,331,019 cases. Hospitals only have a limited number of Covid-19 test kits because of the daily increase in cases. As a result, to prevent the spread of Covid-19 among persons, it is necessary to develop an automatic detection method as quickly as feasible, as well as other diagnosis options. The goal of this research was to employ GLCM to extract features and the Backpropagation Neural Network classification technique to automatically develop a Covid-19 diagnosis system by classifying the lungs into two groups: normal lungs and Covid-19 lungs. Pre-processing, segmentation, feature extraction, and classification were all used in this study\u27s lung classification method. The accuracy of the test findings was assessed to be 85%. The sensitivity value for the normal class was 92.5%, whereas the sensitivity value for the Covid-19 class was 77.5%. The specification value for the normal class was 22.5%, whereas the specification value for the covid-19 class was 7.5%. It may be deduced from the accuracy, sensitivity, and specification percentages that the developed system is capable of categorizing lungs using X-Ray lung pictures
USABILITY EVALUATION MODEL USING CONFIRMATORY FACTOR ANALYSIS ON ONLINE BUYING SITES
Full text linkThe data obtained from the Top brand index shows that there are 3 market leaders of online buying and selling sites in Indonesia, namely lazada.co.id; shopee.co.id and tokopedia.com. This makes the reason for choosing the object of research with the aim of research to find out the usability evaluation model using confirmatory factor analysis. The research data were obtained by distributing questionnaires as many as 150 respondents to 3 market leader online trading sites in Indonesia, namely lazada.co.id; shopee.co.id and tokopedia.com. The five components that will be used to turn into a questionnaire are learning ability, memory, efficiency, errors, satisfaction. The results showed that the design of the model in solving this particular problem is needed to measure the level of suitability of the use of the technology being made, so that research that has been carried out using the CFA method for the usability evaluation model can be concluded that by using 5 latent variables and 22 constituent indicators, resulting latent variables which are very influential in evaluating usability on online buying and selling sites. The results of the data analysis on questionnaire filling conducted by respondents were based on five variables, namely learnability, memorability, efficiency, errors and satisfaction, namely the most important factors in usability evaluation were memorability, efficiency, errors, with each index value of 5.60 in the very high category. Meanwhile, the Learnability factor has the lowest score of 5.56
Penerapan Metode Analytical Hierachy Process (AHP) Untuk Pemilihan Guru Terbaik di Sekolah SD Negeri Periuk 3
No full textSekolah perlu melakukan peningkatan kinerja guru dengan mengadakan pemilihan guru terbaik. Karena itu, penelitian ini mengulas sistem pendukung keputusan pemilihan guru terbaik pada SD Negeri Periuk 3 Tangerang. Namun ada permasalahan yang terjadi, antara lain banyaknya jumlah guru sehingga sulit memilih guru terbaik, belum terdapatnya metode yang dipergunakan sebagai pembobotan nilai kriteria, dan juga belum adanya Sistem Pendukung Keputusan (SPK) yang membantu kepala sekolah. Untuk itu, sebagai solusi dari masalah yang ada, diusulkan pembangunan sitem pendukung keputusan pemilihan guru terbaik, supaya hasil yang dicapai semakin objektif dan dapat meningkatkan kualitas mengajar di sekolah. Sistem ini di bangun dengan menerapkan metode Analytical Hierarchy Process (AHP) juga menggunakan beberapa kriteria untuk menetapkan bobot yaitu nilai kehadiran, nilai kedisiplinan, nilai berpakaian, nilai skill. Selain itu, metode AHP juga dapat memberikan urutan prioritas alternatif. Penelitian ini bertujuan untuk membangun SPK pemilihan guru terbaik yang mempermudah pihak sekolah dalam penentuan guru terbaik serta dapat mengurangi tingkat kesalahan dalam penilaian, dan menghasilkan metode yang tepat untuk pemilihan guru terbaik. Selain itu hasil akhir alternatif guru terbaik yang terpilih, juga dapat menjadi motivasi kepada guru untuk meningkatkan kinerja sehingga menghasilkan kualitas kerja yang baik.
Identifikasi Wilayah Resiko Kerusakan Lahan Terbangun Sebagai Dampak Tsunami Berdasarkan Analisis Building Indices
Full text linkIndonesia has a very large water area and there is the territory that is the confluence of the earth\u27s slabs. It can allowing the occurrence of tsunami natural disasters. The study aims to find out which areas have risks the highest and the lowest land damage. The data used in this study were satellite images taken from 2014-2021 with coverage area in Kulon Progo Regency which consists of 12 sub-districts. This study used indexes vegetation UI, NDBI, IBI, EBBI. With an ANN algorithm get results which is quite accurate with an accuracy rate of 94.50%. Predictions states that the area has a high risk of damage due to the tsunami, namely the villages of Jingkaran, Sindutan, Palihan, Glagah, Karangwuluh, Janten, Temon Kulon, Kedundang, Sogan, Kalidengen, Ngestiharjo, Depok, Kanoman, Panjatan, Wahyuharjo, Pandowan, Nomporejo, and, Kranggan.
IMPLEMENTASI ALGORITMA LINEAR REGRESSION UNTUK PREDIKSI HARGA SAHAM PT. ANEKA TAMBANG TBK
No full textStock investments that provide high returns, but the higher the benefits offered, the higher the risk that will be faced in investing, especially if it is not supported by knowledge of analyzing stocks. This study utilizes the Data Mining prediction technique with the Linear Regression algorithm on the shares of PT. Aneka Tambang Tbk or ANTM. The dataset that will be used is downloaded through the Yahoo Finance website in the period January 2016 - March 2021. In this study the analytical method used is SEMMA (Sample, Explore, Modify, Model, Assess). With RapidMiner Studio 9.9 tools. The result of testing the RMSE (Root Mean Squared Error) value is 17.135, MSE (Mean Squared Error) is 293.599 and the MAPE (Mean Absolute Percentage Error) value is 1.87%. Based on the MAPE, the accuracy of the Linear Regression algorithm in predicting the stock price of PT. Aneka Tambang Tbk provides high-accuracy predictions
Content Classification based-on Latent Semantic Analysis and Support Vector Machine (LSA-SVM)
Full text linkThe diversity of the content of a web page can have a negative impact if used by the wrong user. Almost a half of internet users are children. Therefore, it is important to classify web pages to find out which pages are worthy of being seen by children and that are not feasible. One method that can be used is the Support Vector Machine (SVM) algorithm. SVM is a binary classification whose working principle is to find the best hyperplane to separate the two classes. To obtain better classification accuracy, the SVM is combined with the Latent Semantic Analysis (LSA) algorithm. The data used in this study were taken from the DMOZ web data which has been classified into two categories. The data is then entered into the pre-processing stage for further feature extraction using LSA. The LSA algorithm is used to find out the semantic similarities of words and text contained in web pages. The results of feature extraction are then classified using SVM with RBF kernel. Based on the testing result, we obtain a classification accuracy of 64%
Klasifikasi Resiko Tsunami di Daerah Pantai Selatan Jawa Tengah dengan Menerapkan Algoritma SVM (Studi Kasus Kab. Kebumen)
Full text linkTsunami adalah salah satu bencana alam yang dapat terjadi kapan saja yang keberadaannya tidak dapat dihindari. Kabupaten Kebumen adalah salah satu daerah yang bertempat di bagian pantai selatan Provinsi Jawa Tengah yang memiliki risiko terkenanya bencana tsunami. Hal ini terjadi akibat lokasi dari kabupaten kebumen yang berada dekat dengan perairan dan memiliki banyak luas wilayah pada dataran rendah. Penelitian ini dilakukan agar dapat memberikan informasi mengenai risiko yang dapat terjadi pada wilayah tersebut dengan melakukan prediksi dan klasifikasi dengan menggunakan data dari Citra Landsat-8 OLI yang diambil dari USGS serta memanfaatkan 5 indeks vegetasi yaitu NDVI, NDBI, NDWI, MNDWI, dan WI dengan penerapan metode Support Vector Machine (SVM) yang merupakan salah satu metode machine learning yang digunakan untuk klasifikasi. Pada penelitian ini didapatkan ada 11 desa yang diprediksi berisiko tinggi terkena tsunami dengan nilai akurasi yang didapatkan sebesar 0.945 dan nilai kappa sebesar 0.914 dimana semakin tinggi nilai yang didapatkan maka semakin tinggi juga ketepatan hasil prediksi yang diperoleh
Analisis Manajemen Risiko Sistem Informasi Menggunakan Information System Success Model (ISSM)
Full text linkPT PGAS Solution merupakan perusahaan yang didirikan pada tahun 2009 sebagai anak perusahaan dari PGN yang berfokus pada aspek teknik dan operasional di bidang gas. Perusahaan ini memiliki satuan kerja yang bekerja untuk mengelola informasi mengenai Manajemen Risiko, pengelolaan tersebut menggunakan aplikasi sistem informasi Risk Manajemen Online (Risoles). Pada penelitian ini menggunakan kuesioner berdasarkan 6(enam ) dimensi berdasarkan metode ISSM (Information System Success Model) yaitu System Quality, Information Quality, Use, User Satisfaction, Individual Impact dan Organizational Impact. Kemudian jawaban dari responden di olah menggunakan bantuan software SmartPLS.Hasil dari penelitian ini Menunjukan bahwa terdapat pengaruh Information Quality terhadap Intention to Use/Actual Use dan adanya pengaruh intention to use terhadap net benefit, dimana hasil ini akan menjadi penilaian terhadap pengaruh Aplikasi system informasi terhadap suatu keputusan dalam kinerja individu untuk meningkatkan kualitas kinerja pegawai dan dapat dijadikan acuan dalam pengembangan aplikasi Risoles pada perusahaan PT PGAS Solution
Sistem Pendeteksi Banjir dengan Sensor Ultrasonic berbasis Mikrokontroller di kota Pangkalpinang
Full text linkProvide convenience and information to BPBDs and people living in flood-prone environments that are usually unknown to homeowners without getting information about the arrival of floods in the area. So we need a system that can detect the water level of the river. For that, it is necessary to install a device that can read the water level in the form of an ultrasonic sensor and a microcontroller as well as a GSM SIM900A module. The research method used is literature study, problem analysis, system design and implementation.The results achieved in this study are able to help provide information in the form of sirens and SMS services in the form of short messages about flood alert levels that are sent to the community and the BPBD
The Application of Na ve Bayes Classifier Based Feature Selection on Analysis of Online Learning Sentiment in Online Media
Full text linkThere are problems that still exist in online learning including limited-reach networks, inadequate facilities and infrastructure, and others. This study discussed the analysis of sentiment which used the Na ve Bayes Classifier (NBC) method with XGBoost feature selection as a performance improvement that took data from news portals. The results of this study showed that graph data on the application of online learning forms in Indonesia had a "Negative" opinion. Performance testing of the NBC method based on XGBoost feature selection was conducted four times. The first experiment resulted in an accuracy value of 60.18% with 50/50 split data. The next experiment had an accuracy value of 56.92% with 70/30 split data. After that, the third experiment resulted in an accuracy value of 65.90% with 80/20 split data. The result of the last experiment was an accuracy value of 63.63% with 90/10 split data. After using XGBoost feature selection, it produced an accuracy of 60.18%, 67.69%, 70.45%, and 77.27%. The study also produced the highest average score at 10-Fold Cross-Validation in the second trial with a score of 65.62%