Telematika
Not a member yet
319 research outputs found
Sort by
Multiplatform-Based Digital Market Designs as Marketing and Sales Media of MSME Products in Pleret Village
Get PDFPurpose : This study aims to design a multiplatform-based digital market to help rural MSMEs (Micro, Small, Medium Enterprises) market and sell their products in order to increase the competitiveness of MSMEs.Design/methodology/approach : The design of the digital platform of Ngedolke.com used a design thinking strategy. This study used the black box method to test the functional suitability of the developed application.Findings/result : This study produces an analysis and system design that is used to develop the digital platform of Ngedolke.com. Thus, it can be used to develop the system further.Originality/value/state of the art : The difference between this study and previous studies lies in the use of system design of a design thinking strategy. Besides, the technology used is multiplatform-based
Implementation of Convolutional Neural Network (CNN) in Facial Expression Recognition
Get PDFTujuan: Membantu pengajar melakukan monitoring emosi siswa dengan menerapkan metode Convolutional Neural Network pada aplikasi, serta mengetahui akurasi dalam melakukan pengenalan ekspresi wajah.Perancangan/metode/pendekatan: Menggunakan Convolutional Neural Network untuk mengklasifikasi pengolahan berupa citra. Pengembangan sistem menggunakan metode prototype.Hasil: Berdasarkan hasil pengujian yang dilakukan dengan menggunakan 3589 data ekspresi dasar manusia mendapatkan nilai akurasi sebesar 70,46%, nilai presisi sebesar 71% dan nilai recall sebesar 70%.Keaslian/ state of the art: Berdasarkan penelitian sebelumnya, penelitian ini mempunyai karakteristik yang relatif serupa dalam tema penelitian. Namun memiliki perbedaan pada metode penelitan, perangkat yang digunakan, dan hasil keluaran penelitian.Pada penelitian sebelumnya, dengan objek yang sama yaitu wajah dan emosi wajah, pada metode yang digunakan, perangkat dalam pengambilan citra emosi dan wajah, serta langkah-langkah dalam prosesnya pun berbeda. Pada penelitian ini emosi pada wajah diidentifikasi melalui citra yang diambil secara real-time menggunakan kamera dan dengan menerapkan metode Convolutional Neural Network dengan arsitektur visual group geometry (VGG) dengan 11, 13, 16 dan 19 lapisan yang akan menghasilkan probabilitas ekspresi dalam 7 ekspresi dasar manusia beserta kategorinya
Multimedia Mobile Application of National Heroes History Learning for Children\u27s Character Education
Get PDFPurpose: develope a multimedia application about the history of national heroes from Indonesia.Design/methodology/approach: the method used is the UCD (User Centered Design) method.Findings/result: this multimedia mobile application of national heroes history learning for children\u27s character education has succeeded in meeting user needs.Originality/value/state of the art: a multimedia application about the history of national heroes from Indonesia
Learning and Playing in Early Childhood with Augmented Reality Technology
Get PDFPurpose: Helping the learning process in early childhood through playing and learning activities with Augmented Reality technology.Design/methodology/approach: Using Augmented Reality technology with the Iterative Rapid Paper Prototype system development methodFindings/result: Based on tests conducted on 5 types of android devices, 10 samples of early childhood participants (4-5 years) and 5 groups of objects consisting of 10 types resulted in an increase in learning ability of 33.35% which was sourced from the measurement of the correct answers that were successfully obtained. between learning methods through pictures and learning using Augmented Reality technologyOriginality/value/state of the art: In previous research, the learning model was carried out on elementary school children (aged 6 years and over) and without the implementation of Augmented Reality technolog
Pencak Silat Tournament Information System
Get PDFPurpose:This research was conducted to help manage the implementation of the pencak silat championship. So that the championship can run in an orderly and professional manner.Design/methodology/approach:This research went through several stages, starting from data collection, system requirements analysis, design, implementation, and system testing.Findings/result:Website-based information system for pencak silat tournament.Originality/value/state of the art:Pencak silat is a martial arts rich in techniques, benefits, and carries noble values that should be preserved as the Indonesian nation\u27s successor. To preserve the existence of pencak silat in Indonesia, various pencak silat competitions were held in several cities in Indonesia. In the championship implementation, several things can disrupt the course of the matches. Of course, it will make the championship unprofessional. For this reason, along with the development of science and technology, a system was created that would help manage the implementation of the pencak silat championship so that the championship can run in an orderly and professional manner
Development Of Executive Information Systems Of Cirebon City Government (Case Study: Department Of Communication, Informatics And Statistics)
Get PDFPurpose: Developing an executive information system to meet the information needs of the Mayor, Deputy Mayor, Regional Secretary and the heads of SKPD within the Cirebon City Government.Design / method / approach: Using the drill down method for solving information on executive information systems and the GRAPPLE system development methodResult: The development of an executive information system in Cirebon city government has assisted the executive, consisting of mayors, deputy mayors and regional secretaries and middle executives consisting of skpd within the Cirebon city government. Cirebon city government executive information system consists of five sectors in the city of Cirebon, namely economy, health, population, education and government. The results of the validation testing are 100% and the average user acceptance testing results are 85.29%.Authenticity / state of the art: Based on previous research, this study has the same characteristics but in the development of executive information systems it has differences in objects and methods of software development
Sentiment Analysis On YouTube Comments Using Word2Vec and Random Forest
Get PDFPurpose: This study aims to determine the accuracy of sentiment classification using the Random-Forest, and Word2Vec Skip-gram used for features extraction. Word2Vec is one of the effective methods that represent aspects of word meaning and, it helps to improve sentiment classification accuracy.Methodology: The research data consists of 31947 comments downloaded from the YouTube channel for the 2019 presidential election debate. The dataset consists of 23612 positive comments and 8335 negative comments. To avoid bias, we balance the amount of positive and negative data using oversampling. We use Skip-gram to extract features word. The Skip-gram will produce several features around the word the context (input word). Each of these features contains a weight. The feature weight of each comment is calculated by an average-based approach. Random Forest is used to building a sentiment classification model. Experiments were carried out several times with different epoch and window parameters. The performance of each model experiment was measured by cross-validation.Result: Experiments using epochs 1, 5, and 20 and window sizes of 3, 5, and 10, obtain the average accuracy of the model is 90.1% to 91%. However, the results of testing reach an accuracy between 88.77% and 89.05%. But accuracy of the model little bit lower than the accuracy model also was not significant. In the next experiment, it recommended using the number of epochs and the window size greater than twenty epochs and ten windows, so that accuracy increasing significantly.Value: The number of epoch and window sizes on the Skip-Gram affect accuracy. More and more epoch and window sizes affect increasing the accuracy
Good Morning to Good Night Greeting Classification Using Mel Frequency Cepstral Coefficient (MFCC) Feature Extraction and Frame Feature Selection
Get PDFPurpose:Select the right features on the frame for good accuracyDesign/methodology/approach:Extraction of Mel Frequency Cepstral Coefficient (MFCC) Features and Selection of Dominant Weight Normalized (DWN) FeaturesFindings/result:The accuracy results show that the MFCC method with the 9th frame selection has a higher accuracy rate of 85% compared to other frames.Originality/value/state of the art:Selection of the appropriate features on the frame
Development of a Group Decision Support System with the Multi-Stage Multi-Attribute Group Decision Making (MS-MAGDM) Method on the Intelligent Warehouse Management System
Get PDFPurpose: to find a solution with MS-DAGDM for the problem of different criteria used by decision maker at each stage.Design/methodology/approach: This research was conducted using literature review with a study of the theory of decision-making methods, group decisions, suplier selection processes, and factors that influence decisions in the context of warehousing and MS-MAGDM to solve the problems.Findings/result: This research find that GDSS prototypes which have four methods in making decisions. First, Analytical Hierarchy Process for weighting the division head level. Second, TOPSIS for divison head level decisions. Third, Hybrid Weight Averaging (HWA) manager level. Fourth, Time Weight Averaging (TWA) for manager level decisions.Originality/value/state of the art:The decision-making model of the GDSS system in this study combines four methods at each level of management. The section head level uses AHP for the level weighting and TOPSIS for decision making. Level managers use Hybrid Weight Averaging (HWA) weighting and Time Weight Averaging (TWA) for decisions. The combination of these methods is carried out using a Poisson distribution, for HWA and TWA operators to combine individual decisions into group decisions. Tujuan: Fokus penelitian ini adalah mencari solusi dengan MS-MAGDM untuk permasalahan perbedaan kriteria yang dipergunakan pembuat keputusan dalam setiap stage.Perancangan/metode/pendekatan: Metode yang digunakan yaitu kajian kepustakaan dengan kajian terhadap teori metode pembuatan keputusan, keputusan kelompok, proses pemilihan supplier, dan faktor yang berpengaruh pada keputusan dalam konteks pergudangan serta MS-MAGDM untuk menyelesaikan permasalahan tersebut.Hasil: Hasil penelitian ini berupa purwarupa GDSS yang memiliki 4 metode dalam pembuatan keputusan yaitu Analytical Hierarchi Process (AHP) untuk pembobotan level kepala bagian, TOPSIS untuk keputusan level kepala bagian, Hybrid Weight Averaging (HWA) pembobotan pada level manager dan Time Weight Averaging (TWA) untuk keputusan level managerKeaslian/ state of the art:Model pengambilan keputusan sistem GDSS penelitian ini menggabungkan 4 metode pada setiap tingkatan manajemen. Level kepala bagian menggunakan AHP untuk pembobotan level dan TOPSIS untuk pembuatan keputusan. Level manager menggunakan Hybrid Weight Averaging (HWA) pembobotan dan Time Weight Averaging (TWA) untuk keputusan. Penggabungan metode dilakukan menggunakan distribusi Poisson, untuk operator HWA dan TWA guna memadukan keputusan individu mejadi keputusan kelompok
Content Based Image Retrieval Using Gray Level Co-Occurrence Matrix to Detect Pneumonia in X-Ray Thorax Image
Get PDFPurpose:This study aims to detect the presence of pneumonia or not in thorax x-ray images using the Gray Level Co-Occurence Matrix (GLCM) method as well as find out the accuracy of the accuracy of pneumonia detection accuracy.Design/methodology/approach:The process of detecting pneumonia in thorax x-ray images can use Content Based Image Retriveal (CBIR). CBIR is an image search method by comparing the input image feature with the image feature in the database. Extraction features x-ray texture of thorax in pneumonia detection using Color Histogram, Discrete Cosine Transform and Gray Level Cooccurence Matrix (GLCM). From the day of extraction the feature will be carried out similarity measurements with database images using Euclidean Distance..Findings/result: The test results showed that the GLCM extraction feature with euclidean distance similarity measurements gained 95% accuracy on 100 training data and 20 test data, with the number of images displayed 6. Whereas when testing using data that has been trained produces 100% accuracy.Originality/value/state of the art:The difference between this study and previous research is in the pre-processing method section of imagery. This pre-processing process, x-ray image of thorax is carried out color histogram and discrete cosine transform process. Then continued the extraction of features using GLCM. The output of this system is the result of detection whether normal or pneumonia. Tujuan:Penelitian ini bertujuan untuk mendeteksi adanya Pneumonia atau tidak pada citra x-ray thorax menggunakan metode Gray Level Co-Occurence Matrix (GLCM) serta mengetahui akurasi tingkat akurasi deteksi pneumonia.Perancangan/metode/pendekatan:Proses deteksi penyakit Pneumonia pada citra x-ray thorax dapat menggunakan Content Based Image Retriveal (CBIR). CBIR adalah suatu metode pencarian citra dengan melakukan perbandingan antara fitur citra input dengan fitur citra yang ada didalam database. Ekstraksi fitur tekstur x-ray thorax dalam deteksi pneumonia menggunakan Color Histogram, Discrete Cosine Transform dan Gray Level Cooccurence Matrix (GLCM). Dari hari ekstraksi fitur tersebut akan dilakukan pengukuran kemiripan dengan citra database menggunakan jarak Euclidean Distance.Hasil:Hasil pengujian menunjukkan bahwa fitur ekstraksi GLCM dengan pengukuran kemiripan Euclidean Distance diperoleh akurasi sebesar 95% pada data latih 100 dan data uji 20, dengan jumlah citra yang ditampilkan 6. Sedangkan bila pengujian menggunakan data yang sudah dilatihkan menghasilkan akurasi 100%.State of the art:Perbedaan penelitian ini dengan penelitian sebelumnya adalah pada bagian metode pre processing citra. Proses pre processing ini, citra x-ray thorax di lakukan proses Color Histogram dan Discrete Cosine Transform. Kemudian dilanjutkan ekstraksi fitur menggunakan GLCM. Output dari sistem ini berupa hasil deteksi apakah normal atau pneumonia