Claim Missing Document
Check
Articles

Found 30 Documents
Search

Tuning Parameter pada Pengendali Logika Fuzzy menggunakan Algoritma Ant Colony Optimization Adriana Fanggidae
Telekontran : Jurnal Ilmiah Telekomunikasi, Kendali dan Elektronika Terapan Vol 7 No 2 (2019): TELEKONTRAN vol 7 no 2 Oktober 2019
Publisher : Program Studi Teknik Elektro, Fakultas Teknik dan Ilmu Komputer, Universitas Komputer Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (787.887 KB) | DOI: 10.34010/telekontran.v7i2.2661

Abstract

The Ant Colony Optimization (ACO) algorithm can be applied in tuning parameters in a Fuzzy Logic Controller (FLC) to control the water level of the process tank. Fuzzy input and output consists of seven membership functions, namely large positive (PB), medium positive (PM) and small positive (PS), zero (Z), small negative (NS), medium negative (NM) and large negative (NB) ). First, the initial FLC parameter is searched, then a graph is generated where the values ​​of the FLC parameter are determined in the range of values ​​between 0 and 1.5 times the initial parameter value. ACO algorithm is used to improve the value of the FLC parameter in order to obtain better performance. The expected controller performance is to minimize the maximum surge (overshoot) and rise time. This system is implemented using the LabVIEW program. Water level data is obtained using a potentiometer sensor. The output from the FLC is connected to the stepper motor to regulate the discharge of water input to the process tank. The test results obtained overshoot and a small rise time, for example, for setpoint 8, the system output performance has an overshoot of 2.5% and a rise time of 8909 ms. ACO algorithm succeeded in increasing system performance compared to system performance if using initial parameters. This increase in performance is due to the ACO algorithm acting as a local search algorithm which will look for better system performance around its initial parameter values. This research successfully demonstrated that the ACO algorithm can be used to do tuning from FLC parameters.
Prediksi Penyakit Pada Balita Usia 2-60 Bulan Menggunakan Metode Certaitainty Factor Afriani Afriani Afriani; Sebastinaus Adi Mola; Adriana Fanggidae
Jurnal Dialektika Informatika (Detika) Vol 3, No 1 (2022): Jurnal Dialektika Informatika(Detika) Vol.3 No.1 Desember 2022
Publisher : Universitas Muria Kudus

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24176/detika.v3i1.9081

Abstract

Masuknya COVID-19 pada April 2020 di Kota Kupang sangat memengaruhi kehidupan masyarakat khususnya di bidang kesehatan, terutama pada balita yang sangat rentan diserang oleh berbagai penyakit. Penurunan kunjungan orang tua yang memeriksakan kondisi anaknya di Puskesmas Oesapa Kota Kupang menurun sebanyak 50% di tahun 2020. Pembatasan waktu kunjungan dari 5 jam menjadi 3 jam perhari dan kekhawatiran orang tua akan penularan COVID-19 yang mungkin terjadi di area puskesmas menjadi penyebab penurunan kunjungan yang dimaksud. Oleh karena itu, untuk menjawab permasalahan tersebut dibutuhkan sistem pakar untuk mendiagnosis penyakit pada balita yang dapat membantu para orang tua dalam melakukan diagnosis penyakit pada balita dengan menerapkan certainty factor. Metode certainty factor digunakan untuk mengakomodasi ketidakpastian dalam proses diagnosis penyakit pada balita. Berdasarkan penelitian yang telah dilakukan, sistem terbukti mampu dan akurat dalam mendiagnosis penyakit pada balita usia 2-60 bulan dengan tingkat akurasi sebesar 98%. 
APLIKASI PENENTUAN GOLONGAN DARAH MANUSIA DENGAN METODE SEED REGION GROWING DAN SELF ORGANIZING MAPS David Wewo; Adriana Fanggidae; Kornelis Letelay
J-Icon : Jurnal Komputer dan Informatika Vol 6 No 1 (2018): Maret 2018
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v6i1.349

Abstract

The blood type of human are divided by four group wich is blood type A, B, O & AB. Artificial Neuron Network can help the identify process for blood type. Self organizing maps is a part of arrtificial neuron network who has function for data training and data clasification. The image data are using by blood clotting and obtained after spilled blood sample with the reagent. The real data image are converted into grayscale image, For taking the characteristic are doing by converted real image to image biner with the treshold more than 80 and smaller than 150, image are taken as much as 12 image of clotted blood and 12 image blood wich does not clot, and the next step will do the training process using self organizing maps. The first testing data are doing by the same test data and same with training data too and the result 100%. The second testing data is doing by 12 blood image test data wich is not the same as data training and the result 83.33%.
PENYANDIAN DATA TEKS MENGGUNAKAN ALGORITMA CIPHER FEED-BACK DAN CHAOTIC SKEW TENT MAP Anita M Sonbay; Adriana Fanggidae; Kornelis Letelay
J-Icon : Jurnal Komputer dan Informatika Vol 5 No 2 (2017): Oktober 2017
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v5i2.360

Abstract

Security for documents that contain confidential text and important for a company, institution, or organization from disorders irresponsible organisation is a necessity or a major requirement that must be done, so we need a software that can protect these vital documents. The combination of algorithms Chiper Feed-back (CFB), Chaos Skew Tent Map (CSTM), and the initial value generation techniques with Session Keys capable of encrypting the text properly. Tests conducted on 26 files doc, docx and txt where each test will be analyzed the correlation, standard deviation and variance, as well as the histogram. Testing is done to the file doc, docx, and txt with a size ≤ 35 kb, 30-60 kb, 61-80 kb and > 80 kb. The test results obtained by the value of the average correlation = 0.163326411, standard deviation = 1.068,070, variance = 1.550.358,492, and the ciphertext histogram that visually looks uniform, making it difficult to analyze the statistical analysis of the character or key.
DETEKSI DAN IDENTIFIKASI BARCODE 2D MENGGUNAKAN METODE EKSTRAKSI CIRI GABOR FILTER DAN IDENTIFIKASI CIRI BACKPROPAGATION NEURAL NETWORK Hepiyana V Runesi; Adriana Fanggidae; Meiton Boru
J-Icon : Jurnal Komputer dan Informatika Vol 6 No 2 (2018): Oktober 2018
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v6i2.511

Abstract

Barcode is a device in the form of a black and white matrix to represent 1 and 0, which aims in storing information. It is divided into two types, namely 1D and 2D barcodes. The different between them is 1D barcode has black and white bars, while 2D barcode has square shape. The method used in this research is grayscaling, floating and screening comprehensive using flood fill pixel reduction algorithm, the perimeter of objects, extraction feature using gabor filter algorithm, the learning method uses backpropagation neural network algorythm, and the identification process using the feedforward method to backpropagation neural network algorythm. The data used in this research is a data of 2D barcode on each of it amounted to 20 users who are taken from the BBM (Blackberry Messenger) contact, due to the lack of data thus a data of the 2D barcode is cropped for 8 times to be the training data and twice to be the test data. The test is done in three stages which the first data set consists of 10 users, the second one consists of 15 users and the last one consists of 20 users. The result of the testing system for those data sets show that the first data set obtains an accuracy of 100%, the second one obtains 93,33% and the last one obtains 66%.
ANALISIS METODE SINGLE-POINT CROSSOVER (SPX), TWO-POINT CROSSOVER (TPX) DAN MULTI-POINT CROSSOVER (MPX) PADA FUNGSI NONLINEAR DUA PEUBAH DENGAN BINARY CODING Adriana Fanggidae
J-Icon : Jurnal Komputer dan Informatika Vol 7 No 1 (2019): Maret 2019
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v7i1.872

Abstract

Algoritma genetika merupakan salah satu algoritma evolusioner yang memiliki 4 tahapan penting yaitu pengkodean, seleksi, crossover dan mutasi. Pada tulisan ini, kinerja dari binary coding pada 3 metode crossover SPX, TPX, dan MPX diuji pada 5 fungsi nonlinear dua peubah. Hasil yang diperoleh menunjukkan metode crossover TPX memberikan kinerja yang lebih baik daripada SPX dan MPX.
ANALISIS METODE CYCLE CROSSOVER (CX) DAN METODE PARTIAL MAPPED CROSSOVER (PMX) PADA PENYELESAIAN KASUS TRAVELING SALESMAN PROBLEM (TSP) Theresia Kolo; Adriana Fanggidae; yelly y nabuasa
J-Icon : Jurnal Komputer dan Informatika Vol 7 No 1 (2019): Maret 2019
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v7i1.882

Abstract

Travelling Salesman Problem (TSP) is a form of a problem in optimizing the search for the shortest route by passing through every city in exactly one time. The problem of searching the shortest route of a location can be solved by using many other optimizing algorithms. In this research, genetics algorithm was used by using two crossover methods namely cycle crossover and partial-mapped crossover. The parameters used were crossover probability and mutation probability, the sum of the city, maximum generation, the sum of the population and also threshold. In this research two testing models were used. In the first one, in order to get the generation and the best fitness it used the 80% consistency stopping criteria, and in the second one, in order to get the best testing time, it used the 100 and 500 maximum generation stopping criteria. The result of the first test showed that PMX method is better than the CX one. This was shown through the 8 times of testing which the result was the best PMX generation was 104,0469 and the CX was 350,4563. The second test resulted that the best testing time of the PMX time was 1,1035 second and the CX method was 2,2374 second, thus, it can be concluded that the solution brought by the PMX method is considered better than the CX.
PENGENALAN POLA SIDIK JARI DENGAN METODE LOCAL BINARY PATTERN DAN LEARNING VECTOR QUANTIZATION Adriana Fanggidae; Dony M Sihotang; Adnan Putra Rihi Pati
J-Icon : Jurnal Komputer dan Informatika Vol 7 No 2 (2019): Oktober 2019
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v7i2.1635

Abstract

Fingerprint is the generic structure in the form of a very detailed pattern and a sign that inherent in human beings. Many biometric systems using fingerprint as input data, because the nature of each individual is different although identical twins and do not change unless got a accident. The method used in this research is image segmentation using Otsu thresholding algorithm, feature extraction using Local Binary Pattern (LBP) algorithm and the learning method using Learning Vector Quantization (LVQ) algorithm. The used data is grayscale fingerprint image with 200x300 pixel and *.jpg extension format. The fingerprint image is composed of 25 people, each person has 6 training data and 2 test data. Experiment of training data and test data conducted for four systems, namely the system with characteristics of LBP = 8, 64, 128 and 256 and their respective uses 2 pieces of data set where data set 1 amounted to 15 people and data set 2 amounted to 25 people. The fourth experiment results show that the system is a system with a number of LBP characteristics = 128 is a system with the best combination of high system accuracy and fast learning time.
Implementasi Metode Analisis Gap Dan Profile Matching Untuk Kelayakan Calon Debitur Di Koperasi Simpan Pinjam (Ksp) Kopdit Solidaritas Santa Maria Assumpta yelly y nabuasa; Adriana Fanggidae; Derwin R Sina; Arfan Y Mauko
J-Icon : Jurnal Komputer dan Informatika Vol 7 No 2 (2019): Oktober 2019
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v7i2.1652

Abstract

In granting credit to debtors, it must go through an assessment of whether the debtor is appropriate or not feasible. KSP Koprit Solidaritas has set policy standards in granting credit to accept or reject the risk of bad credit, namely assessing prospective borrowers who meet the conditions of character rating, ability to pay off credit, capital owned, collateral owned and socioeconomic conditions. In this study, the design and manufacture of decision support system applications were carried out using profile matching methods to assess the eligibility of prospective debtors. Profile Matching is used to determine the priority with the highest ranking, which is used as a suggestion from the right system in determining the best alternative. The test results using 60 data obtained an accuracy of 81.667% with an error rate of 18.333% which indicates that the decision support system is functioning optimally following the Profile Matching method.
Sistem pakar mendiagnosa penyakit pada balita usia 0 – 60 bulan menggunakan metode Dempster-Shafer Jesica Cecilia Djami Manu; Sebastianus A S Mola; Adriana Fanggidae
J-Icon : Jurnal Komputer dan Informatika Vol 8 No 1 (2020): Maret 2020
Publisher : Universitas Nusa Cendana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35508/jicon.v8i1.2026

Abstract

Toddlers are susceptible to germs and viruses so that they are susceptible to various types of diseases because the immune system that has not been built properly. Most parents don't know about the symptoms and diseases suffered by toddlers, which is why sometimes the disease gets worse. The solution to the disease experienced by toddlers can be overcome is to take them to a hospital or health center. Binaus Health Center located in Mollo Tengah District, TTS Regency is one of the community health centers where most of the patients are toddlers. The obstacle of Binaus Community Health Center in operating is the specialist doctor of children or toddlers is not available, so an expert system was created as a media for consultation and monitoring of toddlers with the Dempster-Shafer Method whose end result was the diagnosis of diseases suffered by toddlers. Comparison of expert diagnosis results with expert systems obtained an average range of trust of 92.86% for appropriate testing experts above the threshold value and testing below the threshold value obtained miss-classification 7.14%. The expert system does not provide 100% results according to expert diagnosis not because of inference errors but the inappropriate value of expert density. Data testing based on case studies taken from medical record data at the Binaus Health Center, Kec. Mollo Tengah, Kab. TTS has a 100% accuracy rate.