Articles

Found 11 Documents
Search

PERBANDINGAN ANTARA METODE STATISTIKA DAN METODE NEURAL NETWORK PADA MODEL PERAMALAN INDEKS HARGA PERDAGANGAN BESAR Gunaryati, Aris; Suhendra, Adang
Jurnal Ilmiah Teknologi dan Rekayasa Vol 20, No 1 (2015)
Publisher : Universitas Gunadarma

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penelitian yang berhubungan dengan metode peramalan sudah sangat banyak dan akan terus berkembang seiring dengan bertambah kompleksnya permasalahan yang dihadapi di dunia nyata. Metode peramalan yang sering digunakan dalam penelitian adalah metode peramalan kuantitatif dengan data runtun waktu. Beberapa metode peramalan runtun waktu dapat menganalisis pola yang terdapat pada data runtun waktu dengan teknik statistik yang mudah dan sederhana dan memberikan hasil prediksi yang baik, seperti analisis trend, eksponensial smoothing, dekomposisi dan ARIMA Box-Jenkins. Di samping itu ada juga yang menggunakan teknik kecerdasan buatan seperti neural network (jaringan syaraf tiruan) untuk memprediksi. Pada penelitian ini akan dilakukan perbandingan model peramalan dari 13 variabel data indeks harga perdagangan besar periode Januari 2000 ? Agustus 2013 antara metode statistika dan metode neural network. Hasil penelitian menunjukkan bahwa dari 13 variabel indeks harga perdagangan besar yang dianalisis, terdapat 12 variabel memiliki bentuk model peramalan yang paling cocok yaitu ARIMA (92,3%). Berdasarkan hasil peramalan untuk 13 variabel indeks harga perdagangan besar untuk periode bulan Januari ? Agustus 2013, terdapat 11 variabel memiliki model peramalan yang paling akurat yaitu ARIMA (84,62%) Kata Kunci: ARIMA, dekomposisi, neural network, tren, peramalan
EXPERT SYSTEM FOR EARLY DETECTION OF BREAST CANCER WITH THE FORWARD CHAINING METHOD Caniago, Diana; Andryana, Septi; Gunaryati, Aris
Journal of Intelligent Decision Support System (IDSS) Vol 3 No 1, Maret (2020): Exper System, Decision Support System, Datamining
Publisher : Journal of Intelligent Decision Support System (IDSS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (947.672 KB)

Abstract

Breast cancer will be easier to overcome if it is known as early as possible to the importance of self-awareness to perform a routine inspection of BSE. The study presented aims to design a web-based application in the health field in the early detection of breast cancer. Penelian expert system method on this is to use forward chaining to represents the rule and reasoning into a coherent system based on physical symptoms entered. In this system also gets a percentage probability of 72.7%, so it can be quite good. In addition the system can produce two outputs in the form of possibility, the output is both benign and malignant.
HYBRID EXPONENTIAL SMOOTHING NEURAL NETWORK UNTUK PERAMALAN DATA PENGGUNA PITA LEBAR DI INDONESIA Gunaryati, Aris; Fauziah, Fauziah; Andryana, Septi
Jurnal SISKOM-KB (Sistem Komputer dan Kecerdasan Buatan) Vol 2 No 2 (2019): Volume II - Nomor 2 - Maret 2019
Publisher : Teknik Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The country of Indonesia has a huge opportunity to realize the potential for broadband users because the population growth rate is high and internet users are increasing each year. For this reason, acceleration is needed in providing telecommunications services and infrastructure that can increase internet access for users. To find out the estimated number of broadband users in the future, in this study proposed a hybrid exponential smoothing neural network forecasting model. By combining these methods, it is possible to take advantage of the forecasting technique of each while overcoming their drawbacks. This study showed that the RMSE value of hybrid model is 29.86% lower than that of the non hybrid. The MAPE value of hybrid model is 99.93% lower than that of the non hybrid model and the MAE value of hybrid model is 38.52% lower than that of the non hybrid model.
IMPLEMENTATION OF EXPERT SYSTEM FOR EARLY DETECTION OF BREAST CANCER WITH WEB-BASED METHODS CHAINING FORWARD Caniago, Diana; Andryana, Septi; Gunaryati, Aris
Jurnal Mantik Vol 3 No 4, Feb (2020): Manajemen, Teknologi Informatika dan Komunikasi (Mantik)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Breast cancer will be easier to overcome if it is known as early as possible to the importance of self-awareness to perform a routine inspection of BSE. The study presented aims to design a web-based application in the health field in the early detection of breast cancer. Penelian expert system method on this is to use forward chaining to represents the rule and reasoning into a coherent system based on physical symptoms entered. In this system also gets a percentage probability of 72.7%, so it can be quite good. In addition the system can produce two outputs in the form of possibility, the output is both benign and malignant.
ANALISIS PERBANDINGAN ALGORITMA JARINGAN SYARAF TIRUAN PADA PROSES IDENTIFIKASI GERAK BAHU Fauziah, Fauziah; Gunaryati, Aris; Andryana, Septi; Sholihati, Ira Diana
Jurnal SISKOM-KB (Sistem Komputer dan Kecerdasan Buatan) Vol 3 No 1 (2019): Volume III - Nomor 1 - September 2019
Publisher : Teknik Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (70.372 KB)

Abstract

The research related to the analysis of Range of Motion on human movements continues to evolve along with the development of image processing techniques related to human motion can be used to assess the motion done whether it is appropriate. The stage among others is to perform motion acquisition and then convert motion video into multiple image frames. The second phase performs the image processing process consisting of the development phase segmentation algorithm, consisting of background subtraction, grayscale, filtering, threshold, dilation, erosion to silhouette formation. Artificial Neural Network is a concept of learning used and implemented using computer programs and able to complete many calculation processes during the learning process and is system Information processors that have similar characteristics to the human neural network. Results of calculations made from neural network algorithms get an accuracy value of 92.67% using Back propagation of neural networks and 68.13% using; a radialbased neural network function to determine 3 motions are adduction, hyperextension, and extension. DOWNLOAD FILE : FULL TEXT
BIOMETRIK POLA SUARA DENGAN JARINGAN SARAF TIRUAN Agustina, Ina; Fauziah, Fauziah; Gunaryati, Aris
JURNAL TEKNIK INFORMATIKA Vol 9, No 2 (2016): Jurnal Teknik Informatika
Publisher : Prodi Teknik Informatika Universitas Islam Negeri Syarif Hidayatullah Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (428.163 KB) | DOI: 10.15408/jti.v9i2.5605

Abstract

ABSTRAK Metode  yang  digunakan  dalam  penelitian  ini  adalah  biometrik  pengenalan  identitas  berdasarkan karakteristik fisik. Pengenalan pola suara biometrik yang memiliki biaya rendah. Penelitian ini merancang pola penggunaan suara biometrik. sinyal input suara dalam bentuk suara yang direkam dengan durasi sekitar dua (2) detik pada sistem dan sinyal dekomposisi menggunakan transformasi wavelet diskrit. sistem verifikasi menggunakan ekstraksi karakteristik Koefisien Sub Band berdasarkan parameter cepstral (SBC) dengan menggunakanjumlah  koefisien 12. Kata  kunci: Biometrik, pola suara, transformasi wavelet, jaringan saraf
E-VOTING APPLICATION SECURITY USING WEB-BASED CRYPTOGRAPHY RSA TYPE Anjaswari, Istiarati Ayustina; Andryana, Septi; Gunaryati, Aris
Jurnal Mantik Vol 3 No 4, Feb (2020): Manajemen, Teknologi Informatika dan Komunikasi (Mantik)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In 2017 we held the election of the chairman of the Youth South Manggarai in a conventional way, Which were displayed many members who did not came.There are many causes that are still unused ballots and the results of the voting are calculated Relatively long Also have a risk in the results of voting. With the existence of such contradictions, then to Utilize technology has been developed that makes the Electronic Voting author system.However application, the public did not trust e-voting, there will be manipulation of the results approved by each candidate. For that reason, the system can guarantee results of voting using RSA security, public key and private key to approve when you do the voting. Built using the method with waterfall SDLC. The validation test results concluded that the system is valid. Algorithm testing is done by the user entering the first odd value of 15 and the second odd value of 13 getting the value of m 168 and public key 169 from the calculation of the FPB the which is Relatively prime to m. If the public key is inputted Correctly then the user can vote.The voting results show that candidate # 2 has 32 votes, the candidate 1 has 20 votes and the others is non-voters.
THE INTRODUCTION OF 3D APPLICATIONS ANIMAL IN INDONESIA USING AUGMENTED REALITY MARKER-BASED TRACKING METHOD Syihabudin, Bagas; Andryana, Septi; Gunaryati, Aris
Jurnal Mantik Vol 3 No 4, Feb (2020): Manajemen, Teknologi Informatika dan Komunikasi (Mantik)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

In this study, specifically about endangered animals can be done with a variety of media and for users aged 5 to 10 and must be accompanied by parents. This research Utilizes the benefits of marker-based tracking methods to familiarize endangered animals in 3D on sponsored smartphone devices. Augmented Reality Also has a FAST Angle Detection algorithm to support a marker object in it. In the results and testing of 300 angles produce 18.75% of objects Obtained, 81.25% of the objects are not supported, for an angle of 600 produces 27.08% of objects received overall, 54.17% of objects received mostly, 18.75% the object is not liked and the angle of 900 produces 58.4% of the objects taken as a whole, 33.3% of the most objects Obtained, 8.
PERBANDINGAN METODE PERAMALAN EKSPONENSIAL SMOOTHING DAN METODE JARINGAN SYARAF TIRUAN PROPAGASI BALIK UNTUK DATA PENGGUNA PITA LEBAR (BROADBAND) DI INDONESIA Gunaryati, Aris; Fauziah, Fauziah; Andryana, Septi
JUST IT : Jurnal Sistem Informasi, Teknologi Informasi dan Komputer Volume 8 No 2 Tahun 2018
Publisher : Universitas Muhammadiyah Jakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24853/justit.8.2.81-89

Abstract

Negara Indonesia memiliki peluang yang sangat besar untuk merealisasikan potensi pitalebar (broadband), mengingat Indonesia memiliki jumlah penduduk 253 juta orang dan pengguna internet 88,1 juta orang pada tahun 2014. Berdasarkan hal tersebut, diperlukan suatu penelitian tentang peramalan pengguna pitalebar di Indonesia di masa yang akan datang. Pada penelitian ini akan dibuat perbandingan model peramalan dengan metode jaringan syaraf tiruan propagasi balik dan double exponential method untuk mendapatkan hasil yang akurat. Hasilnya adalah antara metode brown dan metode holt mencapai nilai yang signifikan. Pada metode brown didapatkan nilai rata-rata 2458697.713 dan untuk nilai rata-rata menggunakan metode holt 2030153.82. Selain itu, nilai MAE dan MSE metode Brown lebih kecil dibandingkan metode Holt. Dengan demikian, metode Brown lebih cocok dan akurat sebagai model peramalan pada analisis Exponential Smoothing. Untuk nilai MAE dan MSE terkecil diperoleh pada model peramalan dengan analisis jaringan syaraf tiruan metode propagasi balik arsitektur 3, yaitu menggunakan fungsi aktivasi tangen hiperbolik. Oleh karena itu, model peramalan yang paling cocok dan paling akurat untuk memprediksi dan meramal data pengguna pita lebar di Indonesia pada tahun yang akan datang (2016-2020) adalah model peramalan dengan analisis jaringan syaraf tiruan metode propagasi balik arsitektur 3, yaitu menggunakan fungsi aktivasi tangen hiperbolik
Simulasi Vending Machine Dengan Mengimplementasikan Finite State Automata Saputra, Tri Ichsan; Fauziah, Fauziah; Gunaryati, Aris
JOINTECS (Journal of Information Technology and Computer Science) Vol 3, No 3 (2018)
Publisher : Universitas Widyagama Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (530.588 KB) | DOI: 10.31328/jointecs.v3i3.819

Abstract

Pada teori bahasa dan automata menjelaskan tentang mesin-mesin abstrak yang mana didalamnya terdapat finite state automata yang bisa diimplementasikan kedalam vending machine. Vending machine untuk kota seperti Jakarta sudah bukan perihal sulit untuk menemukannya. Cara kerja secara operasional sudah banyak diketahui prosedurnya, namun perihal bagaimana cara kerjanya didalam mesin tersebut, masih belum banyak yang mengetahuinya, melalui simulasi ini diharapkan pengguna juga lebih paham akan hal itu