Claim Missing Document
Check
Articles

Klasifikasi Emosi Berdasarkan Ciri Wajah Menggunakan Convolutional Neural Network Achmad Yusuf; Randy Cahya Wihandika; Candra Dewi
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 11 (2019): November 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (660.1 KB)

Abstract

Apart from being a person's identity, the face is also a supporting tool in direct socializing. A person can convey emotions experienced using expressions raised by their face. Emotion is a feeling to encourage an individual or a response to a stimulus. In consumer research, consumer testing is a method used to predict product acceptance by consumers in a market. Even though it has gone through extensive consumer testing stages before entering the market, the failure rate of new food products is still high. This shows that traditional consumer testing methods are not able to predict market performance and product acceptance by consumers in the long run. To be able to know consumer behavior more deeply, the use of emotional measurement is widely used in consumer testing because emotions affect consumer behavior. In this case, the classification of emotions based on facial characteristics is considered suitable to help improve the quality of consumer testing. The method used in this study is the Convolutional Neural Network (CNN). The data used are data obtained from the Extended Cohn-Kanade Dataset (CK +) taken from 210 subjects with a total of 327 images used. Testing the study using K-fold Cross Validation with a k value of 4. The test results show a certain learning rate value can train architecture better than other learning rate values. The best accuracy results in this study amounted to 86.4% and an average accuracy of 80.7%.
Segmentasi Pembuluh Darah pada Citra Retina Menggunakan Ciri Multi-Scale Line Strength Muhammad Faiz Abdul Hamif; Randy Cahya Wihandika; Agus Wahyu Widodo
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 1 (2020): Januari 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (643.89 KB)

Abstract

One of the disease that should be worried about and can cause the sufferer to have a risk of complications in several parts of the body is diabetic. If this disease attacked the blood vessel in the eye, this disease then called diabetic retinopathy. Diabetic retinopathy examination is carried out every six months including retinal imaging and analysis. The evaluation of retinal images become a serious burden for opthalmologists because the patients with diabetic retinopathy are increasing but the healthcare workers are limited. One way to alleviate the ophthalmologist's workload is to use computer assistance with Multi-Scale Line Strength algorithm for features extraction and Support Vector Machine (SVM) classification algorithm for segmenting the retinal images in a supervised way and the Optic Disc Exclusion algorithm for eliminating the optic disc area in the segmentation result images. The performance of these algorithm is measured in the DRIVE dataset. The accuracy, sensitivity, and specificity obtained from the Multi-Scale Line Strength algorithm combined with SVM are 0,94021, 0,61084, and 0,99693. If those algorithm is combined with the Optic Disc Exclusion algorithm, the performance results are 0,94014, 0,60277, and 0,99694. Both performance results are obtained at window size 13.
Prediksi Laju Pertumbuhan Penduduk Menggunakan Metode Support Vector Regression (Studi Kasus: Kota Malang) Arynda Kusuma Dewi; Muhammad Tanzil Furqon; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 1 (2020): Januari 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (471.505 KB)

Abstract

Population growth rate is a changes population every year in a region. The high population growth rate in Indonesia is something important because it has impact on the economic, social, politic and national defense. Therefore, related parties such as Dinas sosial and BKKBN analyze the factors which related with population growth rate, so it can make some policies to realize balance of population growth. Beside that, population growth prediction is also used by Dispendukcapil to make other budget plans and other needs. In this study, population growth rate is predicted using Support Vector Regression method by comparing the performance of linear kernels with Gaussian kernel RBF used population growth dataset time series in March 2013 until December 2018. The steps to predict population growth rate begin with data normalization, SVR training to get the update lagrange multiplier value and SVR testing to get prediction results and error rates using MAPE. The test results obtained by the MAPE value using a linear kernel 0.0985% and 0.38192% using the Gaussian RBF kernel.
Analisis Sentimen Mass Rapid Transit Jakarta dengan Naive Bayes Classifier dan Deteksi Target Opini Berbasis Aturan pada Twitter Dhanika Jeihan Aguinta; Putra Pandu Adikara; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 13 (2020): Publikasi Khusus Tahun 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Untuk dipublikasikan di 5th International Conference on Sustainable Information Engineering and Technology (SIET)
Analisis Sentimen Mengenai Produk Toyota Avanza Menggunakan Metode Learning Vector Quantization Versi 3 (LVQ 3) dengan Seleksi Fitur Chi Square, Lexicon-Based Features serta Normalisasi Min-Max Jonathan Reynaldo; Putra Pandu Adikara; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 3 (2020): Maret 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (472.804 KB)

Abstract

Car is a means of transportation used by peoples with some excellence and comfort values for better driving experience. Toyota as the manufacturer of Toyota Avanza needs people's opinions to upgrade their products. Opinions from social media need to be classified as positive, neutral or negative opinions so sentiment analysis is needed. For analyzing a sentiment, Learning Vector Quantization 3 (LVQ 3) is used in this research. Chi Square feature selection, lexicon-based features and min-max normalization are used in this research too. Evaluation using confusion matrix with 240 training data and 60 testing data results the accuracy of 38,33% using features from Chi Square feature selection, 33,33% using lexicon-based features, and 36,67% using both of Chi Square feature selection and lexicon-based features.
Peringkasan Artikel Berbahasa Indonesia Menggunakan TextRank dengan Pembobotan BM25 Yurdha Fadhila Hernawan; Putra Pandu Adikara; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 13 (2020): Publikasi Khusus Tahun 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Untuk dipublikasikan di Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK)
Prediksi Luas Serangan Hama pada Tanaman Padi Menggunakan Metode Extreme Learning Machine (ELM) dan Particle Swarm Optimization (PSO) Cornelius Bagus Purnama Putra; Randy Cahya Wihandika; Sigit Adinugroho
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 3 (2020): Maret 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (911.071 KB)

Abstract

Indonesia is one of the countries with the largest population in the world. The majority of the population consuming rice as the staple food, rice becomes an important commodity. Recent global warming has resulted in extreme climate change, so that it can affect crop productivity and the intensity of OPT (Plant Pests) attack on rice plants. In meeting the increasing need for rice, it is necessary to prevent pest attack so that widespread prediction of pest attack area is needed in order to know earlier about upcoming pest attack. This study used hybrid algorithm Extreme Learning Machine and Particle Swarm Optimization with used data on pest attacks and climatology of Sidoarjo Regency from January 2009 to December 2018. Based on the research, the optimal parameters obtained are the ratio of training data 80% and testing data 20%, activation function of TanH, total population of 40, combination acceleration coefficient of 1 & 2, inertia weight limit of 0,4 & 0,9, hidden neuron of 5, and a maximum iteration of 100. Based on these parameters, the average value of the Mean Absolute Percentage Error (MAPE) is 25.143% which is included in the MAPE category of quite good, which is within the range of 20% -50%.
Analisis Sentimen Layanan Astra Honda Motor Menggunakan Metode Naive Bayes dan Identifikasi Aspek pada Layanan Menggunakan DBSCAN Naufal Akbar Eginda; Putra Pandu Adikara; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 3 (2020): Maret 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (364.236 KB)

Abstract

In today's era, transportation is one of the most vital needs. according to BPS Indonesia (badan pusat statistik) in 2017, motorbikes are the most owned form of transportation in Indonesia. one of the most commonly used motorbike brand is honda. astra honda motor provides a range of services available to its customers. of all their services, honda's customers would surely have a number of opinions or feedbacks regarding their services, whether they are positive or negative. in order to classify and identify aspects of the public opinion, naive bayes with lexicon-based features are implemented to classify them and dbscan is implemented to cluster them by taking the top 3 terms generated from each document. the dataset used in this paper consists of 100 datas divided into an equal set of each class, and 25 datas for testing in which 13 are classified positive and 12 are negative. the result of the classification process applying naive bayes with lexicon-based features is a precision value of 53%, recall of 63%, f-measure of 58% and 60% of accuracy. While, the result of the aspect identification came down to 41,4% for precision, 80,5% for recall, 54,6% for f-measure and a mere 37,6% for its accuracy level. as for the cluster evaluation with silhouette coefficient, the best parameter values using dbscan is an epsilon of 0.1 along with minpts of 1.
Pengembangan Aplikasi Perangkat Bergerak: Pencari Konselor Psikologi Terdekat berbasis Lokasi Tifo Audi Alif Putra; Agi Putra Kharisma; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 3 (2020): Maret 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (839.169 KB)

Abstract

Mental health is now a problem that cannot be underestimated. Even mental health is now a problem faced by many countries in the world, including Indonesia. Depression is a major cause of mental health problems. Depression affects 4.4% of the world's population and 5% of the population in Indonesia. In addition to the high number of people with depression, another problem in Indonesia is the lack of available mental health professionals such as counselors and psychiatrists. Therefore, the use of mobile-based systems can be done to connect people who have problems with the nearest counselor to get first aid in the problem. The implementation of mobile application development in this study was developed on the iOS platform using the Swift programming language. The research method used is an iterative waterfall model consisting of needs analysis, system design, system implementation, and testing. Testing in this study consisted of blackbox testing and usability testing. In the blackbox testing the researchers designed a testcase scenario to assess the correct system behavior while to get the comfortability level of the application developed using the SUPR-Qm instrument. The test results on Blacbox Testing stated 100% valid of all designed testcases. For the usability test results obtained 69.25% for patient applications and 74.75% for counselor applications, then if converted to usability rating scale produces an ok category for patient applications and good for counselor applications.
Sistem Prediksi Pertumbuhan Jumlah Penduduk Kota Malang menggunakan Metode K-Nearest Neighbor Regression Diajeng Sekar Seruni; Muhammad Tanzil Furqon; Randy Cahya Wihandika
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 4 (2020): April 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (376.489 KB)

Abstract

Population growth rate in Indonesia keeps growing each and every year. The rapid growth of population may give impacts on many aspects of the country, such as economic development, quality of life and health issues of the residents, and even educational problems. To anticipate the negative effects of population growth, projection of future population is needed as to help the government to develop city development plans. K-Nearest Neighbor (KNN) is one of many methods that could be used to predict future values, be it for classification or regression. KNN Regression is a KNN algorithm used for regression or forecasting problems. In this study, the KNN Regression method is implemented to forecast future population of Malang city, using a time series of monthly population growth consisting of 73 datas in total. The forecasting method starts with preprocessing the time series, calculate the distance between each training and testing data, and estimate the predicted value based on k nearest neighbors. From the testings done in this study, the lowest Mean Absolute Percentage Error (MAPE) value obtained is 0,02526%.
Co-Authors Achmad Arwan Achmad Ridok Achmad Yusuf Adam Hendra Brata Adam Sulthoni Akbar Adinugroho, Sigit Aditya Putra Pratama Agi Putra Kharisma Agung Nurjaya Megantara Agus Wahyu Widodo Akhmad Sa'rony Amar Ikhbat Nurulrachman Anang Hanafi Angky Christiawan Rongre Ani Enggarwati Ardisa Tamara Putri Ardiza Dwi Septian Arif Pratama Arynda Kusuma Dewi Barlian Henryranu Prasetio Bayu Kusuma Pradana Bayu Laksana Yudha Bayu Rahayudi Budi Darma Setiawan Budi Dharma Setiawan Candra Dewi Chandra Tio Pasaribu Cindy Cunday Cicimby Cornelius Bagus Purnama Putra Cusen Mosabeth Dani Devito Daris Hadyan Tisantri Denny Sagita Rusdianto Devinta Setyaningtyas Atmaja Dhan Adhillah Mardhika Dhanika Jeihan Aguinta Diajeng Sekar Seruni Dian Eka Ratnawati Dimi Karillah Putra Dito Rizki Pramudeka Dizka Maryam Febri Shanti Dwi Rahayu Eka Putri Nirwandani Emma Wahyu Sulistianingrum Ersya Nadia Candra Fachril Rachma Zulfidar Fachrur Rozy Faizatul Amalia Fajri Eka Saputra Fanny Aulia Dewi Fera Fanesya Fida Dwi Febriani Fikri Hilman Firda Oktaviani Putri Fitra Abdurrachman Bachtiar Frisma Yessy Nabella Gilang Widianto Aldiansyah Glenn Jonathan Satria Gregorius Ivan Sebastian Hafiz Ari Putra Hamim Fathul Aziz Heykhal Hafiddhan Rachman I Gusti Ngurah Ersania Susena Imam Cholissodin Indriati Indriati Irnayanti Dwi Kusuma Jonathan Reynaldo Kevin Haidar Kevin Nastatur Chatriavandi Koko Pradityo Lailil Muflikhah Lalu Muhammad Ivan Natania Latifa Nabila Harfiya M. Rikzal Humam Al Kholili Moh. Dafa Wardana Mohammad Rizky Hidayatullah Muchlas Mughniy Muh. Arif Rahman Muhamad Ilham Dian Putra Muhamad Wahyu Budi Santoso Muhammad Alif Fahrizal Muhammad Amin Nurdin Muhammad Faiz Abdul Hamif Muhammad Ihsan Diputra Muhammad Shidqi Fadlilah Muhammad Tanzil Furqon Muhammad Tegar Kanugroho Naufal Akbar Eginda Nindy Deka Nivani Nova Amynarto Novanto Yudistira Nur Wahyu Ningtyas Nurul Hidayat Nurul Muslimah Pindo Bagus Adiatmaja Pupung Adi Prasetyo Puspita Sari Putra Pandu Adikara Putu Gede Pakusadewa Qurrata Ayuni Raden Rafika Anugrahning Putri Ratih Kartika Dewi Rayindita Siwie Mazayantri Rekyan Regasari Mardi Putri, Rekyan Regasari Mardi Rizal Setya Perdana Rizky Nur Ariyanti Ruri Armandhani Sarah Najla Adha Satria Dwi Nugraha Satyawan Agung Nugroho Sema Yuni Fraticasari Sevtyan Eko Pambudi Sigit Adinugroho Siti Robbana Sukma Fardhia Anggraini Supraptoa Supraptoa Sutrisno Sutrisno Tahajuda Mandariansah Threecia Agil Regitasari Tifo Audi Alif Putra Tri Kurniawan Putra Utaminingrum, Fitri Valen Novandi Kanasya Vandi Cahya Rachmandika Winda Cahyaningrum Yosendra Evriyantino Yosua Christopher Sitanggang Yudha Prasetya Anza Yuita Arum Sari Yurdha Fadhila Hernawan