Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Not a member yet
1071 research outputs found
Sort by
Evaluasi Topik Tersembunyi Berdasarkan Aspect Extraction menggunakan Pengembangan Latent Dirichlet Allocation
No full textRecently, Sentiment Analysis is used for expression detection of products or services. Sentiment Analysis is one category type with a level of aspect focused on extracting product aspects. One of the common methods used for aspect extraction is Latent Dirichlet Allocation (LDA) using random topic identification, but this method has not been able to find an acceptable topic with some aspects having been found. Undeterminable topics are referred to as the hidden topics. This study purpose is to evaluate and compare the suitability of identifying hidden topics between human and computer evaluation. The study is also focused on aspect extraction using a variety of LDA innovations. The data used in this study used case studies on e-Commerce. Data were processed using feature selection and grouped using LDA development. Then the data results are processed using Latent Topic Identification based on subjective and objective evaluations. The identification of hidden topic results was evaluated using several semantic and lexicon tests. The evaluation results indicate the comparison of two hidden topic identification assessment values is quite relevant with the average difference in value reaching 6%. As a result, computer calculations assist humans in determining topics if each topic has a low coherence value.
atau jasa. Sentiment Analysis mempunyai beberapa klasifikasi dan tingkatan, salah satu jenis Sentiment Analysis tersebut adalah tingkat aspek yang berfokus pada ekstraksi aspek suatu produk. Penelitian ini menggunakan metode yang populer saat ini, yaitu, Latent Dirichlet Allocation (LDA) yang menggunakan identifikasi topik secara acak. Akan tetapi metode ini belum dapat menentukan topik yang sesuai dengan aspek-aspek yang diperoleh. Topik yang belum bisa ditentukan tersebut juga bisa disebut dengan topik tersembunyi. Fokus penelitian ini adalah evaluasi dan perbandingan kesesuaian dari identifikasi topik yang tersembunyi antara penilaian manusia dan perhitungan komputer. Penelitian ini didasarkan pada ekstraksi aspek menggunakan beberapa pengembangan LDA. Data yang digunakan pada penelitian ini menggunakan studi kasus pada e-Commerce dengan jumlah ulasan sebanyak 28.222. Pengolahan data menggunakan seleksi fitur dengan nilai α sebesar 0,99 dan dikelompokkan menggunakan pengembangan LDA yang ditentukan dengan nilai hyperparameter α sebesar 0,9 dan β sebesar 0,05. Kemudian hasil data tersebut diolah menggunakan Latent Topic Identification berdasarkan penilaian subjektif dan objektif. Hasil identifikasi topik tersembunyi dievaluasi menggunakan beberapa pengujian semantik dan leksikon. Hasil evaluasi mengindikasikan bahwa nilai perbandingan kedua penilaian identifikasi topik tersembunyi cukup relevan dengan rata-rata selisih nilai mencapai 6%. Sehingga perhitungan komputer dapat membantu manusia untuk menentukan topik jika setiap aspek memiliki nilai koherensi yang rendah
Klasifikasi Citra Burung Lovebird Menggunakan Decision Tree dengan Empat Jenis Evaluasi
Get PDFLovebird is a pet that many people in Indonesia have known. The diversity of species, coat color, and body shape gives it its charm. As well in this lovebird bird has its uniqueness of various rare colors. However, many ordinary people have difficulty distinguishing the types of lovebirds. This research is needed to improve previous study performance in classifying lovebird images using the Decision Tree J48 algorithm with 4 types of evaluation. In this case, also to reduce the stage of feature extraction to speed up the computational process. Based on available comparisons, the results obtained at the same split ratio with a comparison of 60:40 in Decision Tree J48 have the precision of 1,000, recall of 1,000, f-measure of 1,000, and accuracy value of 100%. Then the Artificial Neural Network with a split ratio of 60:40 has a precision of 0.854, recall of 0.843, f-measurement of 0.841, and an accuracy value of 84.25%. These results prove that by testing the first-level extraction on color features, Decision Tree J48 is superior in classifying images of lovebird species, and Decision Tree J48 can improve performance and produce the best accuracy.
Burung lovebird merupakan hewan peliharaan yang telah dikenal oleh banyak masyarakat di Indonesia. Keanekaragam jenis, warna bulu, serta bentuk tubuhnya memberikan daya tarik tersendiri bagi pecinta. Serta dalam burung lovebird ini juga memiliki keunikan sendiri yang beragam warna-warna langka. Namun, banyak diantara masyarakat awam kesulitan dalam membedakan jenis burung lovebird. Dari segi jenis, masyarakat cenderung melihat sesuai dengan yang ada di pada umumnya. Semakin jenis burung berbeda dan jarang diketahui masyarakat, maka akan memiliki nilai jual yang tinggi pada burung lovebird. Penelitian ini diperlukan untuk memperbaiki kinerja dari penelitian sebelumnya dalam proses klasifikasi citra burung lovebird menggunakan algoritma Decision Tree J48 dengan 4 jenis evaluasi. Serta dalam hal ini juga untuk mengurangi tahapan fitur ekstraksi agar mempercepat pada proses komputasi. Berdasarkan perbandingan yang sudah diketahui, Hasil yang didapatkan pada split ratio yang sama dengan perbandingan 60:40 pada Decision Tree J48 memiliki hasil precision 1.000, recall 1.000, f-measure 1.000 dan nilai accuracy mencapai 100%. Kemudian pada Artificial Neural Network dengan split ratio 60:40 memiliki hasil precision 0.854, recall 0.843, f-measure 0.841 dan nilai accuracy mencapai 84.25%. Hasil ini membuktikan dengan pengujian ekstraksi tingkat pertama pada fitur warna Decision Tree J48 lebih unggul dalam mengklasifikasi citra jenis burung lovebird, serta Decision Tree J48 mampu memperbaiki kinerja dan menghasilkan accuracy yang terbaik.
 
Implementasi Virtual Reality Berbasis Foto 360o Untuk Memvisualisasikan Fasilitas Perguruan Tinggi Surabaya
Get PDFThe college selection process is an important phase because this process affects the future achievement targets of prospective students. One of the factors that is considered in determining a university is the supporting facilities provided during the lecture process. prospective students will continue to search and at the same time consider universities despite the pandemic COVID-19. By implementing 3600 photo-based virtual reality (VR), prospective students or external parties can get information about university facilities anytime and anywhere because it can be accessed online. This study uses the Multimedia Development Life Cycle (MDLC) method in application development, then uses a quantitative approach to test the feasibility of the application. The results showed that 3600 photo-based virtual reality (VR) is an alternative media in conveying information related to the facilities and logistics owned by universities, the variables of smoothness and convenience of operating 3600 photo-based videos have a high enough influence, but on user motivation to use VR is worth less. This happens because users are not used to using this technology.
masa depan calon mahasiswa. Salah satu faktor yang dipertimbangkan dalam menentukan perguruan tinggi adalah fasilitas penunjang yang diberikan selama proses perkuliahan. Calon mahasiswa akan tetap melakukan pencarian sekaligus mempertimbangan perguruan tinggi meskipun terjadi pandemi COVID-19. Dengan mengimplementasi virtual reality (VR) berbasis foto 3600, calon mahasiswa atau pihak eksternal dapat memperoleh informasi mengenai fasilitas perguruan tinggi kapanpun dan dimanapun karena dapat diakses online. Penelitian ini menggunakan metode Multimedia Development Life Cycle (MDLC) dalam pengembangan aplikasi, kemudian menggunakan pendekatan kuantitatif untuk menguji kelayakan aplikasi. Hasil penelitian menunjukkan bahwa virtual reality (VR) berbasis foto 3600 merupakan media alternatif dalam penyampaian informasi terkait fasilitas dan logistik yang dimiliki perguruan tinggi, variabel kelancaran dan kenyamanan pengoperasian video berbasis foto 3600 mempunyai pengaruh yang cukup tinggi, namun pada motivasi pengguna dalam menggunakan VR bernilai kurang. Hal ini terjadi karena pengguna belum terbiasa menggunakan teknologi tersebut
Peningkatan Hasil Klasifikasi pada Algoritma Random Forest untuk Deteksi Pasien Penderita Diabetes Menggunakan Metode Normalisasi
Get PDFDiabetes is a disease caused by high blood sugar in the body or beyond normal limits. Diabetics in Indonesia have experienced a significant increase, Basic Health Research states that diabetics in Indonesia were 6.9% to 8.5% increased from 2013 to 2018 with an estimated number of sufferers more than 16 million people. Therefore, it is necessary to have a technology that can detect diabetes with good performance, accurate level of analysis, so that diabetes can be treated early to reduce the number of sufferers, disabilities, and deaths. The different scale values for each attribute in Gula Karya Medika’s data can complicate the classification process, for this reason the researcher uses two data normalization methods, namely min-max normalization, z-score normalization, and a method without data normalization with Random Forest (RF) as a classification method. Random Forest (RF) as a classification method has been tested in several previous studies. Moreover, this method is able to produce good performance with high accuracy. Based on the research results, the best accuracy is model 1 (Min-max normalization-RF) of 95.45%, followed by model 2 (Z-score normalization-RF) of 95%, and model 3 (without data normalization-RF) of 92%. From these results, it can be concluded that model 1 (Min-max normalization-RF) is better than the other two data normalization models and is able to increase the performance of classification Random Forest by 95.45%.
Diabetes merupakan salah satu penyakit yang disebabkan karena gula darah di dalam tubuh yang tinggi atau melampaui batas normal. Penderita diabetes di Indonesia mengalami peningkatan yang cukup signifikan, Riset Kesehatan Dasar menyebutkan penderita diabetes di Indonesia yang semula dari tahun 2013 sebesar 6,9% menjadi 8,5% di tahun 2018 dengan perkiraan jumlah penderita lebih dari 16 juta orang. Oleh karena itu, sangat diperlukan suatu teknologi yang dapat mendeteksi penyakit diabetes dengan kinerja yang baik, tingkat analisis akurat, sehingga penyakit diabetes dapat ditangani lebih awal untuk mengurangi jumlah penderita, kecacatan, dan kematian. Nilai skala yang berbeda tiap atribut pada data Gula Karya Medika dapat mempersulit proses klasifikasi, untuk itu peneliti menggunakan dua metode normalisasi data yaitu Min-max normalization, Z-score normalization, dan satu tanpa metode normalisasi data dengan Random Forest (RF) sebagai metode klasifikasi. Random Forest (RF) sebagai metode klasifikasi telah teruji di beberapa penelitian sebelumnya, metode ini mampu menghasilkan kinerja yang baik dengan akurasi yang tinggi. Berdasarkan hasil penelitian, akurasi terbaik dihasilkan model 1 (Min-max normalization-RF) sebesar 95.45%, model 2 (Z-score normalization-RF) sebesar 95%, dan model 3 (Tanpa normalisasi data-RF) sebesar 92%. Dari hasil tersebut disimpulkan bahwa model 1 (Min-max normalization-RF) lebih baik dibandingkan dua model normalisasi data lainya dan mampu meningkatkan performansi klasifikasi Random Forest sebesar 95.45%
Pemantauan Perhatian Publik terhadap Pandemi COVID-19 melalui Klasifikasi Teks dengan Deep Learning
Get PDFMonitoring public concern in the surrounding environment to certain events is done to address changes in public behavior individually and socially. The results of monitoring public attention can be used as a benchmark for related parties in making the right policies and strategies to deal with changes in public behavior as a result of the COVID-19 pandemic. Monitoring public attention can be done using Twitter social media data because the users of the media are quite high, so that they can represent the aspirations of the general public. However, Twitter data contains varied topics, so a classification process is required to obtain data related to COVID-19. Classification is done by using word embedding variations (Word2Vec and fastText) and deep learning variations (CNN, RNN, and LSTM) to get the classification results with the best accuracy. The percentage of COVID-19 data based on the best accuracy is calculated to determine how high the public's attention is to the COVID-19 pandemic. Experiments were carried out with three scenarios, which were differentiated by the number of data trains. The classification results with the best accuracy are obtained by the combination of fasText and LSTM which shows the highest accuracy of 97.86% and the lowest of 93.63%. The results of monitoring public attention to the time vulnerability between June and October show that the highest public attention to COVID-19 is in June.Memantau perhatian publik di lingkungan sekitar terhadap suatu kejadian tertentu dilakukan untuk mengatasi perubahan perilaku publik secara individual maupun sosial. Hasil pemantauan perhatian publik dapat dijadikan tolak ukur oleh pihak-pihak terkait dalam membuat suatu kebijakan maupun strategi yang tepat untuk menghadapi perubahan perilaku publik sebagai efek pandemi COVID-19. Pemantauan perhatian publik dapat dilakukan menggunakan data media sosial Twitter karena pengguna media tersebut cukup tinggi, sehingga dapat mewakili aspirasi publik secara umum. Namun, data Twitter mengandung topik yang bervariasi sehingga diperlukan proses klasifikasi untuk mendapatkan data terkait COVID-19. Klasifikasi dilakukan dengan variasi word embedding (Word2Vec dan fastText) dan variasi deep learning (CNN, RNN, and LSTM) untuk mendapatkan hasil klasifikasi dengan akurasi terbaik. Data COVID-19 hasil klasifikasi berdasarkan akurasi terbaik dihitung prosentasenya untuk mengetahui seberapa tinggi perhatian publik terhadap pandemi COVID-19. Percobaan dilakukan dengan tiga skenario yang dibedakan oleh jumlah data train. Hasil klasifikasi dengan akurasi terbaik didapatkan oleh kombinasi fasText dan LSTM yang menunjukkan akurasi tertinggi sebesar 97.86% dan terendah sebesar 93.63%. Hasil pemantauan perhatian publik pada rentan waktu antara Bulan Juni sampai Bulan Oktober menunjukkan bahwa perhatian publik terhadap COVID-19 tertinggi adalah pada Bulan Juni
Optimasi SVM Berbasis PSO pada Analisis Sentimen Wacana Pindah Ibu Kota Indonesia
Get PDFPresident Joko Widodo decided to move the capital city of the country outside Java. The relocation of the capital city is contained in the 2020-2024 National Medium-Term Development Plan. Community response to this has been mixed through national television and social media, especially Twitter. The tendency of Twitter users to respond to the government discourse can be seen with sentiment analysis. Sentiment analysis is one of the areas of Natural Language Processing (NLP) that builds systems for recognizing and extracting opinions. In this study, the Feature Selection PSO algorithm in the classification of the SVM model is proposed to improve the resulting accuracy in the sentiment analysis of moving capital cities. Experiments on the data of 1,319 tweets (457 positive sentiments and 862 negative sentiments) indicate an increase in accuracy by 2.09% from 79.06% to 81.15%, with the classification category is “Good Classification”.Presiden Joko Widodo memutuskan wacana pindah ibu kota negara ke luar Pulau Jawa. Pemindahan ibu kota ini tertuang dalam Rencana Pembangunan Jangka Menengah Nasional 2020-2024. Respon masyarakat terkait hal tersebut sangat beragam baik melaui televisi nasional maupun media sosial khususnya twitter. Kecenderungan respon pengguna twitter dalam menyikapi wacana pemerintah tersebut dapat diketahui dengan analisis sentimen. Analisis sentimen adalah salah satu bidang dari Natural Languange Processing (NLP) yang membangun sistem untuk mengenali dan mengekstraksi opini dalam bentuk teks. Dalam penelitian ini, diusulkan algoritma Feature Selection PSO pada klasifikasi model SVM untuk meningkatkan akurasi yang dihasilkan pada analisis sentimen wacana pindah ibu kota. Pembuktian yang dilakukan melalui eksperimen dengan data 1.319 tweets (457 sentimen positif dan 862 sentimen negatif) menunjukan peningkatan akurasi sebesar 2,09% dari akurasi sebelumnya 79,06% menjadi 81,15% dengan kategori “Good Clasification”. 
Evaluasi Parameter RAW Berdasarkan Multirate Pada IEEE 802.11ah: Simulasi Kinerja Optimum Jaringan IoT
Get PDFIEEE 802.11ah WLAN is a technology standard for IoT networks because it can provide a higher transmission range and data rate than WPAN and LPWAN. To manage channel access up to 8191 at the MAC layer IEEE 802.11ah a Restricted Access Window scheme was introduced. Generally, evaluation and optimization of RAW parameters are only based on constant data rates without taking into account the mutirate support for PHY AP and STA IoT IEEE 802.11ah. This study uses an open source-based NS-3 network simulator. Simulation analysis is run by calculating the value of throughput, delay, packet loss, and energy consumption of each node. Based on testing the effect of the number of slots on throughput, it shows that the resulting throughput values fluctuate with stable dominance, depending on the number of slots used. The effect of the number of slots on packet loss shows that the packet loss value is low for each slot because more packets can be accommodated in the RAW slot queue. The effect of the number of slots on energy consumption decreases at some data rates and some lower energy consumption values, thereby saving energy consumption.
WLAN IEEE 802.11ah merupakan standar teknologi untuk jaringan IoT karena mampu menyediakan jangkauan transmisi dan data rate lebih tinggi dari WPAN dan LPWAN. Untuk mengatur akses kanal hingga 8191 pada lapisan MAC IEEE 802.11ah diperkenalkan skema Restricted Access Window (RAW). Literatur sebelumnya evaluasi dan optimasi parameter RAW hanya berdasarkan data rate konstan tanpa memperhitungkan dukungan mutirate pada PHY AP dan STA IoT IEEE 802.11ah. Penelitian ini bertujuan mengevaluasi dan menganalisis parameter RAW optimal, yaitu parameter Slot Duration Count, Number of Slots, dan RAW Group berdasarkan multirate untuk dapat menghasilkan kinerja WLAN IEEE 802.11ah yang optimal. Penelitian ini menggunakan simulator jaringan NS-3 berbasis open source. Prosedur pengujian dengan mengimplementasikan tahapan RAW berdasarkan multirate pada IEEE 802.11ah untuk optimasi kinerja jaringan IoT. Analisis simulasi dijalankan dengan menghitung nilai throughput, delay, packet loss dan konsumsi energi masing-masing node. Berdasarkan pengujian pengaruh jumlah slot terhadap throughput menunjukkan nilai throughput yang dihasilkan berubah-ubah dengan dominasi stabil, tergantung jumlah slot yang digunakan. Pengaruh jumlah slot terhadap delay cenderung stabil dan beberapa nilai delay juga rendah, rata-rata delay dipengaruhi oleh mekanisme RAW dimana saat paket dikirim tetapi belum mendapat kesempatan pada bagian RAW slot. Pengaruh jumlah slot terhadap packet loss menunjukkan nilai packet loss rendah tiap slot-nya, dikarenakan lebih banyak paket yang dapat ditampung pada antrian RAW slot. Pengaruh jumlah slot terhadap konsumsi energi menurun pada beberapa data rate dan beberapa nilai konsumsi energi lebih rendah sehingga menghemat konsumsi energi. Hal diakibatkan dengan bertambahnya jumlah slot kecepatan mobilitas meningkat, namun nilai rata-rata konsumsi energi hanya naik sedikit
Pengambilan Keputusan Sistem Penjaminan Mutu Perguruan Tinggi menggunakan MOORA, SAW, WP, dan WSM
Get PDFHigher Education Quality Assurance (QA) is regulated in Quality Standards and the number of criteria as well as its relationship with the implementation of the Quality Assurance System (QAS), namely the Internal and External Quality Assurance System (IHSG). . The research is focused on analyzing 4 decision-making methods or Decision Support Systems (DSS) for QAS in MAC. The purpose of this study is to classify standard data and MAC criteria in business processes into a database that is integrated with the QAS decision-making method. The analysis was carried out on 4 multi-criteria decision-making methods that will be used in the QAS-MAC decision-making process, namely: Moora, SAW, WP, and WSM. These methods were tested on a quality standard database and then assessed by comparison, namely relevance, features, accuracy, precision, reliability, effectiveness, efficiency, strengths, and weaknesses. Decision Making Methods as a determinant of Business Process priorities become information for PTMA Leaders in predicting strategic activities. The value of the method analysis shows that 4 decision-making methods are Moora (75%), SAW (75%), WP (94%), and WSM (94%).
Standar Mutu dan banyaknya kriteria dan keterkaitan dengan kegiatan pelaksanaan Sistem Penjaminan Mutu (SPM) yaitu Sistem Penjaminan Mutu Internal (SPMI) dan Sistem Penjaminan Mutu Eksternal (SPME) menjadi persoalan yang diangkat pada penelitian ini terkait Penjaminan mutu Perguruan Tinggi. Standar mutu merupakan standar yang ditetapkan oleh perguruan tinggi berdasarkan mutu dari pemerintah. Penelitian difokuskan pada analisis 4 algoritma pengambilan keputusan atau Sistem Pendukung Keputusan (SPK) untuk SPM di PTMA. Penelitian ini diadakan untuk menyelaraskan antara SPMI dan SPME dengan tujuan penelitian ini adalah untuk mengelompokkan data standar maupun kriteria PTMA dalam proses bisnis menjadi basis data yang diintegrasikan dengan metode pengambilan keputusan pada SPM. Analisis dilakukan terhadap 4 algoritma pengambilan keputusan multi kriteria untuk digunakan dalam proses pengambilan keputusan SPM PTMA, antara lain: Multi-Objective Optimization on the basis of Rasio Analysis (MOORA); Simple Additive Weighting (SAW); Weighted Product (WP); dan Weighted Sum Model (WSM). Metode yang dilakukan adalah dengan menguji data yang ada berkaitan dengan basis data standar mutu untuk kemudian dinilai dengan variabel pembanding antar algoritma yaitu relevansi, ciri, akurasi, presisi, reliabilitas, efektivitas, efisiensi, kelebihan, dan kekurangan. Hasil dan analisis penelitian adalah metode pengambilan keputusan sebagai penentu prioritas Proses Bisnis menjadi informasi bagi Pimpinan PTMA dalam memprediksi kegiatan strategis. Nilai hasil analisis metode diperoleh bahwa 4 algoritma Pengambilan Keputusan yaitu mencapai Moora (75%), SAW (75%), WP (94%), dan WSM (94%)
Big Cats Classification Based on Body Covering
Get PDFThe reduced habitat owned by an animal has a very bad impact on the survival of the animal, resulting in a continuous decrease in the number of animal populations especially in animals belonging to the big cat family such as tigers, cheetahs, jaguars, and others. To overcome the decline in the animal population, a classification model was built to classify images that focuses on the pattern of body covering possessed by animals. However, in designing an accurate classification model with an optimal level of accuracy, it is necessary to consider many aspects such as the dataset used, the number of parameters, and computation time. In this study, we propose an animal image classification model that focuses on animal body covering by combining the Pyramid Histogram of Oriented Gradient (PHOG) as the feature extraction method and the Support Vector Machine (SVM) as the classifier. Initially, the input image is processed to take the body covering pattern of the animal and converted it into a grayscale image. Then, the image is segmented by employing the median filter and the Otsu method. Therefore, the noise contained in the image can be removed and the image can be segmented. The results of the segmentation image are then extracted by using the PHOG and then proceed with the classification process by implementing the SVM. The experimental results showed that the classification model has an accuracy of 91.07%. The reduced habitat owned by an animal has a very bad impact on the survival of the animal, resulting in a continuous decrease in the number of animal populations especially in animals belonging to the big cat family such as tigers, cheetahs, jaguars, and others. To overcome the decline in the animal population, a classification model was built to classify images that focuses on the pattern of body covering possessed by animals. However, in designing an accurate classification model with an optimal level of accuracy, it is necessary to consider many aspects such as the dataset used, the number of parameters, and computation time. In this study, we propose an animal image classification model that focuses on animal body covering by combining the Pyramid Histogram of Oriented Gradient (PHOG) as the feature extraction method and the Support Vector Machine (SVM) as the classifier. Initially, the input image is processed to take the body covering pattern of the animal and converted it into a grayscale image. Then, the image is segmented by employing the median filter and the Otsu method. Therefore, the noise contained in the image can be removed and the image can be segmented. The results of the segmentation image are then extracted by using the PHOG and then proceed with the classification process by implementing the SVM. The experimental results showed that the classification model has an accuracy of 91.07%.  
Chicken Egg Fertility Identification using FOS and BP-Neural Networks on Image Processing
Get PDFThis article aims to test FOS (first-order statistical) in extracting features of embryonated eggs. This test uses the initial step of image processing to get the best input image in feature extraction. The image processing method starts from the image acquisition process, then improves with image preprocessing and segmentation. Image acquisition in this study uses the concept of egg candling in a dark place captured with a smartphone camera. The acquisition results are improved by image preprocessing using gray scaling, image enhancement (by Histogram Equalization), and segmentation of chicken egg image. The segmentation results were extracted using FOS with five parameters: mean, entropy, variance, skewness, and kurtosis. Based on the calculation of these parameters, it is graphed and shows the difference in patterns between fertile and infertile eggs. However, some eggs have a similar pattern, thus affecting the identification process. The identification process used neural networks by the backpropagation method for training and testing. The training results provide an accuracy value of 100% of all training data; however, 80% of the new test data obtained test results at testing. This test is carried out with 100 data, 50 each for training and test data. Based on the test results, which significantly affect the level of accuracy is the feature extraction method. FOS pattern in detecting the fertility of chicken eggs by BP Neural Network is still categorized as low, so it is necessary to improve methods to get maximum results.This article aims to test FOS (first-order statistical) in extracting features of embryonated eggs. This test uses the initial step of image processing to get the best input image in feature extraction. The image processing method starts from the image acquisition process, then improves with image preprocessing and segmentation. Image acquisition in this study uses the concept of egg candling in a dark place captured with a smartphone camera. The acquisition results are improved by image preprocessing using gray scaling, image enhancement (by Histogram Equalization), and segmentation of chicken egg image. The segmentation results were extracted using FOS with five parameters: mean, entropy, variance, skewness, and kurtosis. Based on the calculation of these parameters, it is graphed and shows the difference in patterns between fertile and infertile eggs. However, some eggs have a similar pattern, thus affecting the identification process. The identification process used neural networks by the backpropagation method for training and testing. The training results provide an accuracy value of 100% of all training data; however, 80% of the new test data obtained test results at testing. This test is carried out with 100 data, 50 each for training and test data. Based on the test results, which significantly affect the level of accuracy is the feature extraction method. FOS pattern in detecting the fertility of chicken eggs by BP Neural Network is still categorized as low, so it is necessary to improve methods to get maximum results