This study explores the application of CNN-Transfer Learning for nondestructive chicken egg fertility detection. Four models, VGG16, ResNet50, InceptionNet, and MobileNet, were trained and evaluated on a dataset using augmented images. The training results demonstrated that all models achieved high accuracy, indicating their ability to accurately learn and classify chicken eggs’ fertility state. However, when evaluated on the testing set, variations in accuracy and performance were observed. VGG16 achieved a high accuracy of 0.9803 on the testing set but had challenges in accurately detecting fertile eggs, as indicated by a NaN sensitivity value. ResNet50 also achieved an accuracy of 0.98 but struggled to identify fertile and non-fertile eggs, as suggested by NaN values for sensitivity and specificity. However, InceptionNet demonstrated excellent performance, with an accuracy of 0.9804, a sensitivity of 1 for detecting fertile eggs, and a specificity of 0.9615 for identifying non-fertile eggs. MobileNet achieved an accuracy of 0.9804 on the testing set; however, it faced challenges in accurately classifying the fertility status of chicken eggs, as indicated by NaN values for both sensitivity and specificity. While the models showed promise during training, variations in accuracy and performance were observed during testing. InceptionNet exhibited the best overall performance, accurately classifying fertile and non-fertile eggs. Further optimization and fine-tuning of the models are necessary to address the limitations in accurately detecting fertile and non-fertile eggs. This study highlights the potential of CNN-Transfer Learning for nondestructive fertility detection and emphasizes the need for further research to enhance the models’ capabilities and ensure accurate classification.
Abstrak Sengon (Falcataria moluccana) merupakan jenis pohon yang secara alami tersebar di Indonesia. Masa panen yang cenderung singkat dan nilai eknonomis tinggi, menjadikan sengon banyak diusahakan untuk berbagai keperluan dalam bentuk kayu olahan. Sehingga, kecenderungan hutan tanaman mengarah pada pertanaman sengon dengan skala luas dan bersifat monokultur, dan berdampak munculnya penyakit. Penelitian ini bertujuan mendesain basis data sistem pakar untuk identifikasi penyakit pada sengon. Metode desain basis data yang digunakan adalah dengan pendekatan model entity relationship (ER) untuk menghasilkan model relational, yang selanjutnya dilakukan normalisasi data sampai 3 rd NF. Dari penelitian ini diperoleh model data ER dengan tujuh himpunan entitas yang direlasikan dengan lima himpunan relasi dengan model pemetaaan kardinalitasnya one to many/many to one dan many to many. Hasil Konversi model ER ke model relational diperoleh sepuluh table dengan pemetaan kardinalitasnya one to many/many to one dan sudah dalam kondisi 3 rd NF. Kata kunci: basis data, entity relationship, model data relational, normalisasi, sistem pakar
The prediction and its accuracy of the rainfall is needed due to it would be affected to the various areas of life, such as feasibility aircraft departures and, in general issue, is climate change. This paper aimed to apply a Long Short Term Memory (LSTM) approach to get accurate rainfall forecasting. Also, the LSTM accuracy would be compared to BPNN (Backpropagation Neural Network) algorithm. In this research, LSTM architecture used a hidden layer of 200, a maximum epoch of 250, 1 gradient threshold, and learning rates of 0.005, 0.007, and 0.009. Then, standardize data was used gamma γ of 1.05. Then, the BPNN architectures of [2-50-10-1, epoch 250] have been explored. The accuracy performance is measured by the root means square error (RMSE). The experimental results showed that the LSTM had produced a good accuracy than BPNN, with the value of RMSE was 0.2367 and 0.1938. It means that the forecast accuracy of the LSTM approach outperformed the BPNN to predict the rainfall. This finding would be useful for the climatology station to develop a forecsat rainfall application-based artificial intelligence.
Fungsi dan peran Perguruan Tinggi tidak hanya ketika mahasiswa masih aktif dalam perkuliahan atau sebelum lulus. Namun, setelah menjadi alumni, Perguruan Tinggi masih diperlukan layanan dan kontribusinya, diantaranya dalam melakukan verifikasi dan legalisasi dokumen yang dikeluarkannya, terutama ijazah dan transkip akademik. Di FKIP khususnya, dan lebih luas lagi di Universitas Ahmad Dahlan, sistem layanan legalisasi ijazah dan tanskrip akademik masih bersifat konvensional, yaitu layanan hanya dapat dilakukan on site (ditempat) dan belum tentu dapat diselesaikan secara cepat seperti yang diharapkan. Sehingga, alumni harus datang lebih dari sekali secara langsung ke tempat layanan untuk melakukan legalisasi ijazah dan transkrip akademik. Tentu saja hal ini mengakibatkan terjadi pemborosan dalam hal waktu dan biaya yang dikeluarkan, terlebih bagi alumni yang tinggal jauh dari kampus/luar kota. Oleh karena itu, diperlukan sistem alternatif yang mampu memberikan solusi permasalahan dimaksud. Penelitian ini bertujuan untuk merancang dan mengimplementasikan website sebagai media layanan legalisir ijazah dan transkrip akademik. Pengumpulan data dilakukan dengan metode wawancara, observasi, dan telaah dokumen yang ada. Sedangkan, metode pengembangan sistem yang akan digunakan dalam adalah metode prototype. Metode prototype, lebih berorientasi pada kepuasan pengguna. Oleh karena itu, selama proses sistem dikerjakan terus terjadi interaksi antar pengguna dan developer agar sistem yang dibangun benar-benar mampu menjawab permasalahan yang ada. Hasil dari penelitian yang diuraikan dalam naskah ini berupa perancangan sistem online sebagai media layanan alumni untuk keperluan legalisasi copy ijazah dan transkrip akademik, yang terdiri dari desain basis data dan desain sistem menggunakan Unified Moedelling Language, serta perancangan antar muka halaman utama sistem layanan online (SILON) legalisir ijazah dan transkrip akademik. Kata Kunci : ijazah, layanan online , legalisir, transkrip akademik, website, SILON.
Android technology developments that are currently able to occupy the highest positions of gadgets and computer market, it is certainly due to the sophistication of technology information and applications that are on it that is currently a trend among mobile users because it can help all areas of the job so much easier.The purpose of Technopreneurship in the field of Design and implementation of online fashion store based on android is designed to assist in the sales transaction of business units called "Demi Outfits" has been established since 2013, making it easier for the android-based online transactions.Thus the computer tools and android smartphone is necessary given the various transactions were originally done manually.Given these tools various transactions can be completed quickly and efficiently than when using a manual system.To expedite the process of search services and purchases by customers towards these online stores, we need a system that allows customers to access the service online stores, especially the process of buying fashion online using android based mobile applications store.This application will also provide facilities that assist in the search for collection until the transaction purchases by customers.The system will tell to customers whether the collection to be purchased is available or not and their status are trends or expired.
The primary objective of this study is to analyze multivariate time series data by employing the Long Short-Term Memory (LSTM) model. Deep learning models often face issues when dealing with multivariate time series data, which is defined by several variables that have diverse value ranges. These challenges arise owing to the potential biases present in the data. In order to tackle this issue, it is crucial to employ normalization techniques such as min-max and z-score to guarantee that the qualities are standardized and can be compared effectively. This study assesses the effectiveness of the LSTM model by applying two normalizing techniques in five distinct attribute selection scenarios. The aim of this study is to ascertain the normalization strategy that produces the most precise outcomes when employed in the LSTM model for the analysis of multivariate time series. The evaluation measures employed in this study comprise Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and R-Squared (R2). The results suggest that the min-max normalization method regularly yields superior outcomes in comparison to the z-score method. Min-max normalization specifically resulted in a decreased mean absolute percentage error (MAPE) and root mean square error (RMSE), as well as an increased R-squared (R2) value. These improvements indicate enhanced accuracy and performance of the model. This paper makes a significant contribution by doing a thorough comparison analysis of normalizing procedures. It offers vital insights for researchers and practitioners in choosing suitable preprocessing strategies to improve the performance of deep learning models. The study's findings underscore the importance of selecting the appropriate normalization strategy to enhance the precision and dependability of multivariate time series predictions using LSTM models. To summarize, the results indicate that min-max normalization is superior to z-score normalization for this particular use case. This provides a useful suggestion for further studies and practical applications in the field. This study emphasizes the significance of normalization in analyzing multivariate time series and contributes to the larger comprehension of data preprocessing in deep learning models
This document shows a performance analysis between SDN and a conventional IP network configured with the EIGRP and BGP routing protocols, establishing a configuration scenario with physical network equipment and with an SDN emulator called Mininet. The research methodology is based on the guidelines of the Cisco PPDIOO methodology and is developed in the following phases: 1. Elaboration of physical network topology with Cisco equipment, performing experiments with IPv4 and IPv6, measuring variables such as Jitter, Delay and Throughput. 2. Carrying out the same experiments and tests with SDN, in a network topology with similar characteristics to those already mentioned, but with OpenFlow switches. 3. Analysis of results, for which the behavior of jitter, delay and throughput variations of both scenarios is examined to make a series of comparisons (made with statistical analysis) concerning protocol, addressing, packet size among others. Finally, it was obtained as a result that SDN has a lower delay and jitter than the conventional IP network in some cases, as well as a more favorable throughput.
Zakat berperan untuk mencapai keadilan sosial ekonomi antara orang kaya dan miskin. Saat ini terdapat Lembaga Amil Zakat (LAZ) yang berperan penting dalam pengelolaan dana Zakat, Infaq, dan Shodaqah. Namun ada beberapa faktor kekurangan dalam hal penyaluran dana zakat yaitu siapa yang berhak menerima dana zakat dengan tepat sasaran dan cabang-cabang mana saja yang berpotensi mandiri dalam pengelolaan zakat. Klasifikasi dapat digunakan untuk menilai ketepatan penyaluran zakat dan mengetahui kemandirian tiap-tiap cabang LAZ berdasarkan data-data masa lalu. Data tersebut bisa digunakan untuk menerapkan metode K-NN sehingga dapat mengklasifikasi dana zakat menurut kelasnya. Penelitian ini dilakukan untuk mengkaji tentang algoritma K-NN dan mengimplementasikan Algoritma K-NN dalam klasifikasi data. Data yang digunakan adalah data penyaluran dana zakat di Lazismu DIY dari tahun 2013 sampai 2015.Data penyaluran zakat dari cabang-cabang LAZ yang telah melalui proses cleaning data, integration data, selection data, transformation data,dan analisis diproses menggunakan metode K-Nearest Neighbor (K-NN) untuk mengklasifikasikan cabang-cabang yang berpotensi membantu perekonomian daerah (mandiri) dan penyaluran dana zakat yang tepat sasaran berdasarkan tingkat kemiripan sejumlah nilai variabel k. Proses algoritma K-NN di buatmenghasilkanpattern evaluationdandisajikanmelalui knowledge presentationdenganbantuan web framework. Hasil pengujian dilakukan terhadap 14 cabang Lazismu di DIY menghasilkan tidakada cabang di kelas Super Mandiri, 6 cabang berada dikelas Mandiri, tidakada cabang berada dikelas Cukup Mandiri, dan 8 cabang berada pada kelas Kurang Mandiri. Hasil confusion matrix dengan perbandingan 80:20 dari data uji dan data testing menghasilkan nilai accuracy sebesar 85% dan error-rate sebesar25%. Hasil accuracy>= 85% dikatakan baik dalam klasifikasi tersebut membuktikan bahwa faktor-faktor nilai atribut yang dipilih mendekati nilai significant
Following the United States, China is the world's second-biggest economy and the world's largest exporter of goods. Chinese and Indonesian economies differ significantly in terms of development and implementation and the availability of technology and transportation to support them, particularly in terms of GDP value and other indicators of economic health. The topography and climate of China and Indonesia are covered in detail in this study, as is the economic relationship between China and Indonesia, covered in the second section. According to the economic viewpoints, the economic topic incorporates technology and transportation. Both countries must improve their cooperation to reap mutual benefits in the future.
