Jurnal Teknologi dan Sistem Komputer
Not a member yet
    364 research outputs found

    Comparative analysis of classification algorithms for critical land prediction in agricultural cultivation areas

    Get PDF
    Currently, the identification of critical land, that has been physically, chemically, and biologically damaged, uses a geographic information system. However, it requires a high cost to get the high resolution of satellite images. In this study, a comparison framework is proposed to determine the performance of the classification algorithms, namely C.45, ID3, Random Forest, k-Nearest Neighbor, and Naïve Bayes. This research aims to find out the best algorithm for the classification of critical land in agricultural cultivation areas. The results show that the highest accuracy Random Forest algorithm was 93.10 % in predicting critical land, and the naïve Bayes has the lowest performance, with 89.32 % of accuracy in predicting critical land

    Strategi caching aplikasi berbasis in-memory menggunakan Redis server untuk mempercepat akses data relasional

    Get PDF
    Utilization of an in-memory database as a cache can overcome relational database latency problems in a web application, especially when using a lot of join queries. This study aims to model the academic relational data into Redis compatible data and analyze the performance of join queries usage to accelerate access to relational data managed by RDBMS. This study used academic data to calculate student GPA that is modeled in the RDBMS and Redis in-memory database (IMDB). The use of Redis as an in-memory database can significantly increase Mysql database system performance up to 3.3 times faster to display student data using join query and to shorten the time needed to display GPA data to 52 microseconds from 61 milliseconds.Pemanfaatan basis data in-memory sebagai cache dapat menjadi solusi untuk mengatasi latensi basis data relational dalam pengelolaan data terstruktur di aplikasi web, terutama di data relasional yang banyak menggunakan join query. Penelitian ini bertujuan mengkaji pemodelan data relasional akademik menjadi data yang kompatibel dengan Redis dan menganalisis kinerjanya dalam penggunaan join query untuk mempercepat akses data relasional yang dikelola oleh RDBMS. Kajian ini menggunakan data akademik untuk menghitung IPK mahasiswa yang dimodelkan dalam RDBMS dan in-memory database (IMDB) Redis. Penggunaan Redis sebagai basis data in-memory dapat menaikkan kinerja sistem Mysql secara signifikan hingga 1,7 kali lebih cepat dalam membantu mempersingkat waktu yang dibutuhkan dalam menampilkan data mahasiswa menggunakan join query dan waktu pembacaan data IPK hingga menjadi 52 mikrodetik dari 61 milidetik

    Segmentation of university customers loyalty based on RFM analysis using fuzzy c-means clustering

    Get PDF
    One of the strategic plans of the developing universities in obtaining new students is forming a partnership with surrounding high schools. However, partnerships made does not always behave as expected. This paper presented the segmentation technique to the previous new student admission dataset using the integration of recency, frequency, and monetary (RFM) analysis and fuzzy c-means (FCM) algorithm to evaluate the loyalty of the entire school that has bound the partnership with the institution. The dataset is converted using the RFM approach before processed with the FCM algorithm. The result reveals that the schools can be segmented, respectively, as high potential (SP), potential (P), low potential (CP), and very low potential (KP) categories with PCI value 0.86. From the analysis of SP, P, and CP, only 71 % of 52 school partners categorized as loyal partners

    Unjuk kerja k-nearest neighbor untuk alihaksara citra aksara Nusantara

    Get PDF
    The concept of classification using the k-nearest neighbor (KNN) method is simple, easy to understand, and easy to be implemented in the system. The main challenge in classification with KNN is determining the proximity measure of an object and how to make a compact reference class. This paper studied the implementation of the KNN for the automatic transliteration of Javanese, Sundanese, and Bataknese script images into Roman script. The study used the KNN algorithm with the number k set to 1, 3, 5, 7, and 9. Tests used the image dataset of 2520 data. With the 3-fold and 10-fold cross-validation, the results exposed the accuracy differences if the area of the extracted image, the number of neighbors in the classification, and the number of data training were different.Konsep klasifikasi menggunakan metode k-nearest neighbor (KNN) sederhana, mudah dimengerti, dan diimplementasikan dalam sistem. Tantangan utama dalam klasifikasi dengan KNN adalah menentukan sejauh mana luas perimeter untuk mengukur kedekatan objek dan bagaimana membuat kelas training yang handal. Kajian ini menjelaskan hasil implementasi KNN untuk transliterasi otomatis aksara Jawa, Sunda, dan Batak ke dalam aksara Roman. Kajian menggunakan algoritme KNN dengan nilai k diatur ke 1, 3, 5, 7, dan 9. Pengujian dilakukan pada kumpulan data citra sebanyak 2520 data. Dengan validasi silang 3-fold dan 10-fold, hasil menunjukkan bahwa ada perbedaan akurasi jika area gambar pada saat diekstraksi, jumlah tetangga dalam klasifikasi dan jumlah data latih berbeda

    Sistem pemantauan tanah longsor berdasarkan laju adsorpsi air pada tanah menggunakan sensor kelembapan, kemiringan, dan suhu

    Get PDF
    This study examines the application of a landslide disaster monitoring system based on soil activity information that utilizes humidity, temperature, and accelerometer sensors. An artificial highland was built as the research object, and the landslide process was triggered by supplying the system with continuous artificial rainfall. The soil activities were observed through its slope movement, temperature, and moisture content, utilizing an accelerometer, temperature, and humidity sensors both in dry and wet conditions. The system could well observe the soil activities, and the obtained data could be accessed in real-time and online mode on a website. The time delay in sending the data to the server was 2 seconds. Moreover, the characteristics of soil porosity and its relevance to soil saturation level due to water pressure were studied as well. Kinetic study showed that the water adsorption to soil followed the intraparticle diffusion model with a coefficient of determination R2 0.99043. The system prototype should be used to build the information center of disaster mitigation, particularly in Indonesia.Artikel ini mengkaji aplikasi sistem pemantauan bencana tanah longsor berdasarkan informasi aktivitas tanah yang memanfaatkan sensor kelembapan, suhu, dan accelerometer. Sebuah purwarupa dataran tinggi digunakan sebagai objek studi tanah longsor dan perubahan keadaan tanah dipantau dengan memberikan hujan buatan secara kontinyu kepada sistem. Aktivitas tanah dipantau berdasarkan data kemiringan, suhu, dan juga kelembapan tanah, baik saat keadaan kering maupun basah. Hasil pengukuran menunjukkan bahwa sistem pemantauan dapat merekam aktivitas tanah dengan baik dan informasi tersebut dapat diakses secara online dan real-time melalui sebuah situs web. Estimasi waktu tunda pengiriman data menuju server adalah 2 detik. Karakteristik porisitas tanah dan relevansinya terhadap tingkat saturasi tanah karena tekanan air telah dikaji. Hasil studi kinetik menunjukkan bahwa proses adsorpsi air ke tanah mengikuti model kinetik difusi intra partikel, dengan coefficient of determination R2 sebesar 0,99043. Sistem pemantauan yang dirancang diharapkan menjadi purwarupa awal untuk terbentuknya pusat informasi mitigasi bencana tanah longsor yang dapat diterapkan, khususnya di Indonesia

    Tree-based homogeneous ensemble model with feature selection for diabetic retinopathy prediction

    Get PDF
    Diabetic Retinopathy (DR) is a condition that emerges from prolonged diabetes, causing severe damages to the eyes. Early diagnosis of this disease is highly imperative as late diagnosis may be fatal. Existing studies employed machine learning approaches with Support Vector Machines (SVM) having the highest performance on most analyses and Decision Trees (DT) having the lowest. However, SVM has been known to suffer from parameter and kernel selection problems, which undermine its predictive capability. Hence, this study presents homogenous ensemble classification methods with DT as the base classifier to optimize predictive performance. Boosting and Bagging ensemble methods with feature selection were employed, and experiments were carried out using Python Scikit Learn libraries on DR datasets extracted from UCI Machine Learning repository. Experimental results showed that Bagged and Boosted DT were better than SVM. Specifically, Bagged DT performed best with accuracy 65.38 %, f-score 0.664, and AUC 0.731, followed by Boosted DT with accuracy 65.42 %, f-score 0.655, and AUC 0.724 when compared to SVM (accuracy 65.16 %, f-score 0.652, and AUC 0.721). These results indicate that DT's predictive performance can be optimized by employing the homogeneous ensemble methods to outperform SVM in predicting DR

    Optimasi nilai k dan parameter lag algoritme k-nearest neighbor pada prediksi tingkat hunian hotel

    Get PDF
    Hotel occupancy rates are the most important factor in hotel business management. Prediction of the rates for the next few months determines the manager's decision to arrange and provide all the needed facilities. This study performs the optimization of lag parameters and k values of the k-Nearest Neighbor algorithm on hotel occupancy history data. Historical data were arranged in the form of supervised training data, with the number of columns per row according to the lag parameter and the number of prediction targets. The kNN algorithm was applied using 10-fold cross-validation and k-value variations from 1-30. The optimal lag was obtained at intervals of 14-17 and the optimal k at intervals of 5-13 to predict occupancy rates of 1, 3, 6, 9, and 12 months later. The obtained k-value does not follow the rule at the square root of the number of sample data.Tingkat hunian hotel merupakan faktor terpenting dalam pengelolaan bisnis perhotelan. Prediksi tingkat hunian hotel untuk beberapa bulan ke depan menentukan keputusan pengelola untuk mengatur dan menyediakan semua fasilitas yang diperlukan di hotel tersebut. Penelitian ini melakukan optimalisasi parameter lag dan nilai optimal k dari algoritme k-Nearest Neighbor pada data histori tingkat hunian hotel. Data histori tingkat hunian hotel disusun dalam bentuk data pelatihan supervised dengan jumlah kolom setiap baris sesuai dengan parameter lag dan jumlah target prediksi. Algoritme kNN diterapkan dengan menggunakan validasi silang 10-fold dan variasi nilai k dari 1-30. Dari hasil uji coba didapatkan lag optimal diperoleh pada interval 14-17 dan nilai k optimal pada interval 5-13 untuk prediksi tingkat hunian 1, 3, 6, 9, dan 12 bulan berikutnya. Nilai k terbaik yang diperoleh tidak mengikuti kaidah pada akar kuadrat jumlah sampel data

    Implementasi vigenere cipher 128 dan rotasi bujursangkar untuk pengamanan teks

    Get PDF
    Information that can be in the form of text, image, audio, and video, is a valuable asset that needs to be secured from unauthorized parties. This research aims to study the implementation of Vigenere cipher 128 (VC-128) and square rotation to secure text information. The square rotation is applied to increase the security of the encryption results obtained from VC-128. The randomness of the rotation results was measured using Shannon entropy based on the distance between characters, and the Avalanche Effect measured changes in the encryption results compared to the original text. The square rotation can increase the randomness of the VC-128 encryption results, as indicated by an increase in entropy values. The highest increase in entropy of 34.8 % occurs in repetitive texts with the square size that produces optimal entropy was a 9x9 medium-size square. The Avalanche effect for each test data shows inconsistent results ranging from 44.5 % to 49 %.Informasi merupakan aset berharga yang keamanannya perlu dilindungi dari pihak-pihak yang tidak berhak. Menurut bentuknya, informasi dapat berbentuk teks, citra, audio dan video. Penelitian ini bertujuan untuk melindungi informasi yang disimpan dalam bentuk teks. Metode yang digunakan adalah Vigenere cipher 128 (VC-128) dan rotasi bujursangkar. Rotasi bujursangkar digunakan untuk meningkatkan keamanan dari hasil enkripsi yang diperoleh dari VC-128. Keacakan hasil rotasi diukur menggunakan entropi Shannon berdasarkan jarak antar karakter, sedangkan efek Avalanche digunakan untuk mengukur perubahan hasil enkripsi dibandingkan dengan teks aslinya. Hasil penelitian menunjukkan bahwa rotasi bujursangkar mampu meningkatkan keacakan hasil enkripsi VC-128 yang ditunjukkan dengan adanya peningkatan nilai entropi. Peningkatan entropi tertinggi sebesar 34,8 % terjadi pada teks berulang dengan ukuran bujursangkar yang menghasilkan entropi optimal adalah bujursangkar berukuran sedang, yaitu 9×9. Nilai efek Avalanche untuk setiap data uji memberikan hasil yang tidak konsisten, berkisar antara 44,5 % hingga 49 %

    Optimasi decision tree menggunakan particle swarm optimization untuk identifikasi penyakit mata berdasarkan analisis tekstur

    Get PDF
    The problem of visual impairment is a serious problem with increasing cases, ranging from visual impairment to the cause of blindness. This study examines the development of an identification application for the classification of patients with eye disorders using the Decision Tree (DT) method, which is optimized using Particle Swarm Optimization (PSO). This study used 311 eye image data, consisting of 233 normal eye images and 78 eye images with glaucoma, cataracts, and uveitis. The feature extraction used Gray Level Co-occurrence Matrix (GLCM), while the feature optimization used the PSO and the learning method used DT. This optimized visual impairment classification application can improve system accuracy to 88.09 %.Masalah gangguan penglihatan merupakan masalah serius seiring peningkatan kasus gangguan penglihatan hingga menyebabkan kebutaan. Penelitian ini mengkaji pengembangan aplikasi identifikasi untuk klasifikasi penderita gangguan mata dengan menggunakan metode Decision Tree (DT) yang dioptimasi menggunakan Particle Swarm Optimization (PSO). Kajian ini menggunakan 311 data citra mata, yang terdiri atas 233 citra mata normal dan 78 citra mata berpenyakit glaukoma, katarak, dan uveitis. Gray Level Co-occurrence Matrix (GLCM) digunakan untuk ekstraksi fitur, sedangkan PSO digunakan untuk optimasi fitur dan DT sebagai metode pembelajarannya. Aplikasi klasifikasi gangguan penglihatan teroptimasi ini dapat meningkatkan akurasi sistem menjadi 88,09 %

    Neural network performance based on backpropagation and LVQ as the LoRa RSS fingerprint algorithms for positioning in an open space

    Get PDF
    Penentuan posisi ruang terbuka merupakan salah satu aplikasi penting pada internet of things. Penggunaan GPS tidak cocok untuk perangkat IoT yang berdaya rendah. Sebagai alternatif digunakan perangkat LoRa. Penelitian ini bertujuan untuk menemukan metode yang lebih baik sebagai algoritme fingerprint dalam menentukan posisi objek pada ruang terbuka berdasarkan RSS LoRa. Metode yang digunakan sebagai algoritme fingerprint adalah dua model jaringan saraf tiruan, yaitu backpropagation (BP) dengan empat jenis metode pelatihan dan learning vector quantization (LVQ) dengan dua metode pelatihan. Hasil eksperimen menunjukkan bahwa kinerja LVQ1 lebih baik dibanding LVQ2. Selain itu, kinerja LVQ1 juga lebih baik dibandingkan BP, sedangkan metode BP dan LVQ2 memiliki tingkat keberhasilan hampir sama di sekitar 70 %. Kedua model jaringan saraf tiruan, baik BP maupun LVQ, dapat digunakan sebagai algoritme fingerprint untuk menentukan posisi objek pada ruang terbuka dengan akurasi yang cukup tinggi.Outdoor positioning is one of the important applications in the Internet of things (IoT). The usage of GPS is unsuitable for low-power IoT devices. Alternatively, it can use the LoRa devices. This research aims to find a better method as the fingerprint algorithm for determining the outdoor position using RSS LoRa. The methods used as the fingerprint algorithm were two artificial neural network models, i.e. backpropagation (BP) with four types of training methods and learning vector quantization (LVQ) with two types of training methods. The experiment results show the performance of LVQ1 better than those of LVQ2. Besides, the LVQ1 was also better than the BP method. However, both BP and LVQ2 have a performance that is almost similar to about 70 %. Both of the artificial neural network models, BP and LVQ, can be used as a fingerprint algorithm to determine quite accurate the outdoor object position

    318

    full texts

    364

    metadata records
    Updated in last 30 days.
    Jurnal Teknologi dan Sistem Komputer
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇