Refine Search

New Search

Advanced search

Journal Jurnal Teknologi dan Sistem Komputer

-
175 articles
Page of 18
Articles per Page
by
Dea Alverina, Antonius Rachmat Chrismanto, R. Gunawan Santosa
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 76-83; doi:10.14710/jtsiskom.6.2.2018.76-83

Abstract:This research compared the accuracy of prediction of Grade Point Average (GPA) of the first semester students using C4.5 and CART algorithms in Faculty of Information Technology (FTI), Universitas Kristen Duta Wacana (UKDW). This research also explored various parameters such as numeric attribute categorization, data balance, GPA categories number, and different attributes availability due to the difference of data availability between Achievement Admission (AA) and Regular Admission (RA). The training data used to create decision tree were FTI students, 2008-2015 batch, while the testing data were FTI students, 2016 batch. The accuracy of prediction was measured by using crosstab table. In AA, the accuracy of both algorithms can be achieved about 86.86%. Meanwhile, in RA the accuracy of C4.5 is about 61.54% and CART is about 63.16%. From these accuracy result, both algorithms are better to predict AA rather than RA. Penelitian ini membandingkan akurasi prediksi kategori Indeks Prestasi (IP) semester pertama mahasiswa Fakultas Teknologi Informasi (FTI) Universitas Kristen Duta Wacana (UKDW) menggunakan algoritma C4.5 dan CART. Penelitian ini juga mengeksplorasi berbagai parameter seperti kategorisasi atribut numerik, keseimbangan data, jumlah kategori IP, dan ketersediaan atribut yang berbeda karena perbedaan ketersediaan data antara jalur prestasi dan jalur non-prestasi. Data mahasiswa FTI tahun 2008-2015 digunakan sebagai data latih sedangkan data uji menggunakan data tahun 2016. Akurasi kedua algoritma dalam memprediksi tersebut diukur dengan menggunakan tabel crosstab. Pada jalur prestasi, akurasi kedua algoritma mampu mencapai 86,86%. Pada jalur non-prestasi, akurasi algoritma C4.5 sebesar 61,54% dan algoritma CART sebesar 63,16%. Dilihat dari segi akurasinya, algoritma C4.5 dan CART lebih baik digunakan untuk memprediksi jalur prestasi daripada jalur non-prestasi.
Priyagung Hernawandra, Supriyadi Supriyadi, U. Tresna Lenggana
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 44-50; doi:10.14710/jtsiskom.6.2.2018.44-50

Joko Suryana
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 84-92; doi:10.14710/jtsiskom.6.2.2018.84-92

Abstract:This paper presents a new approach in designing an ultra wideband minimum dispersion antenna optimally to avoid the degradation of broadband communications system performance. Design and iterative optimization are applied to an arbitrary shape of planar monopole antenna using a genetic algorithm combined with the moment method, abbreviated as AGMM method, and implemented with Matlab. Two arbitrary shapes of planar monopole antennas have been implemented in compact physical size using AGMM optimization, each having 9.1 GHz and 7.4 GHz bandwidths, the lowest frequency of 1.9 GHz and 2.7 GHz and fidelity 0.6 and 0.64 for any arbitrary discrete antenna and edge profile antenna. This method can be applied to design any arbitrary shapes of an ultra-wideband antenna with each has wide bandwidth more than 7 GHz, the lowest frequency below 3 GHz and a minimum fidelity of 0,55 that is suitable for high- speed communication, such as 5G system.Makalah ini memaparkan pendekatan baru dalam perancangan optimal antena pita lebar dispersi minimum untuk menghindari penurunan kinerja sistem komunikasi kecepatan tinggi. Desain dan optimasi iteratif diterapkan pada antena planar monopole bentuk sembarang menggunakan algoritma genetika yang digabungkan dengan metode momen yang disingkat sebagai metode AGMM. Metode ini diimplementasikan dengan Matlab. Dua buah tipe antena planar monopole pita lebar bentuk sembarang dan ukuran fisik kompak berhasil dirancang dengan AGMM yang masing-masing memiliki lebar pita 9,1 GHz dan 7,4 GHz, frekuensi terendah 1,9 GHz dan 2,7 GHz serta memiliki fidelity 0,6 dan 0,64 untuk antena diskrit sembarang dan antena profil tepi sembarang. Metode ini dapat diterapkan untuk merancang antena pita lebar bentuk sembarang dengan lebar pita lebih dari 7 GHz, frekuensi terendah < 3 GHz dan memiliki fidelity di atas 0,55 yang cocok untuk komunikasi berkecepatan tinggi, misalnya sistem 5G.
Deni Setiyo Wibowo, Yessy Yanitasari, Dedih Dedih
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 71-75; doi:10.14710/jtsiskom.6.2.2018.71-75

Abstract:This research developed expert system using fuzzy Mamdani to diagnose the spread of diseases affecting the growth of chili plants based on environmental factors. Environmental variables used as fuzzy input parameters are soil pH, air temperature, air humidity and solar irradiation. The input-output relationship uses 81 rules with the AND operator and the MIN implication function. For a case, the results showed that the percentage of the potential spread of the disease showed 60.25%, so the category of potential disease spread with soil PH 7.5 pH, 28 ° C air temperature and air humidity 75 RH and 35% sun irradiance is moderate.Penelitian ini mengembangkan sistem pakar menggunakan fuzzy Mamdani untuk mendiagnosis penyebaran penyakit yang mempengaruhi pertumbuhan tanaman cabai berdasarkan faktor lingkungan. Variabel lingkungan yang digunakan sebagai parameter masukan fuzzy yaitu pH tanah, suhu udara, kelembaban udara dan penyinaran matahari. Relasi masukan-keluaran menggunakan 81 aturan dengan operator AND dan fungsi implikasi MIN. Untuk satu kasus, hasil penelitian menunjukkan nilai persentase potensi penyebaran penyakit adalah 60,25%, sehingga kategori potensi penyebaran penyakit dengan PH tanah 7.5 pH, suhu udara 28°C, kelembaban udara 75 RH dan penyinaran matahari 35% adalah sedang.
Maulana Hasan, Adnan Rafi Al Tahtawi
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 64-70; doi:10.14710/jtsiskom.6.2.2018.64-70

Abstract:This research aims to design fire early warning system using Arduino and Android with Bluetooth communication. The system is constructed using three sensors which are KY-026 flame detector, DS18B20 temperature sensor, and MQ-07 gas sensor. The communication used HC-05 Bluetooth module to send sensor data from Arduino to Android. Testing results show that Arduino through Bluetooth communication able to send sensor data to Android within range of 20 cm at no obstacle condition and range of 10 cm if there exists the obstacle. The system will send ‘normal’ status to Android when fulfilled this conditions: flame sensor value less than 200, temperatures less than 30°C, and gas levels less than 300 ppm. There are seven other conditions that contain ‘warning’ and ‘danger’ data if those condition above not fulfilled.Penelitian ini bertujuan untuk merancang detektor dini kebakaran menggunakan Arduino dan Android dengan komunikasi Bluetooth. Sistem dibuat menggunakan tiga sensor yaitu sensor api KY-026, sensor suhu DS18B20, dan sensor gas MQ-7. Komunikasi menggunakan modul Bluetooth HC-05 yang berfungsi untuk mengirim data sensor dari Arduino kemudian diterima Android. Hasil pengujian sistem menunjukkan bahwa Arduino melalui komunikasi Bluetooth dapat mengirim data ke Android dengan jarak 20 meter jika tidak ada hambatan dan jarak 10 meter jika ada hambatan. Apabila nilai sensor api kurang dari 200, suhu kurang dari 30°C, dan kadar gas kurang dari 100 ppm maka sistem akan mengirim status ‘normal’ ke aplikasi Android. Jika kondisi tersebut tidak terpenuhi, maka ada tujuh kondisi lain yang berisi peringatan ‘warning’ dan ‘bahaya’.
Putu Agus Fredy, Maman Abdurohman
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 139-145; doi:10.14710/jtsiskom.6.4.2018.139-145

Abstract:This paper presents a study on an accurate soil moisture monitoring system based on its humidity from 9 sensor nodes using wireless sensor network (WSN) and M2M platform. The system used IEEE 802.15.4 (Zigbee) protocol. The system was connected to the application via the OpenMTC M2M platform. This monitoring system can measure soil moisture accurately and provide soil water content status on the application. The system was effective in measuring soil moisture at a distance of 0-25 meters where there was a barrier between gateway and sensor, and at a distance of 0-50 meter in line of sight. The position of the sensors that are within 3 meters of each other and the depth of each sensor 3 cm can measure soil moisture properly.
Ari Fadli, Mulki Indana Zulfa, Yogi Ramadhani
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 158-163; doi:10.14710/jtsiskom.6.4.2018.158-163

Abstract:Observation of growing academic data can be carried using data mining methods, for example, to obtain knowledge related to the determinants of timeliness of students graduation. This study conducted a performance comparison of the classification algorithms using decision tree (DT), support vector machine (SVM), and artificial neural network (ANN). This study used students academic data from Faculty of Engineering, Universitas Jenderal Soedirman in the 2014/2015 odd semester until the 2017/2018 odd semester and the attributes that conform to the academic regulations. The analytical method used is CRISP-DM. The results showed that SVM provided the best performance in an accuracy of 90.55% and AUC of 0.959, compared to other algorithms. A Model with SVM algorithm can be implemented in an early warning system for timeliness of student graduation.
Endina Putri Purwandari, Rachmi Ulizah Hasibuan, Desi Andreswari
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 146-151; doi:10.14710/jtsiskom.6.4.2018.146-151

Abstract:Bamboo species can be identified from the bamboo leaf images. This study conducted the identification of bamboo species based on leaf texture using Gray Level Co-occurrence Matrix (GLCM) and Gray Level Run Length Matrix (GLRLM) for texture feature extraction, and Euclidean distance for measure the image distance. This study used the images of bamboo species in Bengkulu province, that are bambusa Vulgaris Var Vulgaris, bambusa Multiplex, bambusa Vulgaris Var Striata, Gigantochloa Robusta, Gigantochloa Schortrchinii, Gigantochloa Serik, Schizostachyum Brachycladum, and Dendrocalamus Asper. The bamboo application was built using Matlab. The accuracy of the application was 100% for bamboo leaf test images captured using a smartphone camera and 81.25% for test images downloaded from the Internet.
Mesran Mesran, Rusiana Rusiana, Maringan Sianturi
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 135-138; doi:10.14710/jtsiskom.6.4.2018.135-138

Abstract:This study aims to develop a decision support system in determining employees which will be laid off. The data used in the research sourced from PT. Mitra Andal Sejati Medan in 2017. This study used the ELECTRE method to conduct an assessment of company employees. The criteria used in the research were absence (C1), appearance (C2), performance (C3), and sales volume (C4). This ELECTRE system can provide an effective decision for termination of employment based on lowest employee rank.
Jumoke Falilat Ajao, David Olufemi Olawuyi, Odetunji Ode Odejobi
Jurnal Teknologi dan Sistem Komputer, Volume 6, pp 129-134; doi:10.14710/jtsiskom.6.4.2018.129-134

Abstract:This work presents a recognition system for Offline Yoruba characters recognition using Freeman chain code and K-Nearest Neighbor (KNN). Most of the Latin word recognition and character recognition have used k-nearest neighbor classifier and other classification algorithms. Research tends to explore the same recognition capability on Yoruba characters recognition. Data were collected from adult indigenous writers and the scanned images were subjected to some level of preprocessing to enhance the quality of the digitized images. Freeman chain code was used to extract the features of THE digitized images and KNN was used to classify the characters based on feature space. The performance of the KNN was compared with other classification algorithms that used Support Vector Machine (SVM) and Bayes classifier for recognition of Yoruba characters. It was observed that the recognition accuracy of the KNN classification algorithm and the Freeman chain code is 87.7%, which outperformed other classifiers used on Yoruba characters.
Page of 18
Articles per Page
by