Jurnal Teknik Informatika dan Sistem Informasi
Not a member yet
499 research outputs found
Sort by
Penerapan Machine Learning untuk Penentuan Mata kuliah Pilihan pada Program Studi Informatika
Get PDFInformatics study program at Nurul Jadid University does not have a general concentration of knowledge, so that sometimes the selection of elective courses by students is not quite right. This study aims to classify the concentration of knowledge with a data mining approach which can then be used as a recommendation for selecting elective courses by students. In this study, we implement a machine learning algorithm to provide recommendations to students regarding what interests are more suitable to be taken based on the values of prerequisite courses in previous semesters. Student data was obtained from the Head of the Center for Data and Information Systems (PDSI) at Nurul Jadid University with 70 student data from Nurul Jadid University batch 2018. The machine learning algorithm used is Neural Network with Python programming language, the tools used are Google Collab. At the beginning of data collection, then pre-processing is carried out to prepare the dataset in order to get good results, and model training is carried out. After training on the model, then further testing is carried out on the model to determine the performance of the model. The result of the accuracy value in the training model process is 0.83 or 83% and the accuracy of the test data is 0.79 or 79%.Program studi informatika di Universitas Nurul Jadid tidak memiliki konsentrasi ilmu pengetahuan yang bersifat umum, sehingga terkadang pemilihan mata kuliah pilihan oleh mahasiswa kurang tepat. Penelitian ini bertujuan untuk mengklasifikasikan konsentrasi pengetahuan dengan pendekatan data mining yang selanjutnya dapat dijadikan sebagai rekomendasi pemilihan mata kuliah pilihan oleh mahasiswa. Pada penelitian ini, kami mengimplementasikan algoritma machine learning untuk memberikan rekomendasi kepada mahasiswa mengenai minat apa yang lebih cocok untuk diambil berdasarkan nilai mata kuliah prasyarat pada semester sebelumnya. Data mahasiswa diperoleh dari Kepala Pusat Data dan Sistem Informasi (PDSI) Universitas Nurul Jadid sebanyak 70 data mahasiswa Universitas Nurul Jadid angkatan 2018. Algoritma machine learning yang digunakan adalah Neural Network dengan bahasa pemrograman Python, tools yang digunakan adalah Google Kolaborasi. Pada awal pengumpulan data, kemudian dilakukan preprocessing untuk mempersiapkan dataset agar mendapatkan hasil yang baik, dan dilakukan pelatihan model. Setelah dilakukan pelatihan pada model, selanjutnya dilakukan pengujian lebih lanjut pada model untuk mengetahui performansi model. Hasil nilai akurasi pada proses model pelatihan adalah 0,83 atau 83% dan akurasi data uji adalah 0,79 atau 79%
Pemanfaatan k-Means Clustering dan Analytic Hierarchy Process terhadap Penilaian Prestasi Kerja Pegawai
Get PDF As a government agency related to education and culture, the Department of Education and Culture of Bengkayang Regency needs qualified employees in their performance. Clustering can be used to determine whether the employee is performing well, moderately, or poorly. The clustering method used in this study is the k-Means method. The research wasconducted by studying and understanding the k-Means method and knowing employee performance data at the Education and Culture Office of Bengkayang Regency. The results of calculations using the k-Means method and the Analytic Hierarchy Process (AHP) method are as many as six employees have good performance (where two of them got the highest score in the AHP calculation), nine employees have moderate performance, and five employees have poor performance. These results can be used as a benchmark for employees in the cluster either to be promoted or rank, while employees in the cluster less are able to begiven employee performance training, in order to be better in the future. With this, employees can become more competitive and superior in the face of increasingly rapid developments. As a government agency related to education and culture, the Department of Education and Culture of Bengkayang Regency needs qualified employees in their performance. Clustering can be used to determine whether the employee is performing well, moderately, or poorly. The clustering method used in this study is the k-Means method. The research wasconducted by studying and understanding the k-Means method and knowing employee performance data at the Education and Culture Office of Bengkayang Regency. The results of calculations using the k-Means method and the Analytic Hierarchy Process (AHP) method are as many as six employees have good performance (where two of them got the highest score in the AHP calculation), nine employees have moderate performance, and five employees have poor performance. These results can be used as a benchmark for employees in the cluster either to be promoted or rank, while employees in the cluster less are able to begiven employee performance training, in order to be better in the future. With this, employees can become more competitive and superior in the face of increasingly rapid developments
Perancangan Aplikasi Pembukuan Menggunakan Metode Agile Scrum
Get PDFThe product and transaction management section's business process in the Sepatu Kaki Kaki has not maximized the existing system application. This makes the business processes carried out by the store less effective. The method used in this study uses one of the agile software development models, namely Scrum. Which has advantages, including flexible requirements, speed up application production time, and require fewer team members. This research aims to produce application development that can manage products effectively and help staff analyze in accordance with the expected targets in the future. The application development uses the Kotlin programming language - Android Native, with BaaS (Backend as a Service) Firebase. The results of this research are an android application to speed up and simplify product management, the latest product stock can be known, and minimize manual recording errors.Proses bisnis pada bisnis toko Sepatu Kaki Kaki pada bagian produk dan manajemen transaksi belum memaksimalkan sistem aplikasi yang ada. Hal ini membuat proses bisnis yang dilakukan oleh toko menjadi kurang efisien. Metodologi yang digunakan yaitu Scrum, salah satu model agile. Kelebihannya antara lain persyaratan yang fleksibel, waktu produksi aplikasi yang lebih cepat, dan anggota tim yang tidak banyak. Tujuan dari penelitian ini adalah untuk mengelola produk secara efektif dan membantu menganalisis penjualan berdasarkan tujuan yang diharapkan di masa depan. Bahasa pemrograman Kotlin - Android Native dengan BaaS (Backend as a Service) Firebase digunakan untuk mengembangkan aplikasi ini. Adapun penelitian ini akan menghasilkan sebuah aplikasi android yang mempercepat dan mempermudah pengelolaan produk, mengetahui persediaan produk terbaru
Serverless Named Entity Recognition untuk Teks Instruksional Pertanian Kota
Get PDFThe evolution of document documentation, classification, and information retrieval includes named entity recognition (NER). The implementation of NER in the agricultural domain, in particular instructional texts or transcriptions of tutorial videos, will make it easier for the general public to understand the specific concepts and terms of urban agricultural activities such as crop production processes and procedures, agricultural methods and tools, harvest cycles, and handling plant pests or diseases. Spacy is an NLP tool, has two methods of developing NER models, namely with Toc2Vec and Transformer. Both methods have advantages and disadvantages, namely different sizes, performance and prediction speeds according to needs. The NER model can be implemented into a Serverless application, using the Functional as Services (FaaS) and Backend as Services (BaaS) approaches. For the subtopic of cultivating fruit crops in agricultural instructional literature, three NER models have been built in this study. First, the IndoBERT-based model, the Toc2Vec-based model with efficiency optimization, and the Toc2Vec-based model with accuracy optimization. The most efficient toc2vec model, with a f1-score of 0.71, is followed by the effective toc2vec model, with a f1-score of 0.60. The COUNT, PERIOD, and VERIETAS entities are consistently predicted incorrectly by the Toc2Vec model, which is unable to forecast numeric entities well. In addition, the Toc2Vec Model's better efficiency optimization directly relates the size of the model to the speed of word prediction per second, and the model is simple to integrate into a FaaS- and BaaS-based Serverless. The capabilities of Serverless M have been successfully tested using the black box method.Named Entity Recognition(NER) merupakan bagian dari pengembangan perangkuman dokumen, klasifikasi dan pencarian informasi. Implementasi NER pada domain pertanian, khususnya teks instruksional atau transkripsi video tutorial, akan memudahkan masyarakat umum memahami konsep dan istilah khusus dari kegiatan pertanian kota seperti proses dan prosedur produksi tanaman, metode dan alat pertanian, siklus panen, dan penanganan hama atau penyakit tanaman. Spacy merupakan alat bantu NLP, memiliki dua metode pengembangan model NER, yaitu dengan Toc2Vec dan Transformer. Kedua metode memiliki kelebihan dan kekurangan, yaitu ukuran, performansi dan kecepatan prediksi yang berbeda beda sesuai kebutuhan. Model NER dapat diimplementasikan menjadi aplikasi Serverless, dengan menggunakan pendekatan Fungsional as Services (FaaS) dan Backend as Services (BaaS). Pada penelitian ini telah dikembangkan tiga model NER untuk data teks instruksional pertanian sub topik budidaya tanaman buah. Model berbasis Toc2Vec dengan optimasi efisiensi, Model Toc2Vec dengan optimasi akurasi dan Model berbasis IndoBERT. Model berbasis Transformer memiliki nilai f1-score terbaik sebesar 0.71 disusul Model Toc2Vec Efisiensi sebesar 0.60 dan Model Toc2Vec Efektif dan 0.57. Model Toc2Vec tidak dapat memprediksi entitas numerik dengan baik, Prediksi entitas COUNT, PERIOD dan VERIETAS selalu tertukar. Selain itu Ukuran model berbanding lurus dengan kecepatan prediksi kata per detik, dalam hal ini Model Toc2Vec optimasi efisiensi unggul, model tersebut mudah diimplementasikan menjadi Serverless berbasis FaaS dan BaaS. Fungsionalitas dari Serverless ML telah berhasil diuji menggunakan metode Blackbox
Implementasi Algoritma $P Point Cloud Recognizer pada Pengenalan Angka Berbasis Game
Get PDFMore devices use gestures as their input method. Recognizing these gestures becomes more important for app development. One of the methods used for gesture recognition is Point Cloud Recognizer or P in games to show that P is implemented with the help of the game engine Unity with C# programming language. 3 sets of numerals 1 to 10 are used as data with P is able to recognize the gesture well.Semakin banyak perangkat memiliki masukan berupa gestur. Mengenali gestur tersebut menjadi semakin dibutuhkan untuk mengembangkan aplikasi. Salah satu metode yang digunakan dalam mengenal gestur adalah Point Cloud Recognizer atau P pada game untuk menunjukkan P diimplementasi dengan menggunakan game engine Unity dan bahasa pemrograman C#. Data yang digunakan ada 3 set numeral 1 sampai 10 dengan konfigurasi jumlah point pada P dapat mengenali gestur tulisan dengan baik
Segmentasi dan Pembentukan Model Regresi Nasabah Berbasis Analisis Recency, Frequency dan Monetary
Get PDFDuring this pandemic, the number of customers of a securities company has increased quite high. This requires securities companies to conduct analysis related to security customer data against transaction data so that the company can find out the segmentation of registered customers and also so that companies can predict the transaction patterns of customers in the company. In processing transaction data, the RFM (Recency, Frequency, Monetary) model can be used as a way to group customers according to their business values. After doing the modeling using RFM, the data is clustered using the K-Means algorithm to find out the segmentation in the RFM model in each group. The RFM model that has been clustered will produce segments based on the RFM group. In this data, a linear regression analysis process is carried out where each group and segment is analyzed and predicted related to variables such as recency, monetary and frequency. The results of data grouping, customer segmentation and also predictions with linear regression can be one of the company's references to make a business decision. From the linear regression process carried out on the RFM attributes, a prediction of the monetary value of the existing recency value is generated and the monetary value of the frequency can also be known with a fairly good error rate.Di masa pandemi ini, peningkatan jumlah nasabah suatu perusahaan sekuritas meningkat cukup tinggi. Hal tersebut mengharuskan perusahaan sekuritas untuk melakukan analisis terkait data nasabah sekuritas terhadap data transaksi agar perusahaan tersebut dapat mengetahui segmentasi dari nasabah yang sudah terdaftar dan juga agar perusahaan dapat memprediksi terkait pola transaksi dari nasabah yang ada di perusahaan tersebut. Dalam melakukan pengolahan data transaksi dapat menggunakan model RFM (Recency, Frequency, Monetary) sebagai cara untuk pengelompokan nasabah sesuai dengan nilai bisnis yang dimiliki. Setelah melakukan pemodelan dengan menggunakan RFM , data yang tersebut dilakukan Clustering dengan menggunakan algoritma K-Means untuk mengetahui segmentasi yang ada dalam model RFM dalam setiap kelompoknya. Model RFM yang sudah dilakukan clustering akan menghasilkan suatu segmen-segmen berdasarkan kelompok RFM. Pada data tersebut dilakukan proses analisis regresi linear dimana pada setiap kelompok dan segmen dianalisis dan diprediksi terkait variabel-variabel seperti recency, monetary dan frequency. Hasil dari pengelompokan data, segmentasi pelanggan dan juga prediksi dengan regresi linear ini dapat menjadi salah satu acuan perusahaan untuk membuat suatu keputusan bisnis. Dari proses regresi linier yang dilakukan atas atribut-atribut RFM, dihasilkan suatu prediksi nilai moneter dari nilai recency yang sudah ada dan juga dapat diketahui nilai moneter dari frequency dengan tingkat error yang cukup baik
Sistem Rekomendasi Suku Cadang Berdasarkan Item Based Filtering
Get PDFReplaceable spare part on workshop have many transaction and possibility thus recommender system is needed to simplify the selection process. We propose recommender system with item collaborative filtering, with high data sparsity. With Single Value Decomposition we reduce the matriks to improve the system and decrease “noise” value. Model will be evaluated using MAE, RMSE, and FCP metrics. The results of recommendation model are MAE = 1.2752, RMSE = 1.4882, dan FCP = 0.4947
Evaluasi Tata Kelola Guna Meningkatkan Kinerja Manajemen Teknologi Informasi Menggunakan Framework COBIT 5
Get PDFThe Salatiga Class I District Court is a government agency that has implemented Information Technology (IT) governance which has problems in IT management, so the purpose of this study is to evaluate information technology governance using the COBIT 5 framework by determining the level of capability and gaps. The results of the recapitulation of the questionnaire with the Likert scale model obtained the average value of the selected domain getting a value of 3.28, which means that the current level of capability (as is) is at level 3 (established process), while the numbers from the range of 0 to 5 levels are expected or target (to be) namely 5 (optimization process). So that a gap or gap is obtained, namely the value of 1.27 which results in a reduction between to be and as is. The selected process domain is in the highest position, namely DSS01 with a value of 3.80 or level 4 (predictable process) while the lowest position is APO12 with a value of 2.82 or level 3 (established process). To achieve the expected conditions, the IT subdivision can improve in dealing with IT risks and IT problems that occur in the Salatiga District Court Class I B. In addition to evaluating IT governance, testing system functionality is also carried out by using the Blackbox Testing method with the Boundary Value Analysis technique. The object is used for testing the system, namely e-Court tested on the registration page and login page. From the results of system testing e-Court On the registration page, there is a discrepancy in the expected name field and on the login page that the test results field - field as expected.
Keywords — Information Technology Governance; COBIT 5; e – Court; Blackbox Testing; Boundary Value Analysis
Pembangunan Dashboard untuk Mendukung Analisis Kartu Rencana Studi dan Kartu Hasil Studi Mahasiswa
Get PDFAt the beginning of every semester, students register for their classes resulting in KRS (student study plan) and at the end of the semester, they will receive KHS (academic transcript). KRS and KHS are recorded into the university’s system to be used for different kinds of analysis including strategic planning and accreditation. The purpose of this research is to create a dashboard to visualize data from KRS and KHS based on several Key Indicator Performance (KPI) using goal-directed design method. Performance metrics method is used to evaluate the user interface of the dashboard with 71% binary success. User Experience Questionnaire (UEQ) is also used to evaluate user experience showing scores of 1.96, 1.93, 1.77, 1.88 and 1.86 for attraction, stimulation, novelty, clarity and efficiency, respectively. These scores can be translated using the Comparison to Benchmark Scale table which shows good for both clarity and efficiency, and excellent for attraction, stimulation and novelty
Risk Management Analysis Using COBIT 4.1 at Vehicle Testing Management Information System
Get PDFThe role of information technology in an organization is growing so fast. Vehicle Testing of Management Information System (VTMIS) is a system that uses information technology to serve users in motor vehicle administration in the Banyumas transportation service area. VTMIS at the Banyumas Transportation Department is an integrated information system for the motor vehicle testing process starting from the registration process, levy payment, and vehicle testing. The purpose of this study is to analyze the risk management at VTMIS in Banyumas Transportation Department using Control Objective for Information and Related Technology (COBIT 4.1) domain Plan and Organize (PO) 9. COBIT 4.1 is a framework for analyzing and ensuring information technology aligns with business management by calculating maturity levels. The data analysis results from the PO9 domain show that VTMIS risk management has a maturity level of 3.46. The maturity level of VTMIS at the Banyumas Transportation Department is at level 3, namely defined, meaning that procedures are in a position of standardization, documentation, and individual communication. The recommendation for VTMIS at the Banyumas Transportation Department needs to carry out risk management in a structured, massive and integrated manne