Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) 2020-04-22T15:32:11+07:00 Yuhefizar Open Journal Systems <p><span id="result_box" class="" lang="id"><strong><span class="">Jurnal </span>RESTI&nbsp; (Rekayasa Sistem dan Teknologi Informasi)</strong> adalah sebuah jurnal <em>blind peer-review</em> <span class="">yang didedikasikan</span> untuk publikasi hasil penelitian yang berkualitas dalam&nbsp; bidang Rekayasa Sistem dan Teknologi Informasi namun tak terbatas secara implisit dengan </span>e-ISSN: <a href="">2580-0760</a>) dan telah terakreditasi <strong>SINTA 2</strong> <strong>sesuai dengan keputusan Dirjen Penguatan Riset dan Pengembangan Kemenristekdikti No. </strong><strong><a href="">10/E/KPT/2019.</a> </strong></p> <div class="itanywhere-activator" style="left: 0px; top: 45px; display: none;" title="Google Translator Anywhere">&nbsp;</div> Optimization of K Value in KNN Algorithm for Spam and Ham Email Classification 2020-04-22T15:32:11+07:00 Eko Laksono Achmad Basuki Fitra Bachtiar <p><em>There are many cases of email abuse that have the potential to harm others. This email abuse is commonly known as spam, which contains advertisements, phishing scams, and even malware. This study </em><em>purpose to know </em><em>the classification of email spam with ham using the KNN method as an effort to reduce the amount of spam. KNN can classify spam or ham in an email by checking it using a different K value approach. The results of the classification evaluation using confusion matrix resulted in the KNN method with a value of K = 1 having the highest accuracy value of 91.4%. From the results of the study, it is known that the optimization of the K value in KNN using frequency distribution clustering can produce high accuracy of 100%, while k-means clustering produces an accuracy of 99%. So based on the results of the existing accuracy values, the frequency distribution clustering and k-means clustering can be used to optimize the K-optimal value of the KNN in the classification of existing spam emails.</em></p> 2020-04-20T22:14:44+07:00 Copyright (c) 2020 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Determining Effective Subjects Online Learning (Study and Examination) with Multi-Attribute Utility Theory (MAUT) Method 2020-04-22T15:31:15+07:00 Tonni Limbong Janner Simarmata <p><em>The Corona pandemic in Indonesia forced the learning system into a drastic change into online learning. Many campuses that were previously quite comfortable with face-to-face learning were forced to be helpless because they had never prepared a backup plan when something unexpected happened, one of which was when the campus was forced to not be able to face-to-face. If the learning system continues to be done online, it might cause the quality of learners to decrease dramatically compared to face-to-face learning. Moreover, the courses that must be filtered include theories and practice courses. Most assignments given to students are not seriously done because of a lack of significant evaluation and supervision. The Faculty of Computer Science, Santo Thomas Catholic University, Medan supports this government policy by implementing online learning using the Zoom application for face-to-face and Edmodo for supplementary lecture material&nbsp; for online learning media. In one semester apart from studying there is a test period which is the Midterm Examination (UTS) and Final Examination Semester (UAS) where the current leadership must wisely determine the type and nature of the exams to be conducted online. To find out the effectiveness of the form of exam questions in online implementation, a Multi-Attribute Utility Theory (MAUT) method was conducted, the test results found that online learning with Zoom and Edmodo was very effective for theoretical courses with a value of 0.88, then the results of this calculation it is recommended that online tests be conducted in the form of theories such as multiple-choice, essay and analysis</em></p> 2020-04-20T21:46:54+07:00 Copyright (c) 2020 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Algorithm Comparation of Naive Bayes and Support Vector Machine based on Particle Swarm Optimization in Sentiment Analysis of Freight Forwarding Services 2020-04-22T15:19:48+07:00 Sharazita Dyah Anggita Ikmah <p><em>The needs of the community for freight forwarding are now starting to increase with the marketplace. User opinion about freight forwarding services is currently carried out by the public through many things one of them is social media Twitter. By sentiment analysis, the tendency of an opinion will be able to be seen whether it has a positive or negative tendency. The methods that can be applied to sentiment analysis are the Naive Bayes Algorithm and Support Vector Machine (SVM).</em><em> This research will implement the two algorithms that are optimized using the PSO algorithms in sentiment analysis. </em><em>Testing </em><em>will be done</em><em> by setting parameters on the PSO in each classifier algorithm.</em> <em>The results of the research that have been done can produce an increase in the accreditation of 15.11% on the optimization of the PSO-based Naive Bayes algorithm. </em><em>Improved accuracy on the PSO-based SVM algorithm worth 1.74% in the sigmoid kernel.</em></p> 2020-04-20T21:21:14+07:00 Copyright (c) 2020 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Nebengin: Android-based Collaborative Transportation Application 2020-04-22T15:18:05+07:00 I Made Sukarsa I Kadek Teo Prayoga Kartika I Putu Arya Dharmadi <p><em>Online transportation services are transportation services that take advantage of advances in information technology. In Indonesia, several online transportation service providers have grown, such as Gojek, Grab and several other startup startups. In addition to the many benefits and conveniences that have been provided, the service price factor is still felt expensive for the community, especially the lower classes. One solution is to build a Collaborative Transportation Application that utilizes the Collaborative Transportation Management (CTM) interaction method. The main features contained in the application, namely the feature of finding driver routes in the direction of the consumer route, online chat, and route management by utilizing the Android mobile application based on Google Map API and Firebase Cloud Messaging. Performance testing using JMeter from a total of 300 virtual users performs 6 HTTP requests resulting in an average response time of 0.852 alias tolerated according to Ap</em><em>d</em><em>ex standards. Based on the results of testing of 15 respondents obtained the results that this application is easy to use, supports traveling activities, fast response time when used, features provided are quite complete, an accurate position tracking system, unidirectional route search is very compatible with consumer routes, and the process of coordination and driver collaboration is very good so that the travel costs borne by consumers become more economical..</em></p> 2020-04-20T20:53:59+07:00 Copyright (c) 2020 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Comparison of SVM, RF and SGD Methods for Determination of Programmer's Performance Classification Model in Social Media Activities 2020-04-22T15:29:37+07:00 Rusydi Umar Imam Riadi Purwono <p><em>The failure of most startups in Indonesia is caused by team performance that is not solid and competent. Programmers are an integral profession in a startup team. The development of social media can be used as a strategic tool for recruiting the best programmer candidates in a company. This strategic tool is in the form of an automatic classification system of social media posting from prospective programmers. The classification results are expected to be able to predict the performance patterns of each candidate with a predicate of good or bad performance. The classification method with the best accuracy needs to be chosen in order to get an effective strategic tool so that a comparison of several methods is needed. This study compares classification methods including the Support Vector Machines (SVM) algorithm, Random Forest (RF) and Stochastic Gradient Descent (SGD). The classification results show the percentage of accuracy with k = 10 cross validation for the SVM algorithm reaches 81.3%, RF at 74.4%, and SGD at 80.1% so that the SVM method is chosen as a model of programmer performance classification on social media activities.</em></p> 2020-04-20T18:05:10+07:00 Copyright (c) 2020 Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)