STMIK Pontianak Online Journals (Sekolah Tinggi Manajemen Informatika dan Komputer)
Not a member yet
653 research outputs found
Sort by
Penerapan Jaringan Syaraf Tiruan Backpropagation Dalam Sistem Pakar Diagnosa Virus TORCH
Virus TORCH is an acronym of the virus Toxoplasma, Rubella, Cytomegalo virus, and Herpes. TORCH virus causes many children born with disabilities. The purpose of this study is to generate and build an expert system software that can help the patient in making an early diagnosis of the TORCH virus, in order to reduce the risk of children born with disabilities caused by this virus. Method research design is a case study and interviews in finding out about the TORCH virus. This expert system making use of Visual Basic .NET in the making of the application. An expert system is paired with a forward chaining neural network Backpropagation as decision-making. This expert system featuring a large selection of symptoms that can be selected user, where each choice of symptoms will bring the user to the selection of the next symptom to get the final result. While the database used is MySQL. Design model used is UML (Unfield Modeling Language) and flowchart. Its design method is a prototype and testing software using black box testing systems have been built. The results of this study is to help the public to better understand and know TORCH suffered virus so that it can be treated promptly to avoid the children who are born with disabilities. Suggestions in the future is that this expert system can be onlin
Implementasi Load Balancing Web Server Menggunakan Apache di Ubuntu 16.04.
When many users access a website at the same time. So when it can make the web server to accommodate excessive loads because it can no longer accommodate the number of requests received. And that usually happens when a single web server or single. However, using the Round Robin load balancing method and also Haproxy as a load balancer can help regulate load sharing on each server. With testing done on 250 users, 500 users, 750 users, up to 1000 users and using a static web. So you can see the CPU usage that is used by the user load that is tested at the time after performing. The results of testing using haproxy run better than a single server and two servers. By using Apache benchmark in calculating Throughput and Time per request between a single server, two servers and running all servers using Haproxy. For testing the presentation on the CPU that is used on 250 users is 31.9% the lowest but there is an increase in testing 1000 users, reaching 54.5%. The test results on two servers in testing 250 users, 500 users, 750 users, and 1000 users produced 62.5% -68.7% results. And by testing on a single server RAM usage in testing 250 users, 500 users, 750 users, and 1000 users which reached about 91% -96%. With better test results using haproxy load balancing than using a single server or two servers. Because the use of a single server and two servers almost reached overload. Which means the time needed to handle the load is load balancing using haproxy better than a single server to more quickly handle requests, time per second and CPU usage
Implementation of the Max-Miner algorithm for product recommendations in Café Lo Aja
Café Lo Aja is a company that sells food and drinks. The business process carried out by this company is buying goods from suppliers and selling products to customers. Sales transaction data will produce heaps of data that get bigger and bigger, so that it can cause new problems. The purpose of this research is to make an application to process a sales transaction data so that it can produce information about consumer purchasing patterns which will be used to help the Cafe Lo Aja to make business decisions. This study uses Cafe Lo Aja sales transaction data in 2019 with the Data Mining Market Basket Analysis method and the Max-Miner Algorithm. This research produces data which is an association rule from a collection of sales transaction data. So that by knowing the pattern of purchasing these products, the cafe manager can predict future market needs that can take into account what stock items must be reproduced and what items should be reduced because of the small percentage of interested ones. By knowing the results of the association, the manager can adjust the layout of the product menu to be better because products that are often purchased will be placed close together. Suggested application development that can be done in further research is the system can be developed into a mobile application, so consumers can order products on their smartphone devices
Education Transformation in the Era of Digital Disruption 4.0
In the digital era of disruption 4.0, featured automatization and connectivitation movement of transformation on education and work competencies were not linear anymore, where every graduate student must master the skills of digital literature, technology, and human resource. The social community service seminar was followed by 66 candidates consist of students, teachers, and structural officials on SMK Negeri 2 Singkawang. The seminar was an open lecturer discussion, question, and answer methods. The seminar emphasized the necessity to build virtual classrooms for teachers and students to collaborate internally or between schools by exchanging information, knowledge, and learning together in increasing their competencies
Implementation of Apriori Algorithm for Determining Product Purchase Patterns
In this study, the implementation of Data Mining association method using Apriori algorithm to determine product purchase pattern. Data obtained from the sales transaction data in the Toko Jaya Putra Bumi Agung in the form of a purchase note that will then be implemented using Apriori algorithm. The data mining technique used is the association rule method to know the pattern between items one and other items using support and confidence. In this study of the calculation process with Apriori algorithm obtained minimum value of support 50% and minimum confidence value 70% then resulting tendency of products purchased by consumers ie if buying cooking oil then buy eggs with confidence 75%. If buying Sumendo coffee then buy sugar with confidence 77.8%. This research is expected to be helpful and useful for the owner of Toko Jaya Putra Bumi Agung to predict and analyze the combinations of what kinds of products are often purchased by consumers simultaneously. For further research need to be developed again by combining other data mining algorithms. Preferably the use of datasets that are used will be better to use larger datasets so that they can obtain higher accuracy values
Smart SIAKAD Android based using RecyclerView
The Academic Information System (SIAKAD) android-based is an information system that can facilitate students and prospective students in obtaining information about the University and academics through mobile devices. The stages of study used waterfall method consisting of analysis, design, coding, testing and maintenance. This application is made using the RecyclerView method in conducting the application development of academic information systems and conducting tests such as Whitebox, blackbox, test devices and UI respondents. The results from the test of the whitebox, the total number of Cyclomatic Complex city (cc) amounted to 11, the Region of 15 and the independent Path of 11. The trial of the blacblox got the result that the app is already able to provide information to your existing needs. The tests of testing whether the function and appearance of the app is feasible to use and for a device trial it is obtained the average app access time of 45 seconds and the visual application display 100%. The others trial is conduct with respondents as many as 23 respondents by testing the application has been running well with the results 52.2% agreed, 39.1% very agreed and 8.7% disagree. Further of development can be directed to the addition of other features that don't exist yet. In addition, it can also be developed for mobile users in addition to Android operating systems such as iOS
Perancangan Neur-O: Sistem Pakar Penyakit Saraf berbasis Ontologi
Penyakit saraf sangat berbahaya jika menyerang seseorang sehingga perlu adanya cara untuk menyelesaikan masalah ini. Banyak cara dalam menangulangi penyakit ini, salah satunya ialah membangun sistem pakar untuk mengedukasi masyarakat tentang bahaya penyakit ini. Sistem ini dirancang dengan metodologi waterfall dimana memiliki 4 fase utama sampai bisa menghasilkan produk. Fase waterfall ialah perancanaan, analisis, desain dan implementasi dan akhirnya sistem telah dibangun. Hasil dari penelitian ini adalah sistem pakar berbasis ontologi yang diberi nama Neur-O. Pada Neur-O ada beberapa kategori yaitu Nama Penyakit, Sinonim, Definisi, Epidermiologi, Gejela baik umum maupun dugaan, Penyebab umum dan pendukung (lain), Agen pembawa, Pemeriksaan utama dan lain, Terapi obat dan fisik serta Pencegahan. Evaluasi menggunakan consistency checking dan evaluasi pakar. Pada consistency checking, ontologi dievaluasi dengan menggunakan inference engine Pellet. Pada proses inferensi, ontologi yang dipakai tidak terdapat error dan memerlukan 5093 milidetik untuk inferensi dan Evaluasi pakar rata-rata nilai ada 4 dimana sistem sudah baik. Berdasarkan hasil yang didapat, diperlukan pengembangan lagi kedepan dimana menambahkan amnesis berupa keluhan pasien dan riwayat penyakit dan pada outputnya dikembangkan ada diagnose banding serta tingkat stadium penyakit
VELOSX: Projectile Motion Visualization Application with Augmented Reality
In learning projectile motion and its velocity, students tend to look up a plain two-dimensional image in a science book. While there’s some educational props, yet they usually a very tradional ones and can not be used for real calculation. The utilization of Augmented Reality (AR) in educational method may raise curiosity and gives a unique way in learning projectile motion as the motion can be seen in a three dimensional. Augmented Reality itself is a combination of real world and virtual objects. This application uses the Vuforia SDK that able to blend the real world and virtual objects. Through this application, we were able to simulate projectile motion and its velocity in more realistic way, have slightly interaction with the reality, and gets input from user so they can learn and see the result of the parameter that they entered. Thus, with the advantage of AR the application gives a more realistic feel compared to the existing ones available in public as it could receive any input and show the output in AR.
Implementation of Cloudflare Hosting for Access Speed on Trading Websites
Companies that are currently developing pretty much do marketing via the website. Website becomes something very important that must be owned by trading companies engaged in the Export and Import sector. Trading websites that are currently running have flaws in terms of speed of access. The process of cloudflare hosting implementation research for access speed on trading websites uses the waterfall method which has stages such as Requirement Analysis, System Design, Implementation, Testing, and Maintenance Operations. The results obtained from this research process in accordance with the waterfall method used is successfully implementing cloudflare hosting to increase the speed of website quality to reach 100%, can even reduce spam, save bandwidth and cut open time for website pages 50% which means higher speeds. Knowing the shortcomings of the website before implementation, through cellular access speed is only 13% while through the desktop 50% and knowing the details of numbers and speed units of the website both before and after implementation. The next research that can be done is to compare the performance of cloudflare with radware, Akamai Prolexic, and Neustar
PERANCANGAN PERANGKAT LUNAK ENKRIPSI SMS MENGGUNAKAN ALGORITMA RC6 DAN RIJNDAEL PADA SMARTPHONE
Sending text messages is done on a smartphone device is basically the text message delivery without doing a safeguard against the content of the messages sent, so that when it is done tapping against the Groove pengirimanya then intercepted text messages can be directly read by the tappers. For that it needs the software as a specific encoding method support so that the message sent is secure. This research uses the form of research studies, literature research method using Research & Development (R & D). As for the technique of data collection using the study documentation and observation to acquire the theory and algorithm rijndael rc6 algorithm. While the method of software design using Rapid Application Development (RAD). The design of software using the java programming language. The results of this design generates a software named "Sms Encryption". Encryption software sms using the rc6 algorithm and rijndael has been executed and in accordance with the expected user. With the software, the confidentiality and authenticity of the information in the form of a text message will be more awake