ejournal.nusamandiri.ac.id (STMIK Nusa Mandiri)
Not a member yet
1504 research outputs found
Sort by
IMPLEMENTATION OF ARAS ALGORITHM ON DECISION SUPPORT SYSTEM TO DETERMINE THE BEST LECTURER
Get PDFIn the world of higher education, lecturers are one of the main components in building quality and quantity. Good quality will give good results as well, to improve the quality of each lecturer, it is necessary to have an award given to lecturers from the campus so that it becomes a motivation for lecturers to improve the quality given to students and the community. Amik Mitra Gama is a private campus located in the Duri Riau area, in this case to improve the quality of education one of the steps taken is to give awards and appreciation to the best lecturers who will be selected every year. To realize this, we need an easy calculation system that is carried out in the form of ranking according to the final value, therefore a decision support system using the ARAS method is chosen because it is very appropriate for the selection process and provides convenience in the calculations which are determined based on ranking. The decision support system using the ARAS method uses 8 criteria that are set as a reference in determining the best lecturers, namely Recent Education, Lecturer Functional Position, Lecturer Certification, Number of Journal Publications, Roles in Research, Journal Publication History, Research Grants, and Community Service. There are 10 lecturers in the field of computers who will be used as alternative data with lecturer codes D01, D02, D03, D04, D05, D06, D07, D08, D09, D10. The results obtained from this study are the lecturer code D04 = 0.0974, D06 = 0.0965, D09 = 0.0932, D07 = 0.0903, D03 = 0.0901 was selected as the best lecturer in 2021/2022. So that the results of this study can help the campus to determine the best lecturers every year fairly and be selected based on rankings
OPTIMALISASI KEAMANAN WIDE AREA NETWORK (WAN) MENGGUNAKAN RAW FIREWALL BERBASIS MIKROTIK PADA PT. PERMATA GRAHA NUSANTARA
No full textFokus penelitian ini adalah optimalisasi keamanan jaringan dengan implementasi metode firewall dengan teknik raw firewall pada perangkat jaringan mikrotik. Firewall membatasi siapa saja yang berhak mengakses suatu internet dalam jaringan, dan siapa saja yang harus diizinkan dan tidak diizinkan untuk lewat, hal ini biasa disebut dengan filtering. Firewall pada jaringan, dapat memantau aktifitas suatu jaringan. Raw Firewall adalah teknik keamanan jaringan yang dalam penggunaannya tidak membutuhkan resource yang besar. Pada penelitian ini dilakukan dua skenario pengujian: (i) pengujian pertama dengan melakukan serangan ping attack sebelum implementasi teknik raw firewall, yaitu menggunakan filter rules, dan (ii) pengujian kedua melakukan serangan ping attack kembali setelah implementasi firewall raw. Hasil yang diperoleh dari penelitian memperlihatkan bahwa penggunaan resource cpu dengan teknik filter rules rata-rata sebesar 41% dan resource cpu setelah implementasi raw firewall rata-rata sebesar 2% saat terjadi serangan. Implementasi raw firewall terhadap ping attack berhasil menurunkan beban pada cpu, sehingga pada kondisi ini kinerja perangkat tidak terganggu
PENERAPAN METODE DESIGN THINKING PADA MODEL PERANCANGAN UI/UX PADA FITUR REPORT HELPDESK TICKETING SISTEM
Get PDFWijaya Karya (Persero) Tbk. Yaitu perusahaan yang bergerak di bidang infrastruktur. Sebagai perusahaan yang besar WIKA memiliki aplikasi untuk membantu proses bisnisnya. Aplikasi-aplikasi tersebut sebagian besar dikelola oleh departemen Sistem Informasi. Dari kekurangan aplikasi tersebut, tentunya akan ada keluhan-keluhan pengguna aplikasi yang telah dibuat. Proses pekerjaan yang dilakukan saat ini masih belum optimal karena belum adanya report ticket yang memudahkan Agen dan Manajer untuk melihat seluruh data tiket dan tampilan dari web helpdesk ticketing system yang kurang menarik sehingga penulis ingin mengubah tampilannya. Berdasarkan masalah di atas, maka diperlukan perancangan sistem pada aplikasi Helpdesk ticketing system. Perancangan ini menggunakan metode design thinking, yang terdiri dari tahapan empathize, define, ide, prototipe dan tes. Sehingga hasil dari perancangan ini memberikan rekomendasi berupa model UI/UX pada aplikasi Helpdesk ticketing system.
 
CLASSIFICATION OF BLOOD DONOR DATA USING C4.5 AND K-NEAREST NEIGHBOR METHODS (CASE STUDY: UTD PMI BALI PROVINCE)
No full textClassification of blood donor data at UTD PMI Bali Province by applying the C4.5 and K-Nearest Neighbor algorithms. The number of blood donor data donors is 34,948, of which 90% of the data, namely 31,454 is used as training data. Meanwhile, 10% of the data, which is 3,494 data, is used as the implementation of data testing using the Orange application. C4.5 obtained an accuracy score of 92.9%, F1 of 92.2%, Precision of 93.1%, Recall of 92.9%, specificity of 68.2%. While K-nearest neighbor obtained an accuracy score of 91%, F1 of 90.1%, Precision of 90.8%, Recall of 91%, specificity of 63%. With the AUC (Area Under Curve) value for the C4.5 algorithm is 0.875 and the K-nearest neighbor is 0.813 Good Classification. The results of the evaluation using the confusion matrix C4.5 obtained an accuracy score of 92.6%, F1 of 95.7%, Precision of 99.4%, Recall of 92.4%, specificity of 96%. While k-nearest neighbor obtained an accuracy score of 90.9%, F1 of 94.6%, Precision of 98.4%, Recall of 91.2%, specificity of 88.4%. Based on the evaluation of the confusion matrix and the ROC Analysis Graph, the C4.5 algorithm obtained higher results than the K-Nearest Neighbor algorithm. Based on the data on the characteristics of blood donors at UTD PMI Bali Province, it shows that the gender is male, Badung area, Age 20 to < 30, the occupation of private employees dominates in blood donation
DESIGN AND DEVELOPMENT OF WEB-BASED INFORMATION SYSTEM FOR OFFICE STATIONERY PROCUREMENT MANAGEMENT
Get PDFOffice Stationery (ATK) is an item used to do written work such as paper, books, ink, pencils, pens, paper clamps, and others. ATK is a supporting tool that has an important role in implementing the administrative function of a company or agency. The management of ATK procurement in the Bogor Agricultural Institute (IPB) is under the coordination of the Directorate of Infrastructure, Facilities, and Campus Environmental Security (DPSPLK) as a coordination unit at the central level. So far, the procurement of ATK in every work unit in the IPB environment uses a direct purchase method. Direct purchasing methods cause problems and weaknesses, including: (1) price differences between providers for the same goods, (2) the quality of providers varies because each work unit directly selects providers, (3) vulnerability to price manipulation, (4) needed storage space for goods in each unit, (5) management and organization of ATK data are carried out partially in each work unit. Atk data management and organization have not been systemized, making it difficult for DPSPLK to reconcile data. An umbrella contract procurement method (framework contract) is applied to deal with problems in the direct purchase method. Management and organization of unsystematic ATK data, an integrated information system is built to facilitate DPSPLK reconciling data for reporting needs. The information system development method uses the waterfall method and testing using the User Acceptance Test (UAT) black-box type. This research resulted in the procurement and management application of ATK used by DPSPL
COMPARING ALGORITHM FOR SENTIMENT ANALYSIS IN HEALTHCARE AND SOCIAL SECURITY AGENCY (BPJS KESEHATAN)
Get PDFTwitter is a social media that can be used to express opinions and exchange information quickly with individuals and institutions such as the Healthcare and Social Security Agency (BPJS Kesehatan). Every word that a Twitter user utters has meaning and stellar emotion. This meaning can be reached through the process of sentiment analysis. Sentiment analysis is the process of understanding and classifying emotions such as positive or negative or complaining or not complaining. This study classifies tweet data related to BPJS Health services into two classifications, namely complain and no complain. Using 1,000 data from Twitter written on the BPJS Kesehatan Twitter account. In text mining, to build a classification, the transform case, tokenize, token filter by length, stemming and stopword techniques are used. Gataframework is used to assist the preprocessing and cleansing process. Rapidminer was used to create sentiment analysis in comparing three different classification methods of the Twitter data. The method used is the Nave Bayes algorithm and the Naïve Bayes algorithm with the addition of a Synthetic Minority Over-sampling Technique (SMOTE) feature and the Naïve Bayes algorithm with an SMOTE feature that is optimized with Adaboost. The Naïve Bayes algorithm is added with the SMOTE feature which is optimized with Adaboost to get the best value with an accuracy value of 69.11%, precision 69.93%, recall 68.89% and AUC 0.77
DESIGN AND IMPLEMENTATION OF INVENTORY INFORMATION SYSTEM IN PUTRA MARIYO TRADING BUSINESS
Get PDFTechnological advancement in information technology applications allows the data recording process to become easier. Putra Mariyo Trading Business is a business institution that sells wood as a building material in various types and sizes. Data processing of incoming and outgoing goods in this business is performed only by writing goods data into a specific book. Consequently, some problems such as the loss of data and miss calculation in processing transactions usually occur. Therefore, in this research, we design a system to address the problems. The system is developed by following the Waterfall software development method. To build the system, we opt to use PHP programming language, CodeIgniter framework, and MySQL as the database server. To determine system reliability, this information system testing uses black box testing which focuses on the functional requirements of the system. To evaluate the performance of the system two testing steps i.e., black-box testing and System Usability Scales (SUS) are adopted. Black-box testing results show that the error percentage of our system is 0%. The SUS testing is conducted to obtain responses from users and the SUS score obtained is 70.1 indicating that the system is at the "good" level and reliable to use
SISTEM INFORMASI PENJUALAN PETI MATI BERBASIS WEB DI CV. GEOJAYA NUSANTARA: WEB-BASED Coffin SALES INFORMATION SYSTEM IN CV. GEOJAYA NUSANTARA
Get PDFAbstract—Sales is an activity that connects customers with sellers through a product or service offered to produce something that is mutually beneficial for both parties. So for that we need a system so that sales transactions can run effectively and efficiently, for example by using a sales information system. CV. Geojaya Nusantara is a company engaged in the sale of coffins. The existing sales system in CV. Currently, Geojaya Nusantara is still carried out conventionally, such as customers who want to buy their products via phone calls or also through messaging applications such as the WhatsApp application. Then the incoming sales data is still done manually, as recorded in a sales book. Therefore, with these problems, a web-based application is needed to support the sales system which aims to simplify all transaction processes, because it can be done anytime and anywhere. The research method used is by conducting direct observations and interviews with employees or shop owners. As for the software development method used is the waterfall method and the programming language used in making the website itself is PHP, HTML, CSS, and JQuery, and for the database it uses MySQL. The expected results in this web-based coffin sales information system at CV Geojaya Nusantara, are expected to be a solution to overcome the problems encountered due to the conventional system, and can also overcome errors in the sales recording input process.
Keywords: coffin sales, information system, web-base
IMPLEMENTATION OF PROFILE MATCHING METHOD FOR THE BEST EMPLOYEE SELECTION SYSTEM PT. JENDELA DIGITAL INDONESIA
Get PDFThe implementation of profile matching method in selecting the best employees at PT Jendela Digital Indonesia aims to assist managers in making decisions with the right calculations and criteria. Employees are one of the factors that play an important role in advancing the company. Employee performance affects the company in obtaining profits. To spur employee performance, the company selects the best employees every period by giving appreciation and bonuses to selected employees.
This selection system using three criteria, namely aspects of cooperation, work performance, and personality. These criteria will be used for calculations using the Profile Matching method. There are five employees who will be submitted to the selection of the best employees in this company. All criteria are given a GAP value and then will provide a ranking. The largest final score will be at the top of the ranking, followed by a smaller final score. The results of this research show that this method can provide results that assist managers in making decisions about the best employees according to the criteria desired by the company.
 
PREDIKSI HARGA SAHAM TWITTER DENGAN LONG SHORT-TERM MEMORY RECURRENT NEURAL NETWORK
Get PDFAbstract— Today the trading business has become a trend to get money easily without having to work hard as long as you have capital. To get maximum results and avoid losses, it is necessary to have expertise in predicting the ups and downs of the stock market value. The purpose of this research is to utilize machine learning technology to predict the fluctuation of stock value by using the Long Short-Term Memory RNN model. From the results of this study, it was found that LSTM+RNN is suitable for use in single-step models.
Keywords: stock price, machine learning, recurrent neural network, lstm
Abstrak—Dewasa ini bisnis trading menjadi suatu trend untuk mendapatkan uang dengan mudah tanpa harus bekerja keras asalkan memiliki modal. Untuk mendapatkah hasil yang maksimal dan menghindari kerugian maka diperlukan keahlian di dalam memprediksi naik turunya nilai bursa saham. Tujuan dari penelitian ini adalah memanfaatkan teknologi machine learning untuk memprediksi naik turunya nilai saham dengan menggunakan model Long Short-Term Memory RNN. Dari hasil penelitian ini didapatkan bahwa LSTM+RNN cocok untuk digunakan pada model single-step.
Kata kunci: harga saham, machine learning, recurrent neural network, lst