J-Icon : Jurnal Komputer dan Informatika

PREDICTION OF BLOOD DEMAND AND SUPPLY: DOUBLE EXPONENTIAL SMOOTHING METHOD APPROACH

2024-02-12T13:09:03+00:00

There is a daily blood transfusion in the hospital. Blood management and distribution systems have been integrated into each UTD-PMI. Different systems were formed to maintain a balance between the demand and supply of blood bags. UTD-PMI Kota Kupang serves blood requests in 12 hospitals and 1 clinic located in the city and Kupang district. The demand for more blood than the supply of blood in the UTD-PMI is a personal problem. The prediction of demand and the blood supply becomes important so that there is no difference in demand and supply of blood in the very large UTD-PMI office. The study used the Double Exponential Smoothing (DES) method to predict blood supply and demand. Data from the last four years is used to predict the amount of blood demand and supply. Predictions with the DES method are divided into two categories. The first category uses the data of the last 3 years to predict data in the fourth year for the next 6 months. The training data used is divided into two types: the last 3 years and the last 2 years. Forecast results compared to real data in 2023. Category 2 uses data from the last four years to predict data for 2023. The training data used is divided into 4 types, each starting from the last 4, 3, 2, and 1 year. To determine the accuracy of the calculation using the mean absolute percentage error (MAPE). The results of determining training data with different ranges result in predictions of blood demand and blood supply, whose results are also different but still in the “good forecast” category.

DESIGN OF JOB VACANCY INFORMATION SYSTEM APPLICATION AS A WEB AND MOBILE-BASED MEDIA PARTNER

2024-02-12T13:13:43+00:00

The increasing number of job seekers in Indonesia has made job vacancy information applications the most accessed application among job seekers, especially for students graduating from vocational schools. Vocational High Schools (SMK) are believed to be schools that are prepared to produce workers who have the skills and expertise needed in the world of work. However, most of the existing job vacancy applications are very limited in presenting information on job vacancies that match the expertise and experience of alumni and even students of SMK Ma’arif 1 Yogyakarta. For this reason, in helping graduates of SMK Ma’arif 1 Yogyakarta, a forum is needed that can facilitate these graduates in obtaining information on job vacancies that are by the skills and location of the alumni’s residence area. The purpose of this project is to develop a system that will assist students at SMK Ma’arif 1 Yogyakarta in locating information about employment openings. This job vacancy application was created using the prototype method because it is very suitable for building Android-based job vacancy information system software. The system design made by the author was successfully implemented into a mobile application using the Kotlin coding language and Android Studio. Based on testing, the application built can change the information system that was originally delivered manually into an electronic system by developing mobile applications so that there is interaction between system developers and application users that can be used to get job vacancy information easily.

APPLICATION OF THE LSTM (LONG SHORT TERM MEMORY) METHOD IN SENTIMENT ANALYSIS TOWARDS THE IMPLEMENTATION OF THE ELECTION DURING THE COVID-19 PANDEMIC

2024-02-21T13:35:00+00:00

The implementation of regional elections in the midst of the Coronavirus disease 2019 (COVID-19) pandemic has become the spotlight and discussion of the wider community. This is because it involves many people so that it becomes one of the risks of transmission of the virus. Therefore, this study analyzes how public sentiment towards the elections amid the COVID-19 Pandemic. One method used in sentiment analysis is Long Short-Term Memory (LSTM). LSTM is a development of the Recurrent Neural Network (RNN) model that can replace problematic RNN nodes in the hidden layer with LSTM cells designed to store previous information. The data used in this research is Indonesian tweet data taken from Twitter with a total of 5053 data classified into positive and negative sentiment classes. The data from the tweets is preprocessed first to make it easy to classify, then selected and transformed to be processed in data mining using LSTM. The LSTM classification was evaluated using the Confusion matrix to obtain an accuracy of 79,78%, a precision of 69% and a recall of 64%. Tests were also carried out with different epochs and dropouts, it was found that the average accuracy value was more than 80%, and the greatest accuracy was obtained with epoch of 45 and dropout of 20.

PUBLIC STREET LIGHTING DAMAGE COMPLAINT SERVICE USING QR CODE AND AUTOMATED CHAT FOR THE TRANSPORTATION DEPARTMENT

2024-02-21T13:50:47+00:00

This research was motivated by observation data which showed results regarding difficulties in receiving complaints from the public, especially regarding damage to Public Street Lighting (PJU). The method used was direct observation at the research site using the SDLC (System Development Life Cycle) model with 6 stages of the SDLC model. that is; Planning, System Analysis, System Design, Implementation, Testing, Validation and Maintenance. In its development, this research used the Google spreadsheet application to create worksheets electronically and Gmail to manage messages electronically. This service has several benefits, including helping in inputting data for reports or public complaints, minimizing the risk of data loss and saving time, as well as making it easier to access information quickly. The results of the research obtained are to create a PJU damage complaint service using QR codes and automatic chatbots to simplify the complaint process and as a medium for complaints as a means of information and cooperation with the public if there is damage to traffic facilities, especially Public Street Lighting (PJU). Trials were carried out at 10 (ten) pole points in different regions, 5 poles in Jasinga and 5 poles in the Jonggol area, their function was to find out whether this complaint service was running well or not. The results of trials carried out in 2 different areas of this complaint service can function well.

APPLICATION OF DATA MINING IN UNDERSTANDING MEDICINE PURCHASE TRANSACTION PATTERNS USING THE APRIORI ALGORITHM AT THE KHARISMA FARMA TIGA PHARMACY

2024-02-21T13:59:57+00:00

Drug storage at the Kharisma Farma Tiga Pharmacy is grouped based on class and form of availability, for categories namely patented, external, over-the-counter, herbal, generic and syrup drugs while for forms of availability, namely drugs in the form of bottles, syrups, creams, capsules, ointments, injection fluids, solutions, roll on and drip. Grouping of drugs based on class and availability in alphabetical order. However, there is a problem when several customers come to the pharmacy at the same time to buy various types of medicines based on different classes or availability of medicines, resulting in a lack of time efficiency in service, pharmacy staff need additional time to search for medicines in their storage place. Seeing that the storage of medicines at the Kharisma Farma Tiga pharmacy is still sorted alphabetically based on class and availability, we need an application that can help recommend arranging the layout of medicine storage by reading customer drug purchase transaction patterns at the Kharisma Farma Tiga Pharmacy so that the storage process medicines according to medicines that are often purchased simultaneously and can increase time efficiency in better customer service, especially in searching for medicines. The application of data mining to determine drug purchasing patterns using the Apriori algorithm at the Kharisma Farma Tiga Pharmacy was built on a web basis. The results obtained in this research are that the application of an a priori algorithm to determine the pattern of drug purchase transactions at the Kharisma Farma Tiga pharmacy has been successfully carried out and can provide recommendations for the layout of drugs that are often purchased based on drug purchases, as for the association rules obtained with a minimum support of 20% and confidence. 75%, namely 2 item sets with 4 rules and 3 item sets with 7 rules. The highest confidence value in the 2 item set is if the consumer buys 60 ml GPU oil then buys 10 ml of capax wind oil, while in the 3 item set is if the consumer buys redoxon effervescent 20s, 60 ml GPU oil then buys 10 ml of capax wind oil.

2D ANIMATION DESIGN AS ONE OF MEDIA USE ON INTRODUCING MORAL OF LIFE THROUGH FICTION STORY

2024-03-04T03:32:04+00:00

Animation 2D is one of the media that has been used for several purposes during this period. Animation 2D is use as study media in most of the schools, but the template is always the same which is similar to Powerpoint slide-show to deliver information or knowledge, especially moral and character-building values. This method that’s usually often use may give impression to viewers that Animation 2D is monotonous. Animation 2D should be delivered in a more different kind of ways in order to attract and interest the viewers, like fiction story. Many knowledges or information, especially moral and character-building values can be delivered through story, because the story itself may entertain and increase the interest of the viewers to watch and it may become one of the entertaining media for viewers. Method used on the research is Multimedia Development Life Cycle (MDLC) which consists of 6 steps such as concept, design, material collecting, assembly, testing, and distribution. The result is the animation 2D fiction story get most people interests, especially teenagers in the age range around 14-18 years old and can deliver moral message implicitly. Animation 2D media can still be further developed into films with detailed story so that it can leave deep impression dan can be understood more by audiences, especially teenagers. By increasing the audience’s understanding of the 2D Animation content story, it may indirectly leave moral message that is quite impressive for the teenagers.

SENTIMENT ANALYSIS OF THE USE OF THE MERDEKA MENGAJAR APPLICATION ON THE GOOGLE PLAY STORE

2024-03-05T13:19:40+00:00

To provide a solution to the learning loss that has occurred in the education sector in Indonesia since the COVID-19 pandemic, the Ministry of Education, Culture, Research, and Technology (Kemendikbudristek) released the "Merdeka" curriculum. To assist teachers in obtaining references, inspiration, and understanding of the "Merdeka" curriculum, Kemendikbudristek launched the "Merdeka Mengajar" platform, which can be downloaded from the Google Play Store. However, the utilization of this application has not yet reached the target number of users expected. To determine the achieved number of users, the analysis process stages are carried out, namely data collection, data pre-processing (pre-processing), data labeling (labeling), word extraction, classification, classification evaluation, sentiment analysis using the naïve Bayes model, and Support Vector Machine (SVM). The research results, using n-gram implementation with the naive Bayes and SVM models, show that the accuracy level generated by each model is 86% and 91%, respectively. Sentiment analysis indicates that 3,225 (57%) user reviews are positive, while 2,421 (43%) are negative. Overall, it can be concluded that the sentiment regarding the use of the independent teaching application is positive. Meanwhile, several factors causing users to provide negative reviews include difficulties during activation, incomplete learning modules, and requests to release the application on IOS.

User Experience Analysis in South Sumatra Diary Application using Heuristic Evaluation Method

2024-04-01T12:43:21+00:00

The application is a mobile application that has great potential to help users record and manage their daily notes. However, a good user experience (UX) is critical in ensuring the success and acceptance of the application. This research aims to carry out UX analysis on the application using the Heuristic Evaluation method. The Heuristic Evaluation method is used to identify potential UX problems by comparing the application with a set of pre-established design principles (heuristics). The research results show a number of findings related to the navigation, consistency, responsiveness and accessibility aspects of the Diary Sumsel application. The results of this research found a low level of improvement analysis after tracing it according to the evaluation that had been carried out. Then UX from the results of this analysis is more than enough to make users feel comfortable in becoming a membership or user list of the Diary Sumsel application in terms of several features that can be utilized in it. This research makes an important contribution in understanding and improving the UX of the Diary Sumsel application, which in turn can increase the acceptability and value of this application for users. By improving the UX, it is hoped that this application can provide greater benefits and better meet user needs.

Analysis Of Interest in The Paylater Feature in the Shopee Application using the UTAUT 2 Method Among Teenagers

2024-04-01T12:54:30+00:00

Technological development is something that cannot be avoided, technology is used to live everyday life starting from studying, communicating, working and shopping. The technology that is currently developing in Indonesia is Financial Technology (Fintech), therefore currently there are many payment options ranging from using paper money to electronic payments. One of the fintech currently developing in Indonesia is paylater, which is a service similar to an online loan without a credit card that can allow consumers to make payments in one go or in installments. The development of E-commerce is currently increasingly rapid, bringing major changes to teenagers' lifestyles, this has an impact on users when checking out payment transactions, including other costs that are not transparent and the process of paying off bills is difficult before waiting for the payment due date each month. The aim of this research is to find out how interested the use of paylater is among teenagers, by adopting the Unified Theory of Acceptance and Use of Technology (UTAUT2) model. The research method used is a quantitative method using a survey method via an online questionnaire involving a total of 65 respondents. This research uses a Likert scale approach and sampling uses the Slovin formula calculation method. The data obtained was analyzed using the Partial Least Square (PLS) method using SmartPLS software. The results of the tests carried out indicate that the variable perceived trust in the UTAUT2 partially has a positive and significant effect on the decision to use paylater by teenagers. Then the hypotheses that were rejected were the Hedonic Motivation, Habit, and Price Value on the UTAUT2 variables.

Evaluation and Interface Redesign Using the User Centered Design (UCD) Method on the Kampus Kita Tendik Application

2024-04-01T13:05:03+00:00

The Kampus Kita Tendik application is a mobile-based application developed by Airlangga University to streamline and optimize the performance of Educational Personnel. From a usability perspective, the application exhibits several UI/UX issues, including inconsistent font placement and size, as well as a disorganized profile menu. Consequently, there is a need for research to redesign the application interface, aiming to deliver an intuitive user experience, ease of navigation, and enhanced functionality. The research methodology employed includes User Centered Design (UCD) for application design and System Usability Scale (SUS) for assessing application usability. Twenty respondents, comprising Educational Personnel, were engaged to evaluate both the initial design and the redesigned interface. The outcomes of this evaluation, conducted through UCD, demonstrate a marked improvement in application design quality. This is evidenced by the SUS score of 87.14, awarded an A+ grade, for the redesigned application—a significant increase from the initial SUS score of 65.57, which received a C grade.