Admin Website UNIKOM

Tag: Garuda

  • Implementation of Speech to Text and Geolocation Technology on the Travelling Merchant System

    Authors: Eko Budi Setiawan, Syifaul Mukminin, Angga Setiyadi, Chrismikha Hardyanto

    DOI: 10.1109/INCITEST64888.2024.11121482

    Abstract

    Travelling merchants are a type of business providing food needs in the informal sector. Currently, when consumers want to make an order, they have to type in the criteria in the form of a fairly long text. Consumers also often have difficulty detecting the location of travelling merchants in their vicinity. This research developed a responsive mobile webbased application utilizing speech recognition technology, specifically speech-to-text. The speech-to-text feature is powered by the Hugging Face Whisper API, which is applied during the food ordering process and geolocation technology in the form of Location Based Service to detect travelling merchants’ locations in real-time. The notification feature uses service workers and socket.io to provide information when a travelling merchant is nearby the buyer’s location. The software was developed using an agile methodology model. Based on the test results, 100% of all features were running well, 82.4% of travelling merchants, and 94.4% of buyers agreed that the implementation of the research results could simplify the process of ordering food and meeting between travelling merchants and buyers. © 2024 IEEE.

    Author keywords

    Geolocation; Hugging face; Push notification; Speech-to-text; Travelling merchant

    This article can be accessed at: https://www.scopus.com/pages/publications/105015715547

  • T-Shirt Vending Machine Using ESP32

    Authors: Mochamad Fajar Wicaksono; Myrna Dwi Rahmatya; Syahrul Syahrul; Sri Nurhayati

    DOI: 10.1109/INCITEST59455.2023.10396857

    Abstract

    At tourist locations, souvenirs are usually found at specific spots, but when utilizing a vending machine, souvenir sales can be made anywhere. The purpose of this research is to create a t-shirt sales system using a vending machine. This vending machine can be placed at tourist attractions, airports, or terminals. The method used in this research is experimental. From the client-side, this system is controlled using ESP32. Meanwhile, from the server-side, the system is supervised by the admin using a web application. The admin is tasked with verifying incoming orders via the web application and sending the barcode to the buyer’s email. ESP32 will wait for the buyer to scan the barcode on the vending machine. If it’s is valid, the ESP32 will inform the buyer via OLED. Furthermore, the ESP32 will activate the DC motor to run the conveyor. The completion of the t-shirt release process is detected using an infrared sensor. ESP32 then informs the server that the process has been completed. Modular testing is carried out to test each tool component’s function. Meanwhile, overall testing is carried out to test system functionality. Based on the results of these two tests, the Vending Machine Using ESP32 works 100% as expected. This tool can read the QR-Code input, validate data to the server, issue t-shirts according to validated orders, and request order deletion to the server when the transaction has been completed. With this tool, it is hoped that it can help sellers and make it easy for tourists. © 2023 IEEE.

    Author keywords

    ESP32; QR-Code; T-Shirt; Vending Machine

    Indexed keywords

    Engineering controlled terms

    Bar codes; DC motors; Infrared detectors

    Engineering uncontrolled terms

    Client sides; D.C. motors; ESP32; QR codes; Server sides; Souvenir sales; T-shirts; Tourist attractions; WEB application; Web applications

    Engineering main heading

    Sales

    This article can be accessed at https://www.scopus.com/pages/publications/85185195782

  • Autonomous Cyber Physical System : Theoretical Model for Monitoring Green Tea Withering Process

    Authors: Sopian Alviana; Taufik Djatna; Shelvie Nidya Neyman; Sri Wahjuni

    Abstract

    Cyber Physical System (CPS) is a relationship between physical entities and virtual entities. CPS has a complexity in conducting relationships between three components, namely physical components, cyber components, and communication. CPS moves towards an autonomous system, so a modeling of the system is needed. Green tea withering is one of the processes in green tea processing by flowing a number of tea leaves into the withering machine. Model Based System Engineering-based modeling can be used to map the needs of autonomous CPS systems. Autonomous CPS modeling is expected to map the needs both in terms of functional, logical, and physical for autonomous CPS in the green tea withering process. This model is expected to represent the behavioral needs of the autonomous CPS system for green tea withering as a case study on autonomous CPS that moves dynamically. Therefore, the depiction of this model can provide an overview of the needs of the autonomous system in the green tea withering process. © 2023 IEEE.

    Author keywords

    autonomous; cyber physical system; green tea; model-based system engineering

    This article can be accessed at: https://www.scopus.com/pages/publications/85185197647

  • Integrating Whitebox and Blackbox Testing for Web Based Point of Sales: A Halstead’s Metric and Equivalence Partitioning Approach

    Authors:  Rangga Sidik, Novia Ardianti Dewi, Aghnia Maulina, Mega Lestari, Syahrul Mauluddin

    DOI: 10.1109/INCITEST64888.2024.11121454

    Abstract

    In developing Point of Sales software, quality is an important concern. Point of sales applications must be able to provide effectiveness in storing company sales data. In the development stage, software testing is an important stage to produce software with good quality. Appropriate testing can provide an overview of the real quality of software. This research was conducted with the aim of testing web-based Point of Sales software to suit the user’s needs. In addition, the functional system can be believed to run well even with the complexity of the source code of the point of sales. We conducted testing by combining white box testing with black box testing. The use of equivalence Partitioning and Halstead’s Metric methods is combined to get test results in terms of functionality and program complexity. So that it can get the effectiveness of the source code and be able to evaluate the level of programming difficulty. From the results of white box testing, we get 9 errors that appear on the shop page, and 13 other test cases are declared valid from 17 test cases conducted using Equivalence Partitioning. Based on the test results with Halstead’s Metric measurement method, the test results show that the source code has a high volume and difficulty level with a high value as well. This can define the level of complexity of PoS. From these findings, PoS software developers can utilize it for the next stage of development in creating a higher quality and reliable Point of Sales system. © 2024 IEEE.

    Author keywords

    equivalence partitioning; halstead metric; point of sales; testing; web application

    This article can be accessed at https://www.scopus.com/pages/publications/105015891816

  • Application of Agglomerative Hierarchical Clustering (AHC) for Grouping Prospective Scholarship Recipients

    Authors: Fakhrian Fadlia Adiwijaya; Arisza Zufar Fathurrahman; Chrismikha Hardyanto

    DOI: 10.1109/INCITEST59455.2023.10395926

    Abstract

    Scholarships are a form of financial aid provided to individuals to support their education. Scholarships can be granted by educational institutions, governments, organizations, or individuals as a means of assisting with educational expenses. Scholarships can take various forms, including cash grants, tuition waivers, assistance in purchasing books or educational supplies, or a combination of these forms of aid. Scholarship recipients are typically selected based on specific criteria such as academic excellence, financial need, special talents or interests, or achievements in a particular field. The current method of selecting scholarship recipients involves evaluating the characteristics of students who are deemed outstanding and actively engaged in the learning process. Due to the large number of students and limited scholarship quotas, this process often consumes a significant amount of time, and the selection of scholarship recipients may not be entirely accurate. To aid in the selection of scholarship recipients, data mining techniques are employed, specifically the clustering method. Agglomerative Hierarchical Clustering (AHC) is one of the clustering algorithms capable of grouping a set of data into specific data clusters. The data analyzed for the determination of scholarship recipients include academic performance data or grades, extracurricular activity data, economic data, and student achievement data. By using data mining and clustering methods, educational institutions or organizations offering scholarships can analyze the data of potential scholarship recipients to identify groups with similarities based on specific criteria. This approach can make the selection and grouping of scholarship candidates more efficient and objective. © 2023 IEEE.

    Author keywords

    AHC; Clustering; Data Mining; Scholarship Management

    Indexed keywords

    Engineering controlled terms

    Cluster analysis; Clustering algorithms; Data mining; Economics; Learning systems

    Engineering uncontrolled terms

    ‘current; Academic excellence; Agglomerative hierarchical clustering; Clusterings; Educational institutions; Financial aids; Government organizations; Prospectives; Scholarship management; Special talents

    Engineering main heading

    Students

    This article can be accessed at https://www.scopus.com/pages/publications/85185202061

  • Big Data Algorithms as Enablers of Sustainable Marketing Practices: A Preliminary Research

    Authors: Rahma Wahdiniwaty; Neng Susi Susilawati Sugiana; M Yani Syafei; Yeffry Handoko Putra; Lili Adi Wibowo

    DOI: 10.1109/INCITEST59455.2023.10396871

    Abstract

    This study presents a systematic literature review on the role of Big Data algorithms in supporting sustainable marketing practices. Sustainable marketing practices have garnered increasing attention in an era where environmental concerns and consumer awareness of environmental issues are on the rise. In this context, Big Data algorithms have gained a progressively crucial role in analyzing large datasets to understand consumer behaviors, predict market trends, and design effective marketing campaigns. The literature review is conducted using a systematic approach to identify, collect, and analyze relevant articles pertaining to the utilization of Big Data algorithms in sustainable marketing practices. The primary findings of this literature review encompass the application of Big Data algorithms in sustainable market needs analysis, the utilization of these algorithms to predict environmentally friendly consumer patterns, and their employment in tailoring sustainable marketing campaigns. Through an in-depth analysis of the existing literature, this research identifies the significant contributions of Big Data algorithms in supporting sustainable marketing practices. However, challenges related to data privacy, analysis integrity, and technological complexity are also revealed in the literature. The practical and theoretical implications of these findings illustrate the potential of Big Data algorithms as valuable tools for marketing practitioners seeking to adopt more sustainable approaches. © 2023 IEEE.

    Author keywords

    Algorithms; Big Data; Marketing Practices; Sustainable Marketing

    This article can be accessed at: https://www.scopus.com/pages/publications/85185195922

  • Travel Route Recommendation System Based on 

    Weather Prediction and Geolocation Technology

    Authors: Eko Budi Setiawan; Gradiyanto Putera Husein; Angga Setiyadi

    DOI: 10.1109/INCITEST59455.2023.10395920

    Abstract

    Information regarding weather conditions is quite challenging to obtain and predict. People who in their daily lives have to travel to a destination, such as workers or traveling salesman, cannot predict the weather they will encounter during their journey. The impact is that people often experience some losses when traveling. This research produces an Android-based application used by the general public to avoid travel losses caused by weather. The application provides travel recommendation information in the form of routes obtained from the Google Directions API and weather predictions obtained from the Dark Sky API. This application helps users predict weather predictions on their path to reduce the risk of rain loss. The results of black-box testing and user acceptance found that Android-based applications managed to get a level of 84% in providing travel recommendations, achieving 80.6% success in predicting the weather and obtaining 76.6% in reducing rain risk. © 2023 IEEE.

    Author keywords

    Geolocation; Recommendation; Route; Travel; Weather

    Indexed keywords

    Engineering controlled terms

    Acceptance tests; Android (operating system); Application programming interfaces (API); Black-box testing; Weather forecasting

    Engineering uncontrolled terms

    Condition; Daily lives; Geolocations; Recommendation; Route; Travel; Travel routes; Weather; Weather prediction; Workers’

    Engineering main heading

    Rain

    This article can be accessed at https://www.scopus.com/pages/publications/85185198321

  • Optimizing Bankruptcy Prediction Through Filterbased Feature Selection

    Authors: Nelly Indriani Widiastuti, Ednawati Rainarli, Kania Evita Dewi

    DOI: 10.1109/INCITEST64888.2024.11121476

    Abstract

    Predicting corporate bankruptcy is a critical task in the financial world. Artificial intelligence techniques have become a promising solution for predicting a company’s condition. One of the crucial phases in the bankruptcy prediction process is feature selection. It is the procedure for choosing from a data set the variables that will be most useful in a prediction model. This study examines and evaluates the important feature selection for machine learning-based bankruptcy prediction. A Taiwan Stock Exchange public dataset, which includes 95 financial ratio features, was used in the study. Several filtering feature selection methods, such as Analysis of Variance (ANOVA), Kendall Tau correlation coefficient (KT), and Mutual Information (MI), were used to identify relevant features. The methods of bankruptcy classification applied were Logistic Regression and Random Forest. According to the test results, ANOVA produces the highest recall value for the Taiwan Stock Exchange dataset, both with the Logistic Regression (LR) and Random Forest (RF). Using 61% of the available features or 58 features, the bankruptcy prediction model with Random Forest performed well without compromising accuracy, precision, recall, or F1 score. © 2024 IEEE.

    Author keywords

    ANOVA; bankruptcy prediction; feature selection; filtering methods; financial ratios

    This article can be accessed at https://www.scopus.com/pages/publications/105015829493

  • Optimizing Bankruptcy Prediction Through Filterbased Feature Selection

    Authors: Nelly Indriani Widiastuti, Ednawati Rainarli, Kania Evita Dewi

    DOI: 10.1109/INCITEST64888.2024.11121476

    Abstract

    Predicting corporate bankruptcy is a critical task in the financial world. Artificial intelligence techniques have become a promising solution for predicting a company’s condition. One of the crucial phases in the bankruptcy prediction process is feature selection. It is the procedure for choosing from a data set the variables that will be most useful in a prediction model. This study examines and evaluates the important feature selection for machine learning-based bankruptcy prediction. A Taiwan Stock Exchange public dataset, which includes 95 financial ratio features, was used in the study. Several filtering feature selection methods, such as Analysis of Variance (ANOVA), Kendall Tau correlation coefficient (KT), and Mutual Information (MI), were used to identify relevant features. The methods of bankruptcy classification applied were Logistic Regression and Random Forest. According to the test results, ANOVA produces the highest recall value for the Taiwan Stock Exchange dataset, both with the Logistic Regression (LR) and Random Forest (RF). Using 61% of the available features or 58 features, the bankruptcy prediction model with Random Forest performed well without compromising accuracy, precision, recall, or F1 score. © 2024 IEEE.

    Author keywords

    ANOVA; bankruptcy prediction; feature selection; filtering methods; financial ratios

    This article can be accessed at https://www.scopus.com/pages/publications/105015829493

  • Design of Public Service Administrative Information System

    Authors: Herry Saputra, Nizar Rabbi Radliya, Eddy Soeryanto Soegoto, Fadhil Fadhilah Ilham

    DOI: 10.1109/INCITEST64888.2024.11121522

    Abstract

    The sub-district office is one of the government agencies in charge of carrying out population administration services at the village level. However, in the implementation of these administrative services, there are still obstacles that must be addressed, such as the process of making certificates and filing documents, which are still too complicated and timeconsuming. To help solve these problems, we need an information system website that can help speed up the administrative service process. This study aims at designing an administrative information system called SIAPIK, which is based on the website at the Tagaraja Sub-district office, Riau, Indonesia. Descriptive analysis method with a qualitative approach was used in this study to determine the needs of the system to be built. In the system development process, we used the concept of an object-oriented approach with a prototype development method. The results showed that the development of this information system could provide convenience in supporting the role and function of the sub-district office in carrying out the population service administration process. In conclusion, the existence of this online administration system will help solve various problems at the sub-district office. This information system used an online system that allows residents to make certificate via desktop or smartphone devices regardless of time and place without having to come directly to the sub-district office. © 2024 IEEE.

    Author keywords

    information system; public service administrative; SIAPIK

    This article can be accessed at https://www.scopus.com/pages/publications/105015848206