Admin Website UNIKOM

Tag: best private campus in Bandung

  • Implementation of Convolutional Neural Network for Sundanese Script Handwriting Recognition with Data Augmentation

    Implementation of Convolutional Neural Network for Sundanese Script Handwriting Recognition with Data Augmentation

    Authors: Irfan Maliki, Ade Syahlan Prayoga

    Abstract

    Sundanese script is one of the cultural heritages that need to be preserved. However, Sundanese script has complexity and uniqueness in its writing, making it difficult to recognize. The recognition can be done automatically using deep learning. One of the problems is that the recognition has a small amount of data and is less varied. In this study, the proposed solution is to use data augmentation. This study focused on how the use of data augmentation can help to improve accuracy in performing the recognition of handwriting image pattern using Convolutional Neural Network (CNN) method. Data augmentation is the process of artificially increasing the amount of data by generating new data points from existing data. The augmentation includes adding small changes to the data or using machine learning models to generate new data points in the latent space of the original data in order to strengthen the data set. Data augmentation applied in this research is flipping, rotation, and translation techniques. Based on the results, it can be concluded that the use of data augmentation to increase the number and variety of data samples has a significant effect on the accuracy value of 0.1707 or about 17.07%. The best accuracy obtained is 0.8 or 80% using a baseline model with data augmentation. These findings yielded good results because the system is able to perform image classification quite well. © School of Engineering, Taylor’s University.

    Author keywords

    Convolutional neural network; Data augmentation; Handwriting recognition; Sundanese script

    This article can be accessed at https://www.scopus.com/pages/publications/85152919072 

  • Efficient Sampling-Based for Mobile Robot Path Planning in a Dynamic Environment Based on the Rapidly-Exploring Random Tree and a Rule-Template Sets

    Efficient Sampling-Based for Mobile Robot Path Planning in a Dynamic Environment Based on the Rapidly-Exploring Random Tree and a Rule-Template Sets

    Authors: Muhammad Aria Rajasa Pohan, Jana Utama

    DOI: 10.5829/ije.2023.36.04a.16

    Abstract

    This study presents an efficient path planning method for mobile robots in a dynamic environment. The method is based on the rapidly-exploring random tree (RRT) algorithm. The two primary processes in mobile robot path planning in a dynamic environment are initial path planning and path re-planning. In order to generate a feasible initial path with fast convergence speed, we used a hybridization of rapidly-exploring random tree star and ant colony systems (RRT-ACS). When an obstacle obstructs the initial path, the path re-planner must be executed. In addition to the RRT-ACS algorithm, we proposed using a rule-Template set based on the mobile robot in dynamic environment scenes during the path re-planner process. This novel algorithm is called RRT-ACS with Rule-Template Sets (RRT-ACS+RT). We conducted many benchmark simulations to validate the proposed method in a real dynamic environment. The performance of the proposed method is compared to the state-of-The-Art path planning algorithms: RRT∗FND and MOD-RRT∗. Numerous experimental results demonstrate that the proposed method outperforms other comparison algorithms. The results show that the proposed method is suitable for the use on robots that need to navigate in a dynamic environment, such as self-driving cars. © 2023 Materials and Energy Research Center. All rights reserved.

    Author keywords

    Dynamic Environment; Efficient Sampling; Path Planning; Rapidly-exploring Random Tree; Rule-Template Sets

    This article can be accessed at https://www.scopus.com/pages/publications/85152951615

  • Computational Bibliometric Analysis Of Evolutionary Game Theory (EGT) Research Using Vosviewer

    Computational Bibliometric Analysis Of Evolutionary Game Theory (EGT) Research Using Vosviewer

    Authors: Rizky Jumansyah, Eddy Soeryanto Soegoto, Chepi Nur Albar

    Abstract

    Evolutionary Game Theory (EGT) applies mathematical theory to the biological context, providing an analytic framework, contest, and strategy. The assumption underlying EGT is that human behavior, whether in games or decision-making problems, often does not arise from rational reasoning. This study aimed to perform a bibliometric analysis of EGT using VOSviewer software to map the research landscape. The approach used was bibliometric and descriptive quantitative, with data obtained from a Google Scholar search for articles with the keyword “Evolutionary game theory” using the publish or perish application. The search yielded 990 articles published between 2017-2021. The results showed that research on EGT peaked in 2017-2019 but declined in 2020 and 2021. In conclusion, this study highlights the importance of bibliometric analysis, particularly in the field of EGT, and provides a reference for future research on this topic. © School of Engineering, Taylor’s University.

    Author keywords

    Bibliometric; Data analysis; Evolutionary game theory; Game; Publish or perish; VOSviewer

    This article can be accessed at https://www.scopus.com/pages/publications/85158098948

  • Pupil Center Detection Using Radial Symmetry Transform to Measure Pupil Distance in the Eye

    Pupil Center Detection Using Radial Symmetry Transform to Measure Pupil Distance in the Eye

    Authors: Utama J, Fitriani V. R. L

    DOI: 10.5829/ije.2023.36.05b.04

    Abstract

    In patients with refractive errors or impaired vision, light rays received by the pupil do not fall directly onto the retina. This can be corrected by wearing monocled glasses. The focal point of the eyeglass lens needs to be adjusted to the center of the user’s pupil. This can be known through the measured pupil distance (PD) value information. The measurement of the PD is very important to determine the center distance of the pupils in both eyes. where the eye does not experience the prism effect. This study aims to apply the radial symmetry transformation (RST) method combined with self-quotient (SQI) imagery to detect the pupillary center and measure PD. This algorithm combines to get more optimal results in detecting the center of the pupil in dark conditions or those exposed to shadow illumination. The program created using the MATLAB software simulates PD measurements for pupillary center detection in bright and dark images conditions. The test was carried out ten times, and the results showed that the system was able to measure PD on low-resolution images of 300 x 300 pixels at 72 dpi in bright image conditions; with measurement uncertainty values in each image of 0.60 mm. As for testing on dark images, the uncertainty values are 0.80 mm. In this case, the standard deviation value is obtained from the effect of the different dimensions of the face object on the tested image. © 2023 Materials and Energy Research Center. All rights reserved.

    Author keywords

    Euclidean Distance; Pupil Distance; Radial Symmetry Transform; Self Quotient Image

    This article can be accessed at https://www.scopus.com/pages/publications/85158848120

  • Designing a Learning System Based on Intelligent Tutoring System Using Design Science Research

    Designing a Learning System Based on Intelligent Tutoring System Using Design Science Research

    Authors: Bobi Kurniawan, Meyliana M, Harco Leslie Hendric Spits Warnars, Bambang Suharjo

    Abstract

    The development of technology especially computers has now been widely used in the field of education. The application of technology in helping the education process today continues to develop toward an intelligent tutoring system. An Intelligent Tutoring System (ITS) is an intelligent guidance system with adjustments to learning models to individual needs. Building an Intelligent Tutoring System not only focuses on building software but also requires adaptation and application of other technologies that can help adapt the curriculum into an intelligent learning system. This study aims to review and design the needs for the development of an Intelligent Tutoring System in the learning system. The framework or design of the Intelligent Tutoring System is designed using the Design Science Research approach which consists of identification and motivation, goal setting, design and development, demonstration, evaluation, and communication. The proposed framework can be a guide in designing an Intelligent Tutoring System, especially in an electronic-based learning system (e-learning). The development of the framework is carried out with stages in the design science research, namely by identifying problems in e-learning, objectives, design and development of intelligent tutoring system based e-learning, evaluation and communication. With this framework, it can make it easier to manage e-learning by using an intelligent tutoring system © School of Engineering, Taylor’s University.

    Author keywords

    Design science research; Framework; Information system; Intelligent tutoring system; Learning

    This article can be accessed at https://www.scopus.com/pages/publications/85164667966

  • Information Technology in Evolutionary Strategic Management: Decision Support System in Smart University

    Information Technology in Evolutionary Strategic Management: Decision Support System in Smart University

    Authors: Senny Luckyardi, Agus Rahayu, Lili Adiwibowo, Ratih Hurriyati

    Abstract

    Decision Support Systems (DSS) refer to computerized information systems that encompass knowledge-based systems and knowledge management to provide assistance for decision-making in businesses or organizations. DSS tools have created from the evolutionary concept of data processing and management information system (MIS). In the connection of strategic management, SS combines human intellectual resources with computer capabilities to enhance the quality of decision-making. A manager/leader and computers must work together as a problem-solving team in organization. Despite the evolutionary strategic management issue of DSS in a company, the utilization of DSS in smart university management are still interesting topic to be studied. Due to tighter competitiveness in higher education, the universities must apply Information and Communication Technologies (ICT) to support their competitive advantage and brand themselves as a smart university. The novelty of the research is in combining DSS in the framework of smart university. Although the concept of Smart in the education area involves the emergence of technologies such as DSS, as the consequence of various terminology in the domain of IT-based university, the elements of smart university are widespread and fragmented. This paper aimed to define how DSS works in diverse explicit circumstances or settings by investigating its smart university readiness index. The method use analysis descriptive and DSS of a private university was analysed as a model in this research. The results showed that DSS has a strong relation with term “smart university” and effective to be used as strategic management tools and to help top management decide an appropriate decision making. Thus, ICT is gradually becoming very imperative for assisting the processes of decision making, especially in smart university domain. © School of Engineering, Taylor’s University.

    Author keywords

    Decision support system; Information technology; Smart university; Strategic management

    This article can be accessed at https://www.scopus.com/pages/publications/85164533986

  • A Bert Model to Detect Provocative Hoax

    A Bert Model to Detect Provocative Hoax

    Authors: Rio Yunanto, Eri Prasetyo Wibowo, Rianto R

    Abstract

    The information flood makes social media users vulnerable to becoming victims of provocative hoaxes or even spreading hoaxes themselves. This research examines the capabilities of two variants of Bidirectional Encoder Representations from Transformers (BERT) models for the Indonesian language (IndoBERT Base Model and Indonesian BERT base model 522M) in developing the detection of provocative hoaxes in the Indonesian language. The proposed method used two variants of the monolingual BERT model for the Indonesian language from the Huggingface library. The proposed method’s architectural flow starts with data collection and labelling from community hoax collector websites, followed by pre-processing. The cleaned data is then divided into training and test data to proceed to the fine-tuning stage, where several layers and weights of the BERT model are adjusted to fit the desired classification task. The experimental results of the study show that the recommended Indonesian BERT variant for the detection of provocative hoaxes is the IndoBERT Base Model with a learning rate of 1e-5, a batch size of 32, and a maximum length limit of 128 tokens, achieving an average training accuracy of 99,22%, with a training time of 21min 52s. The research findings also indicate that a learning rate 1e-5 can produce better test accuracy than a learning rate of 2e-5 or 3e-5. The detection model of provocative hoaxes using Indonesian BERT variants needs to be improved, especially in terms of collecting a large amount of hoax data, to enhance the accuracy of the provocative hoax detection model. © School of Engineering, Taylor’s University.

    Author keywords

    Accuracy; Classification; Hoax; Provocative; Transformer

    This article can be accessed at https://www.scopus.com/pages/publications/85180943734

  • Application of Data Mining for Predicting Horticultural Commodities Price

    Application of Data Mining for Predicting Horticultural Commodities Price

    Authors: Dian Dharmayanti, Arsy Opraza Akma, Eddy Soeryanto Soegoto, Lia Warlina

    Abstract

    In Garut Regency, farmers receive their commodities at a selling price that collectors determined. This issue causes farmers to lose money because their income does not correspond with the market price for their goods. This loss has an impact on reducing farmer productivity. Therefore, this research aims to provide recommendations to farmers regarding commodity selling prices for particular periods based on forecast results. This research implements data mining so commodities selling prices can be predicted. Data mining is a way to find patterns or knowledge from past data. Multiple linear regression methods or algorithms are used to find patterns or knowledge with data mining. Multiple linear regression can be used to predict commodities selling prices based on rainfall factors and total production. The application developed can produce recommendations for farmers for commodity selling prices for specific periods based on prediction results, thereby reducing losses for farmers. © 2024 Taylor’s University. All rights reserved.

    Author keywords

    Data mining; Horticultural commodities; Multiple linear regression; Price prediction

    This article can be accessed at https://www.scopus.com/pages/publications/85184573404

  • Risk Assessment of Work Posture in Manufacture Industry for Drilling Pipes

    Risk Assessment of Work Posture in Manufacture Industry for Drilling Pipes

    Authors: Herry Saputra, Purnama Dicki Gilar, Putra Muhamad Luki, Ramdoni Mochamad Sahri, Suryatno Wiganepdo Soegoto

    Abstract

    The objective of this research is to examine the working postures of individuals in the manufacturing sector involved in drilling pipe production, utilizing the REBA and RULA methodologies. These methods serve to identify potential risks associated with the working positions adopted by workers, distinguishing between correct postures and those that may lead to problems. The research involved data collection through field observations and photographs capturing operators’ postures during work. Data processing entailed creating sketches of the operator’s posture and measuring the angles of each part using CorelDRAW. This facilitated aligning the measurement results with the RULA and REBA worksheets. The results indicated that the RULA method, when assessing the operator’s working posture while handling the spindle, yielded a final score of 5. The assessment outcome indicates a need for a thorough examination of the operator’s body positioning, and prompt implementation of alterations in working posture is essential. Conversely, the application of the REBA method to the operator’s stance while dealing with the iron pipe resulted in a score of 8, signifying a significant risk level. Immediate improvements in the working position are considered imperative. © 2024 Taylor’s University. All rights reserved.

    Author keywords

    REBA; Risk assessment; RULA; Work posture

    This article can be accessed at https://www.scopus.com/pages/publications/85184616974

  • Prediction System for Vitamin A Requirements in West Java Indonesia Using Linear Regression

    Prediction System for Vitamin A Requirements in West Java Indonesia Using Linear Regression

    Authors: Sri Nurhayati, Wicaksono M. F, Diana Effendi

    Abstract

    This research aims to analyse the accuracy of the linear regression method in predicting the amount of vitamin A needed and providing information about predicting vitamin A needs within a specific period. Vitamin A is identified as the most crucial nutrient, necessitating external supplementation due to insufficient and low food consumption. The linear regression method is employed in this study, serving as a data analysis technique to predict unknown data values based on related and known data values. The dataset involves the number of vitamin A administrations in each district/city in West Java, Indonesia. Mean absolute perception error (MAPE) was utilized to assess prediction errors. The system requirements analysis used an object-oriented approach with Unified Modeming Language (UML) tools. The comprehensive prediction results yield an average accuracy of 86%, indicating that the linear regression method effectively predicts vitamin A needs in the subsequent period. Functional testing of the system demonstrates a 100% success rate, aligned with the needs analysis, thereby providing information on predicting vitamin A requirements within a specific period for each district/city in West Java. © 2024 Taylor’s University. All rights reserved.

    Author keywords

    Linear regression; Object-oriented based system; Prediction; Vitamin A

    This article can be accessed at https://www.scopus.com/pages/publications/85184606841