Admin Website UNIKOM

Tag: Times Higher Education Ranking

  • Feature Classification Based on Heterogenous Data 

    Using Hybrid Machine Learning: A Review

    Authors: Agus Nursikuwagus ; Heri Purwanto ; Deshinta Arrova Dewi

    DOI: –

    Abstract

    Heterogeneous data is a dataset with various types including data type and data source. Classification of heterogeneous data is still becoming a discussion in research in the field of intelligence artificial especially in learning classification. Based on data and machine development classification, then study this still relevant done. Machine classification that is still trending now is a hybrid engine known as the collaboration technique such as a fuzzy technique and neural network. The aim of this review paper is to find opportunity research on hybrid machine learning that perform classification on heterogeneous data with multi-class targets. There are several challenges on heterogeneous data such as 1)determining algorithm normalization and text processing as Step beginning from the input layer, 2) function formation variable linguistics for every case allow existence opportunity study for linguistic processes, 3) A membership function algorithm that can adapt of the dataset used can as opportunity research, 4) finding method shaper fuzzy rule as machine inference from a neural network, 5) Process structure of every task, 6) performance like efficiency memory for processing ( management memory ), complexity (process time), and validation architecture (accuracy, precision, recall, f-measure, specification, true prediction, false prediction). The result of the research obtained is the existence opportunity for improving or developing a hybrid classification machine that can handle heterogeneous data with multi-class targets. © 2023 Little Lion Scientific.

    Author keywords

    Classification; Feature; Fuzzy; Heterogenous Data; Neural Network

    This article can be accessed at https://www.scopus.com/pages/publications/85174975477

  • Development of Innovative Model for Waste Management System Using Internet of Things (IoT) and Machine Learning

    Development of Innovative Model for Waste Management System Using Internet of Things (IoT) and Machine Learning

    Authors: Dimas Akmarul Putera, Ansarullah Lawi, Filmada Ocky Saputra, Sri Handayani, Yun Arifathul Fatimah, Ivan Muhammad Reza, Sholikun, Zainal Arifin Hasibuan, Alvendo Wahyu Aranski

    DOI: 10.1109/ICIC64337.2024.10957619

    Abstract

    Effective and efficient waste management system is a significant challenge in coastal areas. This research proposes the development of innovative model for waste management system using IoT and machine learning. The system consists of hardware such as cameras and image sensors installed in coastal locations to monitor the volume and types of waste. The image data collected from these devices is stored and analyzed using image processing techniques. The YOLO (You Only Look Once) algorithm is used to identify and classify the types of waste. The identified data is then used to train a machine learning model, which allows for the prediction of future waste volume and types. These predictions are used to optimize waste collection schedules, reduce operational costs, plan the types of waste processing, and minimize environmental impact. With this holistic approach, the system is expected to enhance the efficiency and sustainability of waste management, providing an innovative solution to waste management issues in coastal areas. © 2024 IEEE.

    Keywords: Environment; Image Processing; IoT; Machine Learning; Waste Management

    This article can be accessed at https://www.scopus.com/pages/publications/105004578932

  • Computation Offloading Quality of Service Improvement with Multi Layer Approach Based on Pure Edge Sim Simulator

    Authors: Agus Mulyana; Sri Wahjuni; Taufik Djatna; Heru Sukoco

    DOI: 10.1109/INCITEST59455.2023.10396945

    Abstract

    Internet of Things (IoT) devices possess limited resources, encompassing computing speed of microcontrollers, memory capacity, battery energy sources, and diverse computational requirements. The prevalent system architecture comprises two tiers, wherein IoT devices connect to the cloud to transmit data or receive commands, relying on a stable internet connection. However, in environments with restricted internet coverage and stability, this situation can become problematic, potentially leading to a reduction in Quality of Service (QoS). A viable solution to address this challenge involves the utilization of computation offloading techniques. This approach involves recognizing computation characteristics through specific rules, determining whether computations should take place at the IoT device layer, the edge computing layer, or within cloud computing. By implementing a multilayer three-tier scheme, the computation offloading process can be effectively optimized. Empirical testing conducted using the Pure Edge Sim simulator demonstrates that the amalgamation of IoT, Edge, and Cloud layers yields the most favorable outcomes for enhancing QoS. © 2023 IEEE.

    Author keywords

    computation offloading; internet of things; multilayer approach; pure edge sim; quality of services

    This article can be accessed at: https://www.scopus.com/pages/publications/85185199021

  • Application of The Thinking Design Method in UI/UX Design of The Lawn Mowing Service Application

    Authors: Muhammad Faiz Islami, Hanhan Maulana

    DOI: 10.1109/INCITEST64888.2024.11121521

    Abstract

    People in Bangkinang have difficulty accessing effective lawn mowing services. Lack of information and traditional ordering methods prevent service providers from expanding their market reach. A digital solution is needed to overcome this problem. In Bangkinang City, many people are engaged in lawn mowing services, but ordering for this service traditionally done through brochures and word of mouth recommendations. This study focuses on developing a lawn mowing service application using design thinking. English This methodology consists of five steps: empathize, define, ideate, prototype, and test, which are used to understand user needs, generate ideas, develop prototypes, and test solutions. The research results in the form of a prototype mobile application for service providers in Bangkinang City, tested using the system usability ladder (SUS) method with an average score of 80.66 for the service user application and 81.5 for the service provider application, indicating that the application is easy to use and well received by users. This research is expected to improve the effectiveness and efficiency of the lawn mowing servise booking process and contribute significantly to the development of information technology in the community. This research aims to design a lawn mowing service application using design thinking, to give users a good experience and can efficiently order services. © 2024 IEEE.

    Author keywords

    Design thinking; system usability scale; usability

    This article can be accessed at https://www.scopus.com/pages/publications/105015889965

  • Elevating PCP Level Implementation: A Strategic Approach

    with Architecture Software in the COVID-19 Era

    Authors: Yeffry Handoko ; Zuriani Ahmad Zukarnain ; Rahma Wahdiniwaty ; Noorihan Abdul Rahman

    DOI: 10.1109/INCITEST59455.2023.10397031

    Abstract:

    Between 2020 and 2021, Indonesia and Malaysia were both impacted by the COVID-19 pandemic. These two nations adopted a strikingly similar approach to restrict access to certain areas, primarily relying on identical criteria. In the case of Indonesia, the PPKM (Community Activity Restriction Enforcement) strategy was employed, this contained a component involving the exposure of COVID-19 affected persons. The determination of the extent of this exposure was primarily based on monitoring the rise in the number of individuals who had been in contact with COVID-19-positive individuals. However, during a pandemic, it’s imperative to consider not only those individuals who are confirmed to have COVID-19 but also those who exhibit symptoms similar to COVID-19 but have not been diagnosed. These individuals can also act as potential agents for the spread of the pandemic within their communities. This is referred to as the “Precaution COVID-19 Pandemic (PCP) Level.”It’s important to note that the PCP level does not have a universally agreed-upon definition at present. Instead, it is influenced more by the number of individuals displaying symptoms akin to those of COVID-19, rather than solely relying on the increase in confirmed COVID-19 cases. In essence, both countries shared a comparable approach to pandemic management during this period, but the emergence of the PCP level underscores the need for a broader perspective when assessing the pandemic’s impact on communities, considering not only confirmed cases but also symptomatic individuals who may contribute to its spread. The PCP Level determination can be used for preventative policy and to supplement the previous Pandemic Level Methods. Two methods are employed to realize this concept: the AHP approach for calculating the Covid-19 pandemic level and the K-Mean algorithm for pattern clustering. Data was gathered from 11 health centers in West Java province. The findings of this study underscore the potential of a multifaceted approach to pandemic precaution in the context of COVID-19. By synergistically integrating the three proposed algorithms and leveraging data on symptoms that are intricately linked to the predominant COVID-19 symptoms, this research illuminates a promising avenue for determining and implementing precautionary measures during the ongoing pandemic. This holistic strategy not only offers a more comprehensive understanding of the virus’s spread but also furnishes decision-makers with a versatile toolkit for tailoring precautionary actions to different circumstances, ultimately contributing to more effective and adaptive pandemic management. © 2023 IEEE.

    Author Keywords:

    clustering; Covid-19; decision support; Precaution

  • Assessing the Effectiveness of Notebar, ConductorBot, and Sub-ConductorBot Methods in Training Novice Angklung Players

    Authors: Bella Hardiyana ; Adam Mukharil Bachtiar ; Kazuki Sugita ; Diana Effendi ; Beri Noviansyah ; Prarinya Siritanawan ; Wen Gu ; Koichi Ota ; Shinobu Hasegawa

    DOI: 10.1109/INCITEST64888.2024.11121467

    Abstract

    This study explores the effectiveness of three distinct angklung training methods: Notebar (NB), ConductorBot (CB), and ConductorBot with Sub-Conductor (SC), specifically targeted towards novice angklung players. The objective was to assess the impact of each method on participant performance, as measured by the difference between Pre-test and Post-test scores. The study’s aims are threefold: (1) to determine the most effective method for enhancing novice players’ skills, (2) to assess whether these methods can provide a sufficient alternative to human conductors during practice, and (3) to explore the potential of these tools in both home and school practice environments. Through Pre-test and Post-test performance comparisons, effect size analysis using Cohen’s d indicated medium impacts across all methods, with SC demonstrating the highest effect size (0.645), followed by NB (0.622) and CB (0.606). The Kruskal-Wallis test, however, revealed no statistically significant differences between the methods, suggesting that all three are similarly effective in improving performance. Participant feedback highlighted areas for improvement in usability, timing synchronization, and fluidity of movement, which are crucial for novice players as they develop foundational skills. These insights informed recommendations for future work, including the refinement of method usability, the enhancement of robotic movement fluidity, adjustments to timing for better synchronization, and the exploration of hybrid methods. This study underscores the importance of both overall method effectiveness and the need for personalized approaches to optimize learning outcomes, particularly for novice angklung players. © 2024 IEEE.

    Author keywords

    angklung; conductorbot; notebar; novice players; sub-conductor; training methods

    Indexed keywords

    Engineering controlled terms

    Learning systems; Robotics

    Engineering uncontrolled terms

    Angklung; Conductorbot; Effect size; Notebar; Novice player; Performance; Post test; Sub-conductor; Test performance; Training methods

    Engineering main heading

    Personnel training

    This article can be accessed at https://www.scopus.com/pages/publications/105015862069

  • Enhancing The Attractiveness of Coloring Books Using Live Texturing Augmented Reality

    Authors: Hanhan Maulana; Jajang Saeful Anshor; Hideaki Kanai

    DOI: 10.1109/INCITEST59455.2023.10396956

    Abstract

    This study aims to build live texturing augmented reality to enhance the attractiveness of coloring books. This research has four main stages, namely data gathering, object preparations, software development and evaluations. This study uses UNITY 3D in building a Live Texturing Augmented Reality System. The AR method used is the Marker method. Users can color the marker. The 3D object will follow the color given to the marker in real time. This system is expected to increase children’s attraction to coloring. It is hoped that children’s understanding of color can also be improved. Furthermore, this system is expected to help children in deciding on color selection so that children’s creativity is getting better. © 2023 IEEE.

    Author keywords

    Augmented Reality; Coloring books; interactive system; Live Texturing; Multimedia

    This article can be accessed at: https://www.scopus.com/pages/publications/85185195328

  • Support Vector Machine for Satellite Images Classification 

    Using Radial Basis Function Kernel Method

    Authors: Nur Suhaili Mansor ; Hapini Awang ; Sarkin Tudu Shehu Malami ; Amirulikhsan Zolkafli ; Mohammed Ahmed Taiye ; Hanhan Maulana 

    DOI: 10.1007/978-981-99-9589-9_23

    Abstract

    Machine learning, particularly Support Vector Machines (SVM), has gained popularity in geospatial data processing and image classification. Geospatial data from various sources may contain errors, impacting image classification accuracy. Traditional pixel-based and object-based methods struggle to classify complex land cover classes accurately. Previous studies explored machine learning algorithms like Random Forests, K-Nearest Neighbors, and Neural Networks. Still, they faced challenges capturing intricate relationships within images and required substantial labeled training data, leading to computational expenses. SVM with polynomial kernels was attempted in some studies, but it suffered from potential overfitting and inefficiency for large datasets. To overcome these issues, this study employed SVM with RBF and Linear kernels to classify multispectral satellite images from the SPOT-6 Satellite Imagery dataset in Sungai Kelang, Malaysia. Previous research evaluated each kernel’s performance accuracy compared using a test dataset, utilizing open-source tools like Jupyter Notebooks and Python libraries to explore SVM’s potential as a high-performance satellite image classification technique. The findings revealed that SVM with RBF kernel outperformed SVM with polynomial or linear kernels in classifying satellite images. The RBF kernel’s robustness allowed SVM to model intricate decision boundaries and capture complex patterns in the image data, making it suitable for tasks with non-linearly separable data. The study introduces a new methodology and theoretical contribution to image classification-related literature, shedding light on the efficacy of SVM-RBF for geospatial data processing. It provides an alternative to traditional approaches for complex image classification tasks. Moreover, the research assists in selecting the optimal algorithm for remote sensing and satellite imagery applications. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

    Author keywords

    Geospatial; Image Classification; Linear Kernel; Radial basis function (RBF) Kernel; Support Vector Machine

    Indexed keywords

    Engineering controlled terms

    Classification (of information); Complex networks; Data handling; Forestry; Large datasets; Learning algorithms; Nearest neighbor search; Radial basis function networks; Remote sensing; Satellite imagery; Statistical tests; Support vector machines

    Engineering uncontrolled terms

    Geo-spatial; Geospatial data processing; Images classification; Kernel-methods; Linear kernel; Performance; Polynomial kernels; Radial basis function kernels; Satellite image classification; Support vectors machine

    Engineering main heading

    Image classification

    This article can be accessed at https://www.scopus.com/pages/publications/85185723250

  • IoT (Internet of Things) as a Tool to Help Regulate Water Altitude on Dams

    Authors: Sri Supatmi, Dita Ardisura Pamungkas, Desta Rifaldi Nugraha, A Febrian, Audi Citra Fadilah

    DOI: 10.1109/INCITEST64888.2024.11121455

    Abstract

    In each dam in the Citarum River, workers from the management officers must go to each dam to arrange the doors of irrigation and record for the report if the officers who visit each dam and make irrigation reports can be minimized to be more efficient. Therefore, this research aims to help regulate water altitude on dams through a dam monitoring system employing the Internet of Things (IoT). IoT is a structure where objects, people who are provided with an exclusive identity, and the ability to move data over a network without requiring two-way between humans to humans is the source of human purpose or interaction to a computer. The way it works from IoT is that objects must have an Internet Protocol (IP) address. Furthermore, Internet Protocol (IP) addresses in these objects are connected to the Internet network. The idea is expected to be an alternative solution for dam officers in managing irrigation in the dam to raise the officers’ performance in working because of the many times created from this monitoring system. Data collection was conducted through interviews in the hall of Management of River Basin Water Resources (PSDA WS) in Citarum, West Java Province. This research result is the prototype of irrigation monitoring tools to facilitate the modeling of the devices. © 2024 IEEE.

    Author keywords

    dam monitoring system; IoT; regulate water altitude

    This article can be accessed at https://www.scopus.com/pages/publications/105015867292

  • Comparison of SAW and MABAC Methods in Determining Strategic Tourism Destinations with Entropy Weighting Integration

    Authors: Muhlas Adiputra, Yeffry Handoko Putra

    DOI: 10.1109/INCITEST64888.2024.11121470

    Abstract

    Bima Regency, with its natural and cultural diversity, has great potential as a strategic tourism destination that requires proper management. However, the main challenge faced is determining the most potential destinations appropriately and supported by accurate data. In an effort to address the problem, this research utilizes the Entropy method for criteria weighting and applies two multi-criteria decisionmaking methods, Simple Additive Weighting (SAW) and MultiAttributive Border Approximation area Comparison (MABAC), to evaluate their effectiveness in selecting tourist destinations in Bima Regency. The results show that both methods consistently identify Lariti Beach as a superior destination. From the comparison of the two methods, the SAW method provides more stable and consistent results, is suitable for less complex conditions, and requires long-term reliability. Meanwhile, the MABAC method shows higher sensitivity to data changes, with more significant fluctuations in results. Thus, the SAW method is considered more appropriate for use in the selection of strategic tourist destinations in Bima Regency, considering that this method provides consistent results as well as efficiency and ease of application. © 2024 IEEE.

    Author keywords

    determination of strategic tourism destinations; DSS; entropy; MABAC; MCDM; SAW

    This article can be accessed at https://www.scopus.com/pages/publications/105015860453