Journal of Informatics and Web Engineering

Editorial Preview

Su-Cheng Haw — 2024-02-14

This editorial highlights all 19 papers in the February issue that deal with the practical aspects of Machine Learning (ML), Artificial Intelligence (AI), Data Mining (DM), the Internet of Things (IoT), and other topics in Computer Science. This issue also includes suggestions for several worthwhile works that deserve further research. With effective from this volume, we will be publishing triannually in our February, June and October issues.

Term Standardisation With LDA Model To Detect Service Disruption Events Using English And Manglish Tweets

Noraysha Yusuf — 2024-02-14

Rapid transit is one of Malaysia's most important transportation modes, where commuters use public transportation to travel. Any disruption in the rapid transit service affects their daily routines. Therefore, detecting such service disruption has become fundamental. In this study, the disruption in Malaysia's rapid transit service was assessed using English and Manglish (a combination of English and Malay) tweets through Latent Dirichlet Allocation (LDA). The gathered tweets were classified into event and non-event tweets and LDA was applied to the event tweets. Manglish event tweets were pre-processed using the proposed term standardisation technique. As a result, LDA has proved its efficiency in topic detection for both English and Manglish tweets with better performance for Manglish tweets; The best event detection rate of the LDA_English model was at the likelihood of 80% while the best detection rate of the LDA_Manglish model was at a likelihood of 60%.

Modelling of Virtual Campus Tour in Minecraft

Liyana Tan Lin — 2024-02-14

Virtual tours have revolutionized the way to explore and experience places from the comfort of our own home. Through advanced technology and immersive digital platforms, virtual tours offer a compelling alternative to tradition face-to-face visits. Whether a famous landmark, museum, real estate or natural wonders, virtual tours offer a unique opportunity to navigate and discover these places form a distance. Meanwhile, creating a virtual tour in Minecraft can provide a unique and immersive experience that sets the users apart from other virtual tour platforms. Minecraft is one of the most popular video games in the world and boasts a large and dedicated community of players. Using Minecraft for a virtual tour allow users to reach a larger audience who are already familiar with the game, increasing the likelihood of engagement and participation. In this paper, the aim is to create a virtual campus tour in Minecraft to give the visitors an immersive and interactive experience with creative freedom. A series of buildings have been built such as Siti Hasmah Digital Library, Common Lecture Complex (CLC) and Smart Lab. Visitors can move around the campus with some gameplay mechanics using mouse and keyboard. Building information was also integrated so visitors can see details about each building during the virtual tour. The virtual tour provides access, comfort and a sense of connection to prospective students, their families and international visitors. Additionally, it serves as a low-cost marketing tool that increases engagement, attracts potential students, researchers and staff and ultimately benefits the University’s recruitment efforts.

A Lung Cancer Detection with Pre-Trained CNN Models

Chai Chee Chiet — 2024-02-14

Lung cancer is a common cancer in Malaysia, affecting the majority of male citizens. The early detection of lung cancer will decrease its death rate. The only way to detect lung cancer is with a CT scan, and it also requires the doctor to check the scan to confirm the disease. In another way, the computer's support for the detection and diagnosis tool will assist doctors in determining lung cancer more accurately and efficiently. There are three main objectives for this research work. The first target is to study state-of-the-art research work to detect and recognize lung cancer from CT scan images. Then, the article will aim to adopt pre-trained convolutional neural network models in lung cancer detection. It also evaluates the performance of convolutional models on lung cancer imagery data. Then, the pre-trained models with a few added layers and modifications to parameters such as epochs, batch size, optimizer, etc. to conduct model training in this article. After that, Python Pylidc is used in image pre-processing to filter the dataset. Overall, pre-trained models such as ResNet-50, VGG-16, Xception, and MobileNet achieve above-state-of-the-art performance in classifying lung cancer from CT scan images in the range of 78% to 86% accuracy. The best detection accuracy result is the pre-trained VGG-16 model with the addition of some fully connected layers, 16 batch sizes, and the Adam optimizer, which achieved 86.71%.

Optimizing Medical IoT Disaster Management with Data Compression

Nunudzai Mrewa — 2024-02-14

In today's technological landscape, the convergence of the Internet of Things (IoT) with various industries showcases the march of progress. This coming together involves combining diverse data streams from different sources and transmitting processed data in real-time. This empowers stakeholders to make quick and informed decisions, especially in areas like smart cities, healthcare, and industrial automation, where efficiency gains are evident. However, with this convergence comes a challenge – the large amount of data generated by IoT devices. This data overload makes processing and transmitting information efficiently a significant hurdle, potentially undermining the benefits of this union. To tackle this issue, we introduce "Beyond Orion," a novel lossless compression method designed to optimize data compression in IoT systems. The algorithm employs advanced techniques such as Lempel Ziv-Welch and Huffman encoding, while also integrating strategies like pipelining, parallelism, and serialization for greater efficiency and lower resource usage. Through rigorous experimentation, we demonstrate the effectiveness of Beyond Orion. The results show impressive data reduction, with up to 99% across various datasets, and compression factors as high as 7694.13. Comparative tests highlight the algorithm's prowess, achieving savings of 72% and compression factor of 3.53. These findings have significant implications. They promise improved data handling, more effective decision-making, and optimized resource allocation across a range of IoT applications. By addressing the challenge of data volume, Beyond Orion emerges as a significant advancement in IoT data management, marking a substantial step towards realizing the full potential of IoT technology.

Sentiment Analysis using Support Vector Machine and Random Forest

Talha Ahmed Khan — 2024-02-14

Sentiment analysis, is commonly known as opinion mining, is a vital field in natural language processing (NLP) that claims to find out the sentiment or emotion expressed in a given text. This research paper demonstrates an exhaustive survey of sentiment analysis, focusing on the application of machine learning techniques. Comprehensive parametric literature review has been completed to determine the sentiment analysis using SVM and Random Forest. Additionally, the paper covers preprocessing techniques, feature extraction, model training, evaluation, and challenges encountered in sentiment analysis. The findings of this research contribute to a deeper understanding of sentiment analysis and provide insights into the effectiveness of machine learning approaches in this domain. Based on the results obtained, two machine learning algorithms named as Random Forest and SVM were evaluated based on their accuracy in a classification task. The Random Forest algorithm achieved an accuracy of 0.78564, while SVM outperformed it slightly with an accuracy of 0.80394. Both Random Forest and SVM have demonstrated their strengths in achieving respectable accuracies in the given classification task. These results suggest that SVM, with its slightly higher accuracy of 0.80394, may be a more suitable choice when accuracy is the primary concern. However, the basic configuration need and characteristics of the problem at hand should be considered when choosing the better algorithm with better results.

Implementation of Grover’s Algorithm & Bernstein-Vazirani Algorithm with IBM Qiskit

Yang-Che Liu — 2024-02-14

Quantum logic gates differ from classical logic gates as the former involves quantum operators. The conventional gates such as AND, OR, NOT etc., are generally classified as classical gates, however, some of the quantum gates are known as Pauli gates, Toffoli gates and Hadamard gates, respectively. Normally classical states only involve 0 and 1, whereas quantum states involve the superpositions of 0 and 1. Hence, underlying principles of algorithm implementation for classical logic gate and quantum logic gate are indeed different. In this paper, we introduce significant concepts of quantum computations, analyse the discrepancy between classical and quantum gates, compare quantum algorithms using Qiskit against equivalent classical algorithms and analyse their performance in terms of runtime.

A Campus-based Chatbot System using Natural Language Processing and Neural Network

Tuan-Jun Goh — 2024-02-14

A chatbot is designed to simulate human conversation and provide instant responses to users. Chatbots have gained popularity in providing automated customer support and information retrieval among organisations. Besides, it also acts as a virtual assistant to communicate with users by delivering updated answers based on users' input. Most chatbots still use the traditional rule-based chatbot, which can only respond to pre-defined sentences, making the users unlikely to use the chatbot. This paper aims to design and build a campus chatbot for the Faculty of Information Science & Technology (FIST) of Multimedia University that facilitates the study life of FIST students. Before the FIST chatbot can be used, natural language processing techniques such as tokenisation, lemmatisation and bag of word model are used to generate the input that can be used to train the neural network model (multilayer perceptron model). It makes the FIST chatbot comprehends user intent by analysing their questions, enabling it to address a broader range of inquiries and cater to the student's need with accurate answers or information related to the Faculty of Information Science & Technology. Besides, we also developed the backend interface allowing the admin to add and edit the dataset in the proposed chatbot and enable it continuously responds to the student with the latest and updated information.

Personalized Healthcare: A Comprehensive Approach for Symptom Diagnosis and Hospital Recommendations Using AI and Location Services

Seng-Keong Tan — 2024-02-14

Utilizing digital advancements, an integrated Flask-based platform has been engineered to centralize personal health records and facilitate informed healthcare decisions. The platform utilizes a Random Forest model-based symptom checker and an OpenAI API-powered chatbot for preliminary disease diagnosis and integrates Google Maps API to recommend proximal hospitals based on user location. Additionally, it contains a comprehensive user profile encompassing general information, medical history, and allergies. The system includes a medicine reminder feature for medication adherence. This innovative amalgamation of technology and healthcare fosters a user-centric approach to personal health management.

Vision-Based Gait Analysis for Neurodegenerative Disorders Detection

Vincent Wei Sheng Tan — 2024-02-14

Parkinson’s Disease (PD) is a debilitating neurodegenerative disorder that affects a significant portion of aging population. Early detection of PD symptoms is crucial to prevent the progression of the disease. Research has revealed that gait attributes can provide valuable insights into PD symptoms. The gait acquisition techniques used in current research can be broadly divided into two categories: vision-based and sensor-based. The markerless vision-based classification model has become a prominent research trend due to its simplicity, low cost and patient comfort. In this study, we propose a novel markerless vision-based approach to obtain gait features from participants' gait videos. A dataset containing gait videos from normal subjects and PD patients were collected, along with a control group of 25 healthy adults. The participants were requested to perform a Timed Up and Go (TUG) test, during which their walking sequences were recorded using two smartphones positioned at different angles, namely side and front. A multi-person pose estimator is used to estimate human skeletal joint points from the collected gait videos. Different gait features associated with PD including stride length, number of steps taken during turn, turning duration, speed and cadence are derived from these key point information to perform PD detection. Experimental results show that the proposed solution achieves an accuracy of 89.39%. The study's findings demonstrate the potential of markerless vision-based gait acquisition techniques for early detection of PD symptoms.

Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study

Pei-Wern Chin — 2024-02-14

The presence issue of inaccurate plant disease detection persists under real field conditions and most deep learning (DL) techniques still struggle to achieve real-time performance. Hence, challenges in choosing a suitable deep-learning technique to tackle the problem should be addressed. Plant diseases have a detrimental effect on agricultural yield, hence early detection is crucial to prevent food insecurity. To identify and categorise the indications of plant diseases, numerous developed or modified DL architectures are utilised. This paper aims to observe the performance of the YOLOv8 model, which has better performance than its predecessors, on a small-scale plant disease dataset. This paper also aims to improve the accuracy and efficiency of plant disease detection and classification methods by proposing an optimised and lightweight YOLOv8 architecture model. It trains the YOLOv8 model on a public dataset and optimises the YOLOv8 algorithm with the integration of the GhostNet module into the backbone architecture to cut down the number of parameters for a faster computational algorithm. In addition, the architecture incorporates a Coordinate Attention (CA) mechanism module, which further enhances the accuracy of the proposed algorithm. Our results demonstrate that the combination of YOLOv8s with CA mechanism and transfer learning obtained the best result, yielding score of 72.2% which surpassed the studies that utilised the same dataset. Without transfer learning, our best result is demonstrated by YOLOv8s with GhostNet and CA mechanism yielding a score of 69.3%.

Adaptive Gaussian Wiener Filter for CT-Scan Images with Gaussian Noise Variance

Kai Liang Lew — 2024-02-14

Medical imaging plays an important role in modern healthcare, with Computed Tomography (CT) being essential for high-resolution cross-sectional imaging. However, Gaussian noise often occurs within the CT scan images and makes it difficult for image interpretation and reduces the diagnostic accuracy, creating a significant obstacle to fully utilizing CT scanning technology. Existing denoising techniques have a hard time balance between noise reduction and preserving the important image details, failing to enable the optimal diagnostic precision. This study introduces Adaptive Gaussian Wiener Filter (AGWF), a novel filter aims to denoise CT scan images that have been corrupted with various Gaussian noise variance without compromising the image details. The AGWF combines the Gaussian filter for initial noise reduction, followed by the implementation of Wiener filter, which can adaptively estimate noise variance and signal power in localized regions. This approach not only outperforms other existing techniques but also showcases a remarkable balance between noise reduction and image detail preservation. The experiment evaluates 300 images from the dataset and each image is corrupted with Gaussian noise variance to ensure a comprehensive evaluation of the AGWF’s performance. The evaluation indicated that AGWF can improve the Signal-to-Noise Ratio (SNR) value and reduce the Root Mean Square Error (RMSE) and Mean Square Error (MSE) value, showing a qualitative improvement in CT scan imagery. The proposed method holds promising potential for advancing medical imaging technology with the implementation of deep learning.

Comparison of Machine Learning Methods for Calories Burn Prediction

Alfred Tan Jing Sheng — 2024-02-14

This paper focuses on the prediction of calories burned during exercise using machine learning techniques. Due to a growing number of obesity and overweight people, a healthy lifestyle must be adopted and maintained. This study explores and compares several machine learning regression models namely LightGBM, XGBoost, Random Forest, Ridge, Linear, Lasso, and Logistic to assess their calories burned prediction performance that can be used in systems such as fitness recommender systems supporting a healthy lifestyle. Our findings show that the LightGBM for predicting calorie burn has a good accuracy of 1.27 mean absolute error, giving users reliable recommendations. The proposed system has a good potential in assisting users in reaching their fitness objectives by offering precise and tailored advice.

GoHoliday: Development of An Improvised Mobile Application for Boutique Hotels and Resorts

Iftiaj Alom — 2024-02-14

One of the main challenges boutique hotels and resorts face is the direct outreach to tourists and customers. As a result, these independent hotels often resort to online platforms such as Agoda and Airbnb to expand their customer base. However, this approach comes at the cost of losing revenue to Online Travel Agencies (OTAs) that solely focus on room sales, hindering the establishment of a strong brand image for boutique hotels and resorts. Considering the heavy reliance on OTAs, this paper focuses on the development of GoHoliday, a cross-platform mobile app prototype that aims to bridge the gap between boutique hotels and users. This mobile application seamlessly integrates a booking engine, an AI assistant for trip planning, and an experience-sharing platform, enhancing the app's capabilities alongside other features. By implementing the GoHoliday mobile application, boutique hotels can maximize their reach and establish a distinct brand identity by directly serving their valuable guests with more personalized arrangements.

Weather-Based Arthritis Tracking: A Mobile Mechanism for Preventive Strategies

Jin-Lun Goh — 2024-02-14

Arthritis is a common joint disorder characterised by symptoms such as swelling, pain, stiffness, and limited joint movement. It primarily affects older individuals, women, and athletes. The advent of information technology has created opportunities for patients to manage their health conditions more effectively. Research indicates that weather can affect arthritis symptoms, with many patients experiencing severe discomfort during rainy weather due to the expansion of already inflamed tissues. However, there is currently no mobile application mechanism available that combines weather forecasting with health recommendations for arthritis patients, which means that patients may not have access to important information that could help them manage their symptoms. Furthermore, few research workflows have focused on weather conditions in online arthritis treatment systems. This research aims to develop a weather-based mobile system for arthritis tracking that provides health advice and alerts based on current and forecast weather conditions, as well as features to help patients track how weather affects their arthritis. This system utilises several tools for its development. The Flutter Framework is used for creating mobile apps, while Firebase is chosen as the cloud-hosted database. Visual Studio Code and Android Studio are utilised as the code editor and Android emulator, respectively. Information about weather forecasts is retrieved via the OpenWeather API. The application mechanism will feature a user-friendly interface to help users stay updated on weather forecasts, and it will collect data in a reliable and user-centric manner for generating robust evidence on health outcomes.

Emojis And Miscommunication in Text-Based Interactions Among Nigerian Youths

Uduak Udoudom — 2024-02-14

This paper explores the dynamic role of emojis in text-based communication among Nigerian youths and the potential implications for miscommunication. Emojis have become integral to contemporary digital conversations, offering users a visual means of expressing emotions, tone, and context within the constraints of text-based interactions. In the context of Nigeria, a country with a diverse linguistic landscape and a youthful population heavily engaged in online communication, understanding the impact of emojis on interpersonal exchanges becomes particularly pertinent. This paper examines the prevalence and patterns of emoji usage among Nigerian youths across various digital platforms. It investigates the cultural nuances and interpretations associated with emojis within the Nigerian context, considering factors such as regional differences, linguistic diversity, and socio-cultural influences. Furthermore, the study examines instances where emojis may contribute to miscommunication or misunderstanding, potentially exacerbating conflicts or hindering effective communication. Through a comprehensive review of existing literature, online discourse analysis, and case studies, this research aims to shed light on the ways in which emojis influence the interpretation of textual messages and the potential challenges they pose to clear and accurate communication. The study concludes that as digital communication continues to be a primary mode of interaction, it is essential for users to recognize the potential for misinterpretation, prompting the need for increased emoji literacy and awareness.

Prediction of Student’s Academic Performance through Data Mining Approach

Muhammad Mubashar Hussain — 2024-02-14

The universities and institutes produce a large amount of student data that can be used in a disciplinary way and useful information can be extracted by using an automated approach. Educational Data Mining (EDM) is an emerging discipline used in the educational environment to deal with big student data and extract useful information. The data mining of students’ data can help the At-risk students as well as the stakeholders by the early warning. This study aims to predict the performance of the students based on student-related data to increase the overall performance. In existing studies, insufficient attributes and complexity of network models is a problem. The student’s current records and grades need to be analyzed. In this approach, the Levenberg Marquardt Algorithm (MLA) deep learning algorithm is used. The data consists of the class test, attendance, assignment and midterm scores. The neural network model consists of four input variables, three hidden and one output layer. The performance of the deep neural network is evaluated by accuracy, precision, recall and F1 score. The proposed model gained a higher accuracy of 88.6% than existing studies. The study successfully predicts the student's final grades using current academic records. This research will be beneficial to the students, educators and educational authorities as a whole.

Hybrid Crow Search and RBFNN: A Novel Approach to Medical Data Classification

Marai Ali — 2024-02-14

The Radial Basis Function Neural Network (RBFNN) is frequently employed in artificial neural networks for diverse classification tasks, yet it encounters certain limitations, including issues related to network latency and local minima. To tackle these challenges, researchers have explored various algorithms to enhance learning performance and alleviate local minima problems. This study introduces a novel approach that integrates the Crow Search Algorithm (CSA) with RBFNN to augment the learning process and address the local minima issue associated with RBFNN. The study evaluates the performance of this innovative model by comparing it to state-of-the-art models like Flower-pollination-RBNN (FP-NN), Artificial Neural Network (ANN), and the conventional RBFNN. To assess the efficacy of the proposed model, the study employs specific datasets, such as the Breast Cancer and Thyroid Disease datasets from the UCI Machine Repository. The simulation results illustrate that the proposed model surpasses other models in terms of accuracy, exhibiting lower Mean Squared Error (MSE) and Mean Absolute Error (MAE) values. Specifically, for the Breast Cancer dataset, the proposed model attains an accuracy of 99.9693%, MSE of 0.000307024, and MAE of 0.00789449. Likewise, for the Thyroid Disease dataset, the proposed model achieves an accuracy of 99.9535%, along with MSE of 0.000464932 and MAE of 0.0057098. For the diabetes dataset, the proposed model demonstrates an accuracy of 98.8073%, MSE of 0.003024, and MAE of 0.009449. In summary, this analysis underscores the enhanced accuracy and effectiveness of the proposed model when compared to traditional approaches.

Electric Vehicle Health Monitoring with Electric Vehicle Range Prediction and Route Planning

Jayapradha Jayaram — 2024-02-14

The automotive industry is experiencing a revolutionary wave due to the rapid spread of electric vehicles (EVs), which is paving the way for a fundamental and long-lasting revolution in the way we approach transportation. The global movement to reduce greenhouse gas emissions and lessen the environmental impact of traditional internal combustion engine vehicles has seen a significant boost in the popularity of electric vehicles as people come together to support environmentally conscious and sustainable mobility solutions. But the ecology surrounding electric vehicles must continue to flourish if the particular problems that EVs present are to be successfully addressed. Chief among these are the formidable foes of range anxiety and battery health management. Range anxiety is a real issue felt by many potential EV owners worry about becoming stuck because their battery has run out before reaching their destination. This psychological barrier is very noticeable and makes present and future EV owners doubtful. In addition, the longevity and health of EV batteries are essential to their continued effectiveness and affordability. The driving range and operating efficiency of the vehicle are directly affected by the gradual degradation of the battery due to several factors like aging, charging patterns, and temperature. This research presents an integrative and holistic approach to address these pressing issues, enhancing and elevating the whole EV ownership experience by combining Electric Vehicle Health Monitoring (EVHM) with Electric Vehicle Range Prediction (EVRP) and Route Planning (EVRP). Combining these three essential elements creates an all-encompassing plan created to not only lessen these enormous obstacles but also accelerate the switch to electric vehicles by giving consumers the knowledge and assurance they require for a smooth, eco-friendly, and sustainable mobility in the future.

An In-Depth Analysis on Efficiency and Vulnerabilities on a Cloud-Based Searchable Symmetric Encryption Solution

Prithvi Chaudhari — 2024-02-14

Searchable Symmetric Encryption (SSE) has come to be as an integral cryptographic approach in a world where digital privacy is essential. The capacity to search through encrypted data whilst maintaining its integrity meets the most important demand for security and confidentiality in a society that is increasingly dependent on cloud-based services and data storage. SSE offers efficient processing of queries over encrypted datasets, allowing entities to comply with data privacy rules while preserving database usability. Our research goes into this need, concentrating on the development and thorough testing of an SSE system based on Curtmola’s architecture and employing Advanced Encryption Standard (AES) in Cypher Block Chaining (CBC) mode. A primary goal of the research is to conduct a thorough evaluation of the security and performance of the system. In order to assess search performance, a variety of database settings were extensively tested, and the system's security was tested by simulating intricate threat scenarios such as count attacks and leakage abuse. The efficiency of operation and cryptographic robustness of the SSE system are critically examined by these reviews.