Nur Atiqah Sia Abdullah

Nur Atiqah Sia Abdullah

Auto-generated profile, based on publicly available information

Nur Atiqah Sia Abdullah

Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA, Shah Alam, Selangor, Malaysia

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 January 2021 in Pertanika Journal of Science and Technology
Verified
Journal Article
Open access
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
NRNur Ida Aniza Rusli
Nur Ida Aniza Rusli
With the explosive growth of social media, the online community can freely express their opinions without disclosing their identities. People with hidden agendas can easily post fake opinions to discredit target products, services, politicians, or organizations. With these big data, monitoring opinions and distilling their sentiments remain a formidable task because of the proliferation of diverse sites with a large volume of opinions that are portrayed in multilingual. Therefore, this paper aims to provide a systematic literature review on multilingual sentiment analysis, which summarises the common languages supported in multilingual sentiment analysis, pre-processing techniques, existing sentiment analysis approaches, and evaluation models that have been used for multilingual sentiment analysis. By following the systematic literature review, the findings revealed, most of the models supported two languages, and English is seen as the most used language in sentiment analysis studies. None of the reviewed literature has catered the combination of languages for English, Chinese, Malay, and Hindi language on multilingual sentiment analysis. The common pre-processing techniques for the multilingual domain are tokenization, normalization, capitalization, N-gram, and machine translation. Meanwhile, the sentiment analysis classification techniques for multilingual sentiment are hybrid sentiment analysis, which includes localized language analysis, unsupervised topic clustering, and then followed by multilingual sentiment analysis. In terms of evaluation, most of the studies used precision, recall, and accuracy as the benchmark for the results.

1 September 2024 in Indonesian Journal of Electrical Engineering and Computer Science
Verified
Journal Article
Open access
PDF file
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
NRNur Ida Aniza Rusli
Nur Ida Aniza Rusli
NYNurshaheeda Shazlin Yuslee
Nurshaheeda Shazlin Yuslee
With the extensive technological advancements and expansion, the persistent issues regarding the creation and rapid dissemination of fake news have become a prevalent and recurrent concern. The manipulation of news content has critical repercussions, such as causing public mistrust, fear, harm, and misinformation. Addressing that, this study developed a supervised machine learning algorithm that can accurately classify social media data as fake news. The methodology of the proposed fake news detection model involved five main components: data acquisition from Twitter, data preprocessing, data transformation, model development using Naïve Bayes, decision tree, and support vector machine (SVM) and model evaluation using accuracy, precision, recall and F1-score. The results revealed that decision tree recorded the highest accuracy for both textual data (100%) and metadata (94.54%) and consistently outperformed both Naïve Bayes and SVM in terms of precision, recall, and F1-score metrics, with a score of 100% for the classification of textual data-based datasets. Regarding the metadata-based classification, decision tree also demonstrated excellent performance, with the highest F1-score of 94% for fake news data. Meanwhile, SVM exhibited the highest precision and recall performance for the metadata-based classification. Overall, the application of the decision tree classifier was deemed the most effective in Twitter fake news detection.

30 November 2024
Verified
Book Chapter
Open access
PDF file
MRMuhamad Ridhwan Mohamad Razali
Muhamad Ridhwan Mohamad Razali
HAHana Fakhira Almarzuki
Hana Fakhira Almarzuki
RBRabiahtul Adawiyah Bahar
Rabiahtul Adawiyah Bahar
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
Every individual requires food for nutritional support, without which the body suffers from various diseases such as gastric issues and low blood sugar. While there are many recipe websites, finding a suitable recipe can be time-consuming, especially for those who lack cooking knowledge with available ingredients. This project addresses this challenge by developing a recipe recommendation model using a deep learning algorithm to suggest recipes based on recognized food ingredients. The model utilizes more than 2,000 images categorized into three food ingredient groups: rice, apple, and chicken. The data collection process includes thorough preprocessing, data cleaning, and annotation. The dataset is then augmented and split into training, validation, and testing sets in an 80–10-10 ratio. To build the food ingredient recognition model, the You Only Look Once (YOLO) technique, specifically YOLOv8, is employed. The system is designed as a web-based application, providing an accessible interface for users. After extensive training, validation, and testing, the model achieved an impressive accuracy of 98%, demonstrating its capability to accurately detect the specified ingredients. Future research aims to enhance the system's functionality by addressing current limitations. Planned improvements include integrating a comprehensive database for storing recipes, which would facilitate easier data insertion and the addition of nutritional information for each recipe. Furthermore, incorporating live-camera integration could enhance user engagement and practicality.

2 August 2024 in International Journal of Interactive Mobile Technologies (iJIM)
Verified
Journal Article
Open access
PDF file
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
MRMarshima Mohd Rosli
Marshima Mohd Rosli
NANur Diyana Abd Aziz
Nur Diyana Abd Aziz
NANOOR ALICEZAH
NOOR ALICEZAH
The recent healthcare transformations emphasize the importance of individuals maintaining a healthy lifestyle through proper nutrition and physical activity to reduce the risk of severe illnesses. Patients often search for information on their own, leading to uncertainty about appropriate diets or fitness activities. Consequently, many individuals cross-check information or health advice from various sources. However, some people hesitate to verify online health-related information with their clinicians, fearing that it may be perceived as a challenge to their expertise and authority. This study aimed to determine a useful way to monitor a patient’s lipid profile and provide recommendations for meal plans and fitness activities. A content-based approach that utilizes a vector space model is employed in the development of a recommender method. The vector space model uses meal plan keywords to suggest similar items, and selection rules are applied to identify relevant meal plan and fitness activity options. This approach has been integrated into a mobile application for healthcare, enabling patients to receive personalized recommendations based on their lipid levels. To assess the usability of the mobile application, an initial user study was conducted, which revealed that most respondents had a positive opinion of the application. In the future, the application could be enhanced with a wider variety of meal plans and additional features.

30 November 2024
Verified
Book Chapter
Open access
PDF file
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
FHFarahnatasyah Abdul Hanan
Farahnatasyah Abdul Hanan
ARAbdul Hakim Abdul Rahman
Abdul Hakim Abdul Rahman
Parkinson’s Disease (PD) is a neurodegenerative disorder characterized by motor impairments, and early detection is crucial for effective intervention. PD patient diagnose the disease only after having serious symptoms and the diagnosis requires huge amount of time. The self-serves diagnostic can be applied with the help of computer aided technology. This paper aims to identify the features of Parkinson's disease symptoms by assessing on the spiral drawing images and develop a Parkinson's disease prediction system using machine learning model. By leveraging the power of machine learning algorithms, ResNet50 and Random Forest with Histogram Oriented Gradient (HOG) models are trained on spiral and waves imaging datasets comprising spiral drawings of healthy individuals and with PD symptoms. Data are pre-processed. After the data is cleaned, the models were trained with different experiments for ResNet50 model, also random state for Random Forest with HOG and to be compared to find the best model with highest accuracy and lowest loss. As the results, the best model is ResNet50 model with 50 epoch and 32 batch size as it has the highest validation accuracy (0.8967) and lowest validation loss (0.245). The ResNet50 Model 2 is chosen to be embedded in the PD detection system, which is being developed using Python. The model can be enhanced further in the future by increasing the number of features on detecting spiral drawings such as speed and pressure of pen when drawing.

30 November 2024
Verified
Book Chapter
Open access
PDF file
HHHafizatul Hanin Hamzah
Hafizatul Hanin Hamzah
WEWan Athirah Wan Endut
Wan Athirah Wan Endut
AAAzmina Mohd Ariffin
Azmina Mohd Ariffin
NANur Atiqah Sia Abdullah
Nur Atiqah Sia Abdullah
Efficient waste management plays a crucial role in ensuring the sustainability of our environment and safeguarding public health. However, the lack of knowledge and awareness of waste management practices among the public will increase waste production and cause an unprecedented environmental crisis. Waste separation is crucial for addressing environmental problems, but the diversity of household and company waste complicates proper separation. This paper aims to develop a user-friendly mobile application to help with waste sorting processes by identifying various waste materials. This waste separation solution includes image detection using Faster Regional Convolutional Neural Networks (Faster R-CNN). The comprehensive study involved meticulously curating and utilizing a dataset of 250 images of five categories, which consist of glass, metal, paper, cardboard, and plastic. After a thorough pre-processing process, the datasets are then used for training data and testing data. The test consists of 7 trials. Remarkably, the outcomes yielded a good confidence value for cardboard and plastic, which are 97.99% and 95.24% respectively. Meanwhile, paper achieved 87.54%, glass achieved 70.24%, and metal achieved 65.88% in confidence value. This means the algorithm can detect and classify successfully the five waste categories. Furthermore, we embedded this best model in an Android application to demonstrate its potential for consumer and industrial usage, underscoring the transformative potential of technology in enhancing waste management practices and fostering environmental sustainability initiatives.