A5007851816- Computational Drug Discovery Methods
- Analytical Chemistry and Chromatography
- Handwritten Text Recognition Techniques
- Spectroscopy and Chemometric Analyses
- Natural Language Processing Techniques
- Molecular spectroscopy and chirality
- Metabolomics and Mass Spectrometry Studies
- Advanced Chemical Sensor Technologies
- Computational Physics and Python Applications
- Metaheuristic Optimization Algorithms Research
- Text and Document Classification Technologies
- Spam and Phishing Detection
- Image Retrieval and Classification Techniques
- Receptor Mechanisms and Signaling
- Cell Image Analysis Techniques
- Machine Learning in Bioinformatics
- Vehicle License Plate Recognition
- Machine Learning in Materials Science
- Numerical Methods and Algorithms
- Information Retrieval and Data Mining
- Pressure Ulcer Prevention and Management
- Multimedia Learning Systems
- Identification and Quantification in Food
- Data Mining and Machine Learning Applications
- Authorship Attribution and Profiling
Technical University of Malaysia Malacca
2011-2023
Sheffield Hallam University
2022-2023
Intel (Malaysia)
2013-2018
Purpose: The purpose of this study is to analyze text documents from Twitter about public policies in handling COVID-19 that are currently or have been determined. classified into positive and negative sentiments by using Multinomial Naive Bayes.Methods: In research, CRISP-DM used as a method for conducting sentiment analysis, starting the business understanding process, data understanding, preparation, modeling, evaluation. Bayes has applied building classification based on documents....
The uniqueness of shape and style handwriting can be used to identify the significant features in confirming author writing. Acquiring these leads an important research Writer Identification domain. This paper is meant explore usage feature selection order find most features. proposes a hybrid method Particle Swarm Optimization Computationally Inexpensive Sequential Forward Floating Selection for Identification. promising applicability proposed has been demonstrated worth receive further...
Swarm intelligence-based feature selection techniques are implemented by this work to increase classifier performance in classifying Amphetamine-type Stimulants (ATS) drugs. A recently proposed 3D Exact Legendre Moment Invariants (3D-ELMI) molecular descriptors as structure representational for ATS These utilized the dataset study. However, a large number of may cause degradation classifier. To complement issue, research applies three swarm algorithms with k-Nearest Neighbor (k-NN) wrapper...
Similarity searching of molecular structure has been an important application in the Chemoinformatics, especially drug discovery. is a common method used for identification structure. It involve three main principal component similarity searching: representation; weighting scheme; and coefficient. In this paper, we introduces Weighted Tanimoto Coefficient based on weighted Euclidean distance order to investigate effect weight function result searching. The coefficient one popular...
The uniqueness of shape and style handwriting can be used to identify the significant features in confirming author writing. This paper is meant propose a novel feature selection framework for Swarm Optimized Computationally Inexpensive Floating Selection (SOCIFS), by exploring exi sting frameworks, compare performance proposed against various methods Writer Identification order find most features. promising applicability has been demonstrated result worth receive further exploration...
Drug abuse among the people around world is getting serious and increase nowadays. An Amphetamine-Type Stimulant (ATS) drug one of popular drugs in world. This kind a combination two types drugs; amphetamine ecstasy. In this project, we are focusing on different techniques used to compare result between 2D 3D ATS molecule. Besides that, project also looking for best technique produce images. United Moment Invariant (UMI) Hough Transform (HT) feature extraction methods that will be extract...
Researchers are often perplexed when their machine learning algorithms required to deal with complex number. Various strategies commonly employed project number into real number, although it is frequently sacrificing the information contained in This paper proposes a new method and four techniques represent as without having sacrifice contained. The proposed also capable of retrieving original from representing little none loss. promising applicability has been demonstrated worth receive...
The uniqueness of shape and style handwriting can be used to identify the significant features in confirming author writing. Acquiring these leads an important research Writer Identification (WI) domain. This paper is meant explore usage improved discretization method alternative Cheap Computational Cost Class-Specific Swarm Sequential Selection (C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> S ) WI framework for Optimized...