Farahwahida Mohd
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Farahwahida Mohd has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Siti Afiqah Zainuddin
Universiti Malaysia Kelantan
3 shared publications - Siti Salwa Hasbullah
Centre for Artificial Intelligence Technology, Universiti Kebangsaan Malaysia
3 shared publications - Tahirah Abdullah
Universiti Malaysia Kelantan
3 shared publications - Eko Purwanto
Universiti Kuala Lumpur Malaysian Institute Information Technology
2 shared publications - Eko Purwanto
Universitas Duta Bangsa Surakarta
2 shared publications - Liyana Ahmad Afip
Universiti Malaysia Kelantan
2 shared publications
Latest Publications
- Open Access
- 29 December 2025
- Published by Telecommunications Association Inc. in Journal of Telecommunications and the Digital Economy
- Vol. 13 (4) , 33-52
A Hybrid AI Model Combining RF and SVM for Accurate Sales Prediction in SMEs to Support Digital Transformation
Small and Medium Enterprises (SMEs) are increasingly adopting digital transformation to enhance their competitive advantage and operational efficiency. Accurate sales prediction is crucial for informed decision-making and business strategy optimisation. This research presents a Hybrid AI model, which combines the machine learning techniques of Random Forest (RF) and Support Vector Machine (SVM). Unlike traditional ensemble methods, this hybrid model leverages AI-driven decision-making processes to provide a more intelligent, adaptive, and accurate approach to sales forecasting in SMEs. By integrating RF's ensemble learning power with SVM's robust classification abilities, the model effectively addresses sales prediction challenges in dynamic business environments. The hybrid model was evaluated using key statistical metrics, achieving a Mean Squared Error (MSE) of 0.7254, Mean Absolute Error (MAE) of 0.4605, and an R-squared (R²) value of 0.99999, indicating outstanding predictive accuracy and a near-perfect fit to the data. These results demonstrate the model's capacity to capture complex sales data relationships, proving its practical applicability for sales forecasting. This study highlights the potential of hybrid AI models in advancing sales forecasting techniques, particularly in the context of digital transformation for SMEs.
- 29 December 2025
- Published by Telecommunications Association Inc. in Journal of Telecommunications and the Digital Economy
- Vol. 13 (4) , 33-52
Open AccessA Hybrid AI Model Combining RF and SVM for Accurate Sales Prediction in SMEs to Support Digital Transformation
Small and Medium Enterprises (SMEs) are increasingly adopting digital transformation to enhance their competitive advantage and operational efficiency. Accurate sales prediction is crucial for informed decision-making and business strategy optimisation. This research presents a Hybrid AI model, which combines the machine learning techniques of Random Forest (RF) and Support Vector Machine (SVM). Unlike traditional ensemble methods, this hybrid model leverages AI-driven decision-making processes to provide a more intelligent, adaptive, and accurate approach to sales forecasting in SMEs. By integrating RF's ensemble learning power with SVM's robust classification abilities, the model effectively addresses sales prediction challenges in dynamic business environments. The hybrid model was evaluated using key statistical metrics, achieving a Mean Squared Error (MSE) of 0.7254, Mean Absolute Error (MAE) of 0.4605, and an R-squared (R²) value of 0.99999, indicating outstanding predictive accuracy and a near-perfect fit to the data. These results demonstrate the model's capacity to capture complex sales data relationships, proving its practical applicability for sales forecasting. This study highlights the potential of hybrid AI models in advancing sales forecasting techniques, particularly in the context of digital transformation for SMEs.
- Open Access
- 1 August 2025
- Published by Lembaga KITA in International Journal Software Engineering and Computer Science (IJSECS)
- Vol. 5 (2) , 763-770
Development and Evaluation of an Expert System for the Early Diagnosis of Dental Diseases
Dental diseases, such as caries, periodontitis, and gingivitis, affect public health worldwide, especially in regions where healthcare access remains restricted. The study develops an expert system for early dental disease diagnosis using Forward Chaining and Certainty Factor methods. The system overcomes deficiencies found in previous approaches, such as Naive Bayes and Dempster-Shafer, which demonstrate insufficient accuracy and unclear result interpretation. The developed expert system incorporates a knowledge base containing 7 diseases, 40 symptoms, and 7 diagnostic rules. Forward Chaining enables inference of potential diagnoses from reported symptoms, while the Certainty Factor evaluates diagnostic reliability by calculating confidence levels. System evaluation through Black Box testing achieved 92% diagnostic accuracy, and usability assessments revealed 85% user satisfaction rates, demonstrating that the system proves reliable, accurate, and accessible. Research findings indicate the expert system offers viable solutions for improving dental disease diagnosis in underserved and remote areas, potentially enhancing oral health outcomes through early detection and prompt intervention.
- 1 August 2025
- Published by Lembaga KITA in International Journal Software Engineering and Computer Science (IJSECS)
- Vol. 5 (2) , 763-770
Open AccessDevelopment and Evaluation of an Expert System for the Early Diagnosis of Dental Diseases
Dental diseases, such as caries, periodontitis, and gingivitis, affect public health worldwide, especially in regions where healthcare access remains restricted. The study develops an expert system for early dental disease diagnosis using Forward Chaining and Certainty Factor methods. The system overcomes deficiencies found in previous approaches, such as Naive Bayes and Dempster-Shafer, which demonstrate insufficient accuracy and unclear result interpretation. The developed expert system incorporates a knowledge base containing 7 diseases, 40 symptoms, and 7 diagnostic rules. Forward Chaining enables inference of potential diagnoses from reported symptoms, while the Certainty Factor evaluates diagnostic reliability by calculating confidence levels. System evaluation through Black Box testing achieved 92% diagnostic accuracy, and usability assessments revealed 85% user satisfaction rates, demonstrating that the system proves reliable, accurate, and accessible. Research findings indicate the expert system offers viable solutions for improving dental disease diagnosis in underserved and remote areas, potentially enhancing oral health outcomes through early detection and prompt intervention.
- 11 August 2024
- Published by Springer Nature
- p. 97-103
A Self-Efficacy Sign Language Web Prototype: MUTAPP
- 11 August 2024
- Published by Springer Nature
- p. 97-103
A Self-Efficacy Sign Language Web Prototype: MUTAPP
- 9 August 2024
- Published by Springer Nature
- p. 291-301
Global Research Trend in E-commerce Logistics: A Bibliometric Analysis
- 9 August 2024
- Published by Springer Nature
- p. 291-301
Global Research Trend in E-commerce Logistics: A Bibliometric Analysis
- 25 July 2024
- Published by Springer Nature
- p. 15-22
Factors that Influence Sellers in Selection E-Marketplaces: A Systematic Literature Review
- 25 July 2024
- Published by Springer Nature
- p. 15-22
Factors that Influence Sellers in Selection E-Marketplaces: A Systematic Literature Review
- Open Access
- 7 April 2024
- Published by Narotama University in IJEEIT International Journal of Electrical Engineering and Information Technology
- Vol. 7 (1) , 22-29
Determining the Factors that Influence e-Marketplace Selection from Seller’s Perspective Using a Binary Logistic Regression Model
IJEEIT | International Journal of Electrical Engineering and Information Technology, open submission for research from scope on electrical Enginering and Information Technology
- 7 April 2024
- Published by Narotama University in IJEEIT International Journal of Electrical Engineering and Information Technology
- Vol. 7 (1) , 22-29
Open AccessDetermining the Factors that Influence e-Marketplace Selection from Seller’s Perspective Using a Binary Logistic Regression Model
IJEEIT | International Journal of Electrical Engineering and Information Technology, open submission for research from scope on electrical Enginering and Information Technology
- 1 January 2024
- Published by Springer Nature
- p. 99-104
Review of the Use of Digital Media Products by Small and Medium Enterprises in Malaysia During the Covid-19 Pandemic Crisis
- 1 January 2024
- Published by Springer Nature
- p. 99-104
Review of the Use of Digital Media Products by Small and Medium Enterprises in Malaysia During the Covid-19 Pandemic Crisis
- 17 May 2023
- Published by Springer Nature
- p. 97-103
“Hello Dr” Application for Mobile Devices
- 1 January 2023
- Published by Springer Nature
- p. 715-725
An Application of Actor Network Theory (ANT) in Business-to-Business Procurement Platform
- 1 January 2023
- Published by Springer Nature
- p. 715-725
An Application of Actor Network Theory (ANT) in Business-to-Business Procurement Platform
- 1 January 2023
- Published by Springer Nature
- p. 191-201
Is Your Organization’s Operational Risk Management Practice Up to Snuff? A Multiple-Case Study on Businesses in Kelantan, Malaysia
- 1 January 2023
- Published by Springer Nature
- p. 191-201
Is Your Organization’s Operational Risk Management Practice Up to Snuff? A Multiple-Case Study on Businesses in Kelantan, Malaysia