Volume-14 Issue-6, March 2026 – International Journal of Recent Technology and Engineering (IJRTE) https://www.ijrte.org Exploring Innovation | ISSN:2277-3878(Online) | A Periodical Journal | Reg. No.: C/819981 | Published By BEIESP Mon, 30 Mar 2026 04:20:35 +0000 en-US hourly 1 https://www.ijrte.org/wp-content/uploads/2021/02/cropped-IJRTE-Logo-WS-32x32.png Volume-14 Issue-6, March 2026 – International Journal of Recent Technology and Engineering (IJRTE) https://www.ijrte.org 32 32 A834515010526 https://www.ijrte.org/portfolio-item/a834515010526/ Wed, 25 Mar 2026 12:02:26 +0000 https://www.ijrte.org/?post_type=portfolio&p=50076 Smishing (SMS phishing) is a cyber threat that is growing rapidly, and it’s a tactic in which attackers use SMS messages to deceive users and trick them into revealing their private information or unintentionally installing harmful applications. As mobile devices are used everywhere, detecting smishing messages has become a very important yet difficult task in cybersecurity. In the present study, the authors conduct a comparative analysis of several deep learning models, namely, the Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi LSTM), Convolutional Neural Network (CNN), and a combination of CNN-LSTM, for the detection of smishing. The experiments are conducted on publicly available SMS datasets, and performance is evaluated using accuracy, precision, recall, F1 Score, and a confusion matrix. The findings indicate that deep learning-based approaches yield significantly better results than traditional methods, with hybrid architectures leading in overall performance.

The post A834515010526 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]>

A Comparative Analysis of Deep Learning Models for Smishing Detection in SMS Message
Aqsa Shaikh1, Mariya Shaikh2, Srivaramangai R3

1Aqsa Shaikh, Student, Department of Information Technology, University of Mumbai, Mumbai (Maharashtra), India.

2Mariya Shaikh, Student, Department of Information Technology, University of Mumbai, Mumbai (Maharashtra), India.

3Srivaramangai R., Head, Department of Information Technology, University of Mumbai, Mumbai (Maharashtra), India.

Manuscript received on 01 March 2026 | Revised Manuscript received on 09 March 2026 | Manuscript Accepted on 15 March 2026 | Manuscript published on 30 March 2026 | PP: 14-21 | Volume-14 Issue-6, March 2026 | Retrieval Number: 100.1/ijrte.A834515010526 | DOI: 10.35940/ijrte.A8345.14060326

Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Smishing (SMS phishing) is a cyber threat that is growing rapidly, and it’s a tactic in which attackers use SMS messages to deceive users and trick them into revealing their private information or unintentionally installing harmful applications. As mobile devices are used everywhere, detecting smishing messages has become a very important yet difficult task in cybersecurity. In the present study, the authors conduct a comparative analysis of several deep learning models, namely, the Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi LSTM), Convolutional Neural Network (CNN), and a combination of CNN-LSTM, for the detection of smishing. The experiments are conducted on publicly available SMS datasets, and performance is evaluated using accuracy, precision, recall, F1 Score, and a confusion matrix. The findings indicate that deep learning-based approaches yield significantly better results than traditional methods, with hybrid architectures leading in overall performance.

Keywords: Cybersecurity, Deep Learning, LSTM, Smishing.
Scope of the Article: Computer Science and Engineering

Download Download Download PDF

The post A834515010526 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]> E832914050126 https://www.ijrte.org/portfolio-item/e832914050126/ Wed, 25 Mar 2026 11:50:37 +0000 https://www.ijrte.org/?post_type=portfolio&p=50071 Predicting student performance remains a challenge in many education systems, especially in developing countries like Zambia, where robust predictive tools are scarce. This study shows how data mining methods can be utilised to improve the accuracy of performance prediction by leveraging mock examination results to meet the needs of school management. For this study, a dataset containing 1,170 instances and 17 attributes was constructed and analysed using four classification algorithms (J48, PART, BayesNet, and Random Forest). The findings indicate that although each classifier produced results with high accuracy above 99%, Random Forest performed best, delivering perfect predictions with 100% accuracy. These results emphasise the importance of data mining in generating reliable forecasts of student performance, enabling early detection of at-risk learners and timely interventions by school managers, teachers, and parents. The study recommends adopting Random Forest as the most suitable classifier for predicting student performance. By incorporating predictive analytics into educational management, schools can strengthen decision-making, refine teaching approaches, and ultimately improve learning quality.

The post E832914050126 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]>

Using Data Mining to Predict Secondary School Student Performance for Zambia
Mainess Kandah Namuchile1, Christopher Mulwanda2

1Mainess Kandah Namuchile, Department of Computer Science, School of Engineering and Technology, Mulungushi University, Zambia.

2Dr. Christopher Mulwanda, Researcher, Centre for Environmental Justice, Plot 37741, Pitta Road, off Twin-Palm Road, Ibex Hill, Lusaka, Zambia.

Manuscript received on 29 December 2025 | First Revised Manuscript received on 18 February 2026 | Second Revised Manuscript received on 03 March 2026 | Manuscript Accepted on 15 March 2026 | Manuscript published on 30 March 2026 | PP: 7-13 | Volume-14 Issue-6, March 2026 | Retrieval Number: 100.1/ijrte.E832914050126 | DOI: 10.35940/ijrte.E8329.14060326

Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Predicting student performance remains a challenge in many education systems, especially in developing countries like Zambia, where robust predictive tools are scarce. This study shows how data mining methods can be utilised to improve the accuracy of performance prediction by leveraging mock examination results to meet the needs of school management. For this study, a dataset containing 1,170 instances and 17 attributes was constructed and analysed using four classification algorithms (J48, PART, BayesNet, and Random Forest). The findings indicate that although each classifier produced results with high accuracy above 99%, Random Forest performed best, delivering perfect predictions with 100% accuracy. These results emphasise the importance of data mining in generating reliable forecasts of student performance, enabling early detection of at-risk learners and timely interventions by school managers, teachers, and parents. The study recommends adopting Random Forest as the most suitable classifier for predicting student performance. By incorporating predictive analytics into educational management, schools can strengthen decision-making, refine teaching approaches, and ultimately improve learning quality.

Keywords: Data Mining: Performance Prediction: Classification Algorithms: Random Forest: Zambia Secondary School.
Scope of the Article: Computer Science and Engineering

Download Download Download PDF

The post E832914050126 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]> F834314060326 https://www.ijrte.org/portfolio-item/f834314060326/ Wed, 25 Mar 2026 11:39:12 +0000 https://www.ijrte.org/?post_type=portfolio&p=50067 Ransomware is a rapidly increasing hazard to essential networks, including the health care, finance, energy, and government sectors. Traditional security solutions have shown deficiencies in their ability to rapidly recognise zero-day ransomware attacks. This research project proposes a hybrid artificial intelligence-honeypot framework for proactive detection and mitigation of ransomware within critical infrastructure. Honeypot-based security technologies will be combined with artificial intelligence-based behavioural analysis of attackers to identify potential ransomware signatures at the earliest possible stage. Machine learning algorithms provide continuous estimates of file system interactions, network traffic patterns, and system calls captured in honeypot environments to detect and profile malicious behaviour. This research will contribute to the effectiveness of combining deception-based security measures with AI-based behavioural models, thus enhancing the resiliency of ransomware defence solutions in critical infrastructure.

The post F834314060326 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]>

Surveying Hybrid Intelligence Approaches that Combine Honeypots and AI for Ransomware Defence in Critical Infrastructure
Ibrahim Shaikh1, Omkar Nachare2, Srivaramangai Ramanujam3

1Ibrahim Shaikh, MS. (Cybersecurity) Student, Department of Information Technology, University of Mumbai, Vidyanagri, Kalina, Santacruz, Mumbai, (Maharashtra), India.

2Omkar Nachare, MS. (Cybersecurity) Student, Department of Information Technology, University of Mumbai, Vidyanagri, Kalina, Santacruz, Mumbai, (Maharashtra), India.

3Srivaramangai Ramanujam, Professor, Department of Information Technology, University of Mumbai, Vidyanagri, Kalina, Santacruz, Mumbai, (Maharashtra), India.

Manuscript received on 01 March 2026 | Revised Manuscript received on 09 March 2026 | Manuscript Accepted on 15 March 2026 | Manuscript published on 30 March 2026 | PP: 1-6 | Volume-14 Issue-6, March 2026 | Retrieval Number: 100.1/ijrte.F834314060326 | DOI: 10.35940/ijrte.F8343.14060326

Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Ransomware is a rapidly increasing hazard to essential networks, including the health care, finance, energy, and government sectors. Traditional security solutions have shown deficiencies in their ability to rapidly recognise zero-day ransomware attacks. This research project proposes a hybrid artificial intelligence-honeypot framework for proactive detection and mitigation of ransomware within critical infrastructure. Honeypot-based security technologies will be combined with artificial intelligence-based behavioural analysis of attackers to identify potential ransomware signatures at the earliest possible stage. Machine learning algorithms provide continuous estimates of file system interactions, network traffic patterns, and system calls captured in honeypot environments to detect and profile malicious behaviour. This research will contribute to the effectiveness of combining deception-based security measures with AI-based behavioural models, thus enhancing the resiliency of ransomware defence solutions in critical infrastructure.

Keywords: Ransomware, Honeypot, Artificial Intelligence, Machine Learning, Cybersecurity, Critical Networks
Scope of the Article: Computer Science and Engineering

Download Download Download PDF

The post F834314060326 appeared first on International Journal of Recent Technology and Engineering (IJRTE).

]]>