Skip to content
Teesta University, Modern More, R. K. Road, Rangpur-5404, Bangladesh +8801716504535
Department of Computer Science and Engineering | Teesta University Internal Assessment
CSE homePublications

Publications

Journal

A Controlled Cross-Language Performance Evaluation of Merge Sort under Diverse Data Types and Input Patterns

Md Sydul Islam, Pranto Halder, Md. Serazul Islam Saykat.

Abstract

Abstract
aim/background: This study evaluates the execution-time performance of iterative bottom-up Merge Sort across C, C++, Go, Java, PHP, and Python under controlled and identical algorithmic conditions. It investigates how programming-language design, runtime behavior, numeric data type, and input ordering influence practical performance despite Merge Sort maintaining O(nlogn) complexity across all cases.

methods: Identical Merge Sort logic and identical input files were used for all six languages. Twelve experimental conditions were created from four numeric datasets—positive floating-point, negative floating-point, positive integer, and negative integer—and three input patterns: ascending, descending, and random. Each dataset contained 100,000 elements. Execution time was recorded in milliseconds, and overall mean, standard deviation, 95% confidence interval, and efficiency rank were calculated across the twelve conditions for each language.

results: Go achieved the lowest overall mean execution time at 11.50 ms, followed by C at 19.83 ms, C++ at 22.75 ms, and Java at 23.58 ms. PHP and Python were substantially slower, with mean execution times of 156.33 ms and 556.75 ms, respectively. Random inputs generally produced slightly higher execution times, particularly in PHP and Python. The reported 0 ms C++ timings for floating-point datasets were interpreted as timer-resolution or compiler-optimization artifacts rather than literal zero execution time.

conclusion: The findings show that programming-language execution models, memory representation, compiler behavior, and runtime overhead can substantially affect measured Merge Sort performance even when the algorithm and datasets are held constant. Lower-overhead compiled languages performed more efficiently in this experiment, while dynamically typed and interpreted environments incurred greater runtime costs. Because each condition was measured only once, the findings should be interpreted as descriptive benchmark results rather than definitive statistical evidence.
Key words: Merge Sort, Performance Evaluation, Programming Languages, Data Types, Input Patterns, Execution Time, Algorithm Optimization, Runtime Efficiency Introduction

DOI Open paper PDF

Risk-Bounded Temporal Graph Decision Learning for DDoS Mitigation in Software-Defined Internet of Things Networks

Pranto Halder

Abstract

Abstract

Distributed denial-of-service (DDoS) attacks remain a major availability threat in software-defined Internet of Things (SDN-IoT) networks because the controller must both recognize abnormal traffic and choose a mitigation action without causing unnecessary service disruption. This paper presents a risk-bounded temporal graph decision framework that converts controller-visible telemetry into a history-aware graph state, scores four candidate actions (ALLOW, RATE_LIMIT, BLOCK, and ISOLATE_SUBNET), filters candidates through predicted benign-loss, controller-cost, and risk constraints, and records decision provenance in a permissioned hash-linked audit log. The archived temporal-transition dataset contains 1282 one-step transitions with an eight-window history and a contiguous source-time split of 910/196/176 train/validation/test samples. The original action-selection replay contains 130 decision steps: the temporal model with the risk selector reaches 0.5077 accuracy and 0.6667 F1-score, compared with 0.9385 accuracy and 0.9344 F1-score for the strongest observable baselines, and it collapses toward RATE_LIMIT in 128/130 decisions. These negative results are analyzed rather than hidden. Inspection of the archived transition pipeline also shows incomplete factual support for the four-action space because the source stress logs record operational status rather than an explicit four-action treatment label. Consequently, the learned temporal policy is retained as shadow-mode feasibility evidence rather than an autonomous controller. Supporting controlled validation adds a timestamped temporal detector split, topology-scale tests, five repeated live/stress runs at a common threshold, benign flash-crowd testing, and audit-overhead measurements. Zero unsafe-action and controller-overload events in the 130-step replay are, therefore, reported only as empirical constraint satisfaction under the evaluated conditions, not as a safety guarantee. The contribution is an auditable, nonleaky, constraint-filtered research architecture and a transparent diagnosis of the remaining adaptive-policy learning problem.

DOI Open paper PDF

Optimizing Capsule Endoscopy via (SHAP) Perturbations and Optimal Data Distribution for Pancreatic Disease Classification

Anurag Sinha, Pranto Halder, Aditya Pandey, Avi Mohan Kumar Shukla, Shravan Kumar, Ashutosh Rastogi, Asima Akter Chowdhury, Suryansh Rai, Sagar Singh

Abstract

Abstract
Accurate classification of pancreatic endocrinogenesis-related abnormalities in capsule endoscopy images remains challenging because of class imbalance, high intra-class variability, and limited annotated data. This study proposes an optimal data distribution framework based on Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) perturbations to enhance automated classification performance. The workflow combines perturbation-guided data augmentation, feature-importance analysis, and deep-learning classifiers, including convolutional neural networks (CNNs), U-Net, and You Only Look Once (YOLO). The approach is evaluated on the Kvasir-CapsuleSeg dataset using accuracy, macro F1-score, balanced area under the receiver operating characteristic curve (AUC), and confusion-matrix analysis. Results indicate that perturbation-guided augmentation improves robustness and class balance, with consistent gains across models and particularly strong performance for underrepresented classes. These findings suggest that explainability-driven data optimization can improve automated capsule endoscopy analysis and support more interpretable pancreatic disease classification.
Keywords
Optimal data distribution; local interpretable model-agnostic explanations; SHapley Additive exPlanations; pancreatic disease classification; capsule endoscopy; explainable artificial intelligence; image analysis

DOI Open paper PDF

Quantum-classical hybrid learning framework for Parkinson’s disease prediction using acoustic and clinical features

Jiahua Li, N. Rama Rao, C. H. N. Santhosh Kumar, G. Madhukar Rao, Pranto Halder, Shakti Singh & Doğan Keskin

Abstract

Abstract

Parkinson’s disease (PD) is a progressive neurodegenerative disorder affecting millions worldwide, where early and accurate diagnosis plays a critical role in improving patient outcomes. While classical machine learning methods have shown strong performance in PD classification using clinical and acoustic data, they remain limited in capturing complex nonlinear feature interactions in high-dimensional spaces. This study explores the application of quantum machine learning (QML) techniques for PD prediction, focusing on Variational Quantum Classifiers (VQC) and Quantum Support Vector Machines (QSVM), and compares their performance with strong classical baselines including Logistic Regression, SVM, Random Forest, Grid Search–optimized models, and XGBoost. Experiments were conducted using stratified train–test splits and validated through cross-validation and statistical testing. Results show that classical models achieve competitive performance, with XGBoost reaching 91.9% accuracy. However, quantum models demonstrate slightly improved performance, with the VQC using EfficientSU2 achieving the highest accuracy of 93.1%, followed by the Real Amplitudes VQC (92.6%) and QSVM (92.2%). These improvements are consistent across evaluation metrics including F1-score and recall. Statistical analysis using 10-fold cross-validation indicates that the VQC (EfficientSU2) shows a significant improvement over classical baselines (p = 0.031), while other quantum models do not achieve statistical significance over XGBoost (p > 0.05). McNemar’s test on the independent test set confirms no significant difference in final prediction distributions between quantum and classical models. Overall, the findings suggest that quantum machine learning models are competitive with state-of-the-art classical approaches, offering marginal but consistent performance gains. This study highlights the potential of quantum-enhanced learning frameworks for biomedical classification tasks, while also emphasizing that their advantages remain incremental rather than definitive under current simulation-based implementations.

DOI Open paper PDF

Quantum Machine Learning and Blockchain Integration for Enhanced Security and Efficiency in UGU, UAV, and Drone Networks: A Comprehensive Review

Halder, P. & Islam, M. S.

Abstract

Abstract
The rapid evolution of autonomous systems—such as Unmanned Ground Units (UGUs), Unmanned Aerial Vehicles (UAVs), and drone networks—has generated unprecedented demands for secure, intelligent, and efficient data processing. This study presents a comprehensive review of recent advancements integrating Quantum Machine Learning (QML) and blockchain technologies to address these challenges. Following a systematic review approach, relevant studies published between 2020 and 2024 were analyzed to evaluate their contributions to enhancing decision-making, communication reliability, and data protection in autonomous networks. The findings reveal that QML enables adaptive learning and optimization in dynamic environments, while blockchain ensures data immutability, decentralized trust, and robust security. Together, these technologies offer a promising foundation for next-generation autonomous systems characterized by transparency, resilience, and computational efficiency. The paper concludes by outlining open research issues, including scalability, interoperability, and quantum resource constraints, and highlights future research opportunities for developing unified QML–blockchain frameworks in autonomous networks.
Key words: Keywords: Quantum Machine Learning (QML), Blockchain, Unmanned Ground Units (UGUs), Unmanned Aerial Vehicles (UAVs), Drones, Autonomous Networks, Data Security, Decentralized Systems, Artificial Intelligence, Quantum Computing.

DOI Open paper PDF

Cross-language evaluation of sorting algorithms on various input patterns: A comparative study

Md Sydul Islam, Pranto Halder, Shah Rafatur Rahman, Shahriar Shakil, Nur Shazwani Kamarudin, Ohahidur Zaman, Md Soyaeb Islam, Md Abdulla Al Mamun

Abstract

Abstract
Sorting algorithms are essential in computing applications, yet their practical performance varies across programming languages and input data structures. Existing studies often focus on single-language implementations or limited datasets, leaving a gap in understanding cross-language behavior under diverse input patterns.
This study presents a comparative evaluation of six sorting algorithms—Bubble, Heap, Insertion, Merge, Quick, and Selection Sort—implemented in C, C++, Go, Java, PHP, and Python. Experiments were conducted using identical arrays of 100,000 elements across six input patterns, including ascending, descending, nearly sorted, random, duplicates, and repeated patterns. Execution times were analyzed using the Coefficient of Variation (CV), Relative Slowdown Factor (RSF), and Pattern Sensitivity Coefficient (PSC).
The results show that programming language environment has a dominant influence on execution efficiency. C and C++ deliver the fastest performance, followed by Go and Java, while PHP and Python exhibit substantially higher runtimes due to interpreter and runtime overhead. Among algorithms, Heap Sort and Merge Sort provide the most consistent performance across input patterns, whereas Quick Sort performs best on random data but shows sensitivity to ordered inputs. Insertion Sort is highly efficient for nearly sorted arrays but exhibits large performance variation, while Bubble and Selection Sort remain inefficient across all environments.
The most effective performance is achieved when robust algorithms are paired with low-overhead languages, particularly Heap Sort and Merge Sort in C and C++. The empirical findings are further supported by theoretical bounds derived from decision-tree complexity and factorial approximations, providing mathematical grounding for observed performance trends.
This study provides a structured cross-language benchmarking framework and practical guidance for selecting efficient sorting strategies in real-world computing environments.
Keywords
Sorting algorithms
; Cross-language performance
; Execution time analysis
; Programming language benchmarking
; Coefficient of variation (CV)
; Relative slowdown factor (RSF)
; Pattern sensitivity coefficient (PSC)

DOI

Resource-Efficient Conjunctivitis Screening Using MOGONET and VGG-Based Transfer Learning

Pranto Halder, Anurag Sinha, Md. Khalid Hasan, Amit Kumar, Jibran Gulzar, Anmol Chauhan, Divyank Sharma, Pratyush Singh, Sagar Singh, Kshitij Tandon, Kshitij Ranjan, Aditya Pandey.

Abstract

Abstract
Background/Aim: Conjunctivitis, commonly known as pink eye, poses a significant diagnostic challenge due to overlapping presentations across viral, bacterial, and allergic subtypes. This study presents MOGONET, a multi-objective generalized normal distribution optimization framework built upon VGGNet, specifically designed to address conjunctivitis identification through resource-aware deep learning optimization.
Methods: The proposed pipeline integrates contrast-limited adaptive histogram equalization-based image enhancement, data augmentation, and multi-level Otsu thresholding segmentation with transfer learning using a VGG16 backbone. A MOGNDO-based outer-loop search simultaneously maximizes classification accuracy, minimizes binary cross-entropy loss, and minimizes computational cost (FLOPs) using Pareto-front-based non-dominated sorting. Experiments were conducted on 1,230 conjunctival eye images sourced from Kaggle and Shutterstock, with a stratified 72%/18%/10% train/validation/test split.
Results: MOGONET achieves 98.32% test accuracy, 97.54% precision, 96.92% recall, and 96.22% F1-score. Compared with the unoptimized Visual Geometry Group-based baseline, MOGONET reduces FLOPs by 79.18% (from 238.06M to 49.56M), and model parameters by 77.78% (from approximately 9M to 2M).
Conclusion: MOGONET demonstrates that multi-objective optimization can substantially reduce computational complexity in conjunctivitis classification without sacrificing diagnostic performance, offering a promising framework for deployment on resource-constrained clinical screening platforms.
Key words: Conjunctivitis, Generalized Normal Distribution Optimization, multi-thresholding, deep learning

DOI Open paper PDF

Unifying Algorithmic Performance: Insertion Sort Across Diverse Programming Paradigms

Islam, M. S. & Halder, P. (2026)

Abstract

Abstract
Abstract— Aim/Background
This study aims to systematically evaluate the practical runtime performance of the Insertion Sort algorithm across six widely used programming languages: C, C++, Go, Java, PHP, and Python. While Insertion Sort’s theoretical time complexity is well established, real-world execution performance may differ significantly due to variations in language paradigms, compilation strategies, and runtime environments. The research focuses on identifying how language-specific characteristics influence actual execution time.
Methods
The Insertion Sort algorithm was implemented uniformly in all six languages using a shared pseudocode to ensure consistency. Experiments were conducted under controlled conditions using input sizes of 1K, 10K, and 100K elements. Each dataset was tested with four different data arrangements: ascending, descending, nearly sorted, and random. Runtime measurements were collected and compared. Additionally, a Relative Execution Time Ratio (RETR) metric was introduced to provide a normalized and quantitative comparison of performance across languages.
Results
The experimental findings show clear performance differences among the programming languages. Go demonstrated the fastest runtimes across most scenarios, benefiting from its efficient compiled execution model. Java performed second-best, leveraging just-in-time (JIT) compilation and runtime optimizations. C and C++ showed moderate performance, influenced by compiler optimization levels and memory management mechanisms. PHP and Python exhibited significantly slower execution times, particularly for large and poorly ordered datasets, mainly due to interpreter overhead and runtime inefficiencies. The RETR metric effectively highlighted these relative performance gaps.
Conclusion
The study confirms that programming language choice significantly impacts real-world algorithm performance, even for well-understood algorithms like Insertion Sort. Compiled languages with efficient runtime optimizations generally outperform interpreted languages, especially for larger datasets. These results emphasize the importance of selecting appropriate programming languages for performance-critical applications and aligning algorithm implementation with both data characteristics and execution environments.
Key words: Insertion Sort, Comparative Performance Analysis, Programming Languages, Algorithm Benchmarking, Runtime Comparison, Sorting Algorithms, Compiled vs Interpreted Languages, Language Efficiency, Experimental Evaluation, Relative Execution Time Ratio (RETR).

DOI Open paper PDF

Language Aware Sorting: Empirical and ML Driven Performance Evaluation of Insertion Sort

Islam, M. S., Halder, P. & Shakil, S.

Abstract

Abstract
Aim/Background: Insertion Sort is widely taught in computer science, yet its practical runtime can vary substantially depending on the programming language and the structure of the input data. This study aims to provide a language-aware empirical comparison of Insertion Sort and to complement benchmark results with machine-learning–based runtime classification.
Methods: Insertion Sort was implemented and evaluated in six languages (C, C++, Go, Java, PHP, and Python). Execution time was measured under a controlled hardware/software environment using identical datasets of three sizes (1,000; 10,000; and 100,000 elements) and four input patterns (ascending, descending, nearly sorted, and random). To interpret performance trends, supervised machine learning models were trained to classify runtime behavior into Fast, Moderate, and Slow categories, and a Relative Execution Time Ratio (RETR) metric was introduced to normalize runtimes against the fastest observed case.
Results: Go achieved the best overall performance, followed by Java, while C and C++ produced mid-range results. Python and PHP were consistently the slowest, with the largest slowdowns observed for large and reverse-ordered inputs. For runtime classification, Random Forest outperformed XGBoost, achieving 93% accuracy compared with 87% for XGBoost. RETR enabled a clearer cross-language comparison by standardizing runtimes relative to the fastest baseline.
Conclusion: Insertion Sort performance is strongly influenced by both programming language and input characteristics. Combining benchmarking with ML-based classification and normalized metrics (RETR) offers a more interpretable view of runtime behavior and supports better alignment of algorithm choices with language and data context in real-world applications.
Key words: Keywords— Insertion Sort, Programming Languages, Runtime Performance, Data Patterns, Machine Learning, Random Forest, XGBoost, Relative Execution Time Ratio (RETR)

DOI Open paper PDF

Challenges and Opportunities of 5G Integration in Bangladesh: Impacts on Education, Research, Business, and Healthcare

Halder, P. & Islam, M. S

Abstract

Abstract
Background: 5G has been officially launched in Bangladesh but remains limited to pilot zones (mainly Dhaka), while nationwide mobile broadband demand continues to grow. This study examines the prospects and constraints of 5G adoption in Bangladesh and maps likely sectoral implications for education, research, business, and healthcare.

Methods: An exploratory research design was employed using secondary datasets from the Bangladesh Telecommunication Regulatory Commission (BTRC), the World Bank, and a University of Minnesota (Twin Cities) measurement dataset. IBM SPSS Statistics 25 and Python 3.6.9 were used for correlation analysis and feature-based visualizations to interpret relationships between key network features and performance indicators.

Results: Throughput shows a positive association with 5G radio type (Pearson 0.669, Kendall 0.492) and with web protocol (Pearson 0.758, Kendall 0.653). Throughput decreases with distance (Pearson −0.85), orientation (Pearson −0.782), NLOS effects (with multipath Pearson −0.533, without multipath Pearson −0.678), and server-type differences (Pearson −0.787), highlighting mmWave sensitivity to propagation and deployment conditions.

Conclusion: While 5G can enable high data rate and low-latency services relevant to key sectors, successful nationwide impact in Bangladesh depends on mitigating mmWave limitations, strengthening infrastructure readiness, and addressing cost, security, and policy coordination across stakeholders.

Key words: Keywords: 5G technology, Internet of Things, Mobile Communications, Cellular Network, TCP protocol

DOI Open paper PDF

Conference

An Edge-Based IoT Video Streaming System with Bounded Buffering for Multi-Client Monitoring

Noman Zuhaer Antor; Pranto Halder; Masud Ibn Afjal; Abdulla Al Mamun

Abstract

Edge video monitoring in barns often needs realtime viewing by multiple users over weak Wi-Fi and limited uplink. This paper presents an edge-based IoT video streaming system where an ESP32-CAM publishes JPEG frames to an edge gateway, and multiple mobile clients subscribe via WebSocket. To avoid memory growth under slow receivers, the gateway uses bounded buffering with controlled frame dropping, enabling fair and stable multi-client delivery. We also analyze protocol and encoding overheads to guide practical deployment for continuous monitoring.

DOI

SecureD-GIoT: Collaborative DDoS Mitigation for Greenfield IoT Using SDN and a Permissioned Ledger

Pranto Halder, Masud Ibn Afjal, Md. Abdulla Al Mamun

Abstract

(Accepted; Paper ID:64 - the 5th IEEE International Conference on Signal Processing, Information, Communication and Systems 2026)

The Business Landscape of 5G in Bangladesh: Opportunities and Challenges

Pranto Halder, GM Arafat Prodhan, Md. Serazul Islam Saykat

Abstract

Accepted; Paper Id: 25; International Conference on Business Innovation and Sustainable Development (ICBISD 2026).

Open paper PDF

Book Chapter

IoT-Enabled Low-Cost Smart Home Security Framework with Multi-Hazard Sensing: Latency and Detection Accuracy Analysis

Riyad Al Araf, Shabbir Hossain Shaptak, Sydul Islam, and Pranto Halder

Open paper PDF

Symposiums

No publications listed in this category yet.

Help · CSE Assistant

Ask about CSE faculty, research, notices, or publications.