Research Interests
Explainable AI (XAI) attributions, Trustworthy ML, baseline Time-Series Forecasting, Agriculture AI diagnostics, and Educational Data Mining.
Prospective Fall 2026 Thesis MSc Applicant
Md. Salauddin
Sarker
Seeking a funded thesis-track MSc position abroad to research how Explainable AI (XAI) and trustworthy machine learning can safeguard agricultural, time-series, and industrial decision systems.
3.84
B.Sc. CGPA (Dean's x2)
Q1
Featured Springer Pub
6.5
IELTS English Band
5
Papers & Manuscripts
For Supervisors
MSc Research Fit.
I am targeting a funded, thesis-based MSc abroad where I can contribute interpretable, reproducible, and deployable machine-learning pipelines from day one.
- IntakeFall 2026 · thesis track
- FundingSeeking fully funded
- EnglishIELTS 6.5
- B.Sc. CGPA3.84 / 4.00
- Output3 published · 1 Q1
Qualifications
Featured Q1 review (Springer Nature), IEEE conference papers, CGPA 3.84/4.00, two-time Dean's List, IELTS 6.5.
Research Practice
Reproducible experiments, systematic baselines, clean documented notebooks, PostgreSQL pipelines, and rigorous validation.
Availability
Open to funded MSc/PhD thesis supervision, Research Assistant (RA) tracks, and graduate research for Fall 2026.
Proposed MSc Research Agenda
Q1
Are post-hoc explanations actually faithful?
Measuring whether SHAP, LIME, and Grad-CAM attributions align with verified domain evidence in high-stakes vision (crop pathology, medical screening) — not just plausible-looking heatmaps.
Q2
Trustworthy ML on small, imbalanced clinical data.
Combining feature selection (RF/RFE/ANOVA), resampling (SMOTE/IHT), and interpretable ensembles so medical models stay both accurate and explainable on limited patient datasets.
Q3
When do simple baselines beat complex forecasters?
Rigorous evaluation of non-stationary time-series where honest OLS/ARIMA baselines expose whether deep models add real predictive value or just complexity.
Directions drawn from my published work — open to reshaping around a supervisor's active projects.
Scholarship Aspiration
Trustworthy
Data Science.
Architecting machine learning systems that maintain interpretability and robustness under high-stakes, real-world constraints.
My academic path started within Educational Technology & Engineering (EdTE) at the University of Frontier Technology, Bangladesh (UFTB). I quickly became interested in a core vulnerability: complex deep neural classifiers excel at mapping parameters, but their decision boundaries are opaque.
This inspired my research direction. By using local post-hoc explanations like SHAP, LIME, and Grad-CAM, I work to establish transparent, faithful attributions. My goal is to ensure model attributions align with verified physical domain evidence before predictions are integrated into agricultural or time-series systems.
Explainable AI (XAI)
Developing faithful, post-hoc local attributions for deep agricultural leaf-pathogen computer vision models and educational student-risk classifiers.
Dynamic Forecasting
Rigorously evaluating model stability and error margins across non-stationary time-series datasets against robust ARIMA/OLS baselines.
"The complexity of industrial data requires the precision of neural interpretation."
Scientific Philosophy
Technical Toolkit
Build
Python
SQL
Java
PHP
Dart
Linux / Shell
Model
PyTorch
TensorFlow
scikit-learn
XGBoost
ARIMA / OLS
Computer Vision
Explain
SHAP
LIME
Grad-CAM
Spatial Attention
RF / RFE / ANOVA
Data & ERP
PostgreSQL
Odoo 17
ETL Pipelines
Tableau
Metabase
Languages
Bengali — Native
English — Professional (IELTS 6.5)
Japanese — Basic (JLPT N5 track)
Publication Footprint
Research Evidence.
Verified scientific output in Q1 journal, conference, and manuscript-stage tracks.
3Peer-reviewed published
2Under review / in prep
Q1Springer · JIF 13.9
98.29%Best model accuracy
SCIE · Scopus · IEEEIndexing reach
Peer-Reviewed & Published
Springer Nature Q1 Journal
Author 4 of 8
SCIE + Scopus Indexed
Leveraging explainable AI for sustainable agriculture: a comprehensive review of recent advances
Aditya Rajbongshi, Fatema Tuz Johora, Arafat Hossain, Md. Salauddin Sarker, Md Habibur Rahman, Md Wahidur Rahman, Fahad T. Alotaibi, Mohammad Ali Moni.
Artificial Intelligence Review (2026) Vol 59, 105 · JIF: 13.9 | CiteScore: 26.3
My contribution: deep learning, machine learning, and XAI analysis synthesised into the review.
Abstract: This systematic review details current advances in Explainable AI (XAI) within precision agriculture. It synthesizes over 150 papers to map out local post-hoc attribution methods (LIME, SHAP, and Grad-CAM) across leaf pathogen detection, soil nutrient forecasting, and yield predictions. We propose a structured 4-tier taxonomy that evaluates model transparency, and discuss key challenges concerning explainability faithful bounds.
@article{rajbongshi2026leveraging,
title = {Leveraging explainable AI for sustainable agriculture: a comprehensive review of recent advances},
author = {Rajbongshi, Aditya and Johora, Fatema Tuz and Hossain, Arafat and Sarker, Md. Salauddin and Rahman, Md Habibur and Rahman, Md Wahidur and Alotaibi, Fahad T. and Moni, Mohammad Ali},
journal = {Artificial Intelligence Review},
volume = {59},
pages = {105},
year = {2026},
doi = {10.1007/s10462-025-11459-5}
}
IEEE Conference Paper
Author 2 of 6
IEEE Xplore Indexed
Bridging Climates: Weather Forecasting Using OLS and ARIMA Models with a Multi-Country Dataset
Arafat Hossain, Md. Salauddin Sarker, Aditya Rajbongshi, Most. Rakiba Khanom Jisa, Maria Afrin Bindu, Kayes Mohammad Abdullah.
IEEE 2nd International Conference on Computing, Applications and Systems (COMPAS 2025)
My contribution: methodology, code implementation, and the results section.
Abstract: This research presents a comparative diagnostic study of standard statistical models—Ordinary Least Squares (OLS) regression and Autoregressive Integrated Moving Average (ARIMA)—for predicting historical climate indicators. We utilize 61 years of longitudinal meteorological records across Bangladesh, Saudi Arabia, Japan, and Russia. Standard residual tests confirm OLS yields lower prediction errors (RMSE 0.30) for baseline variables in agricultural forecasting.
@inproceedings{hossain2025bridgingclimates,
title = {Bridging Climates: Weather Forecasting Using OLS and ARIMA Models with a Multi-Country Dataset},
author = {Hossain, Arafat and Sarker, Md. Salauddin and Rajbongshi, Aditya and Jisa, Most. Rakiba Khanom and Bindu, Maria Afrin and Abdullah, Kayes Mohammad},
booktitle = {2025 IEEE 2nd International Conference on Computing, Applications and Systems (COMPAS)},
year = {2025},
publisher = {IEEE},
doi = {10.1109/COMPAS67506.2025.11381792}
}
IEEE Conference Paper
Author 5 of 6
IEEE Xplore Indexed · 94.74% Accuracy
Bridging Educational Gaps: Predicting Retakes Among Bangladeshi Undergraduates with Machine Learning and Explainable AI
Nusrat Zahan Nila, Md. Ali Mahmud Pritom, Fatema Tuz Johora, Aditya Rajbongshi, Md. Salauddin Sarker, Md. Ashrafuzzaman.
IEEE International Conference on Computer and Information Technology (ICCIT 2025)
My contribution: explainable AI (XAI) analysis and the results section.
Abstract: Course failures and retakes severely interrupt academic timelines in higher education. We present a predictive model using records from 474 Bangladeshi undergraduates, deploying an ensemble soft voting classifier that achieves 94.74% risk-prediction accuracy. We incorporate SHAP and LIME explanatory pipelines to isolate core socio-economic and preparatory vectors responsible for academic risk.
@inproceedings{nila2025bridgingeducational,
title = {Bridging Educational Gaps: Predicting Retakes Among Bangladeshi Undergraduates with Machine Learning and Explainable AI},
author = {Nila, Nusrat Zahan and Pritom, Md. Ali Mahmud and Johora, Fatema Tuz and Rajbongshi, Aditya and Sarker, Md. Salauddin and Ashrafuzzaman, Md.},
booktitle = {2025 International Conference on Computer and Information Technology (ICCIT)},
year = {2025},
publisher = {IEEE},
doi = {10.1109/ICCIT68739.2025.11491491}
}
Under Review & In Preparation
B.Sc. Thesis
First author
Medical AI · Under Review · 98.29% Accuracy
Enhancing Parkinson's Disease Prediction Using Feature Selection Models (RF, RFE, and ANOVA) and Explainable AI: A Comparative Analysis of Machine Learning Performance
Md. Salauddin Sarker, et al.
Undergraduate (B.Sc.) thesis · Explainable Medical AI · Speech-based Parkinson's detection · Manuscript under review
My contribution: dataset collection, methodology design and implementation, and the results section — primarily code implementation.
Abstract: Speech disorder is a significant challenge for Parkinson's patients, impairing effective communication; early detection is essential to minimise movement issues and disability. While prior work detects Parkinson's with machine learning, a gap remains in the explicability of those models. We propose a system that both detects Parkinson's disease and analyses the reasons behind predictions to identify the actual causes, increasing transparency and trustworthiness. Using a medical dataset of 188 patients (mixed text and numerical features), we apply three feature-selection methods after preprocessing, and two balancing algorithms (SMOTE and IHT) to address class imbalance — producing six unique datasets. An ensemble learning approach is compared against seven individual models; the weighted ensemble achieves the highest accuracy at 98.29%. Two explainable AI (XAI) techniques ensure transparency and interpretability of the chosen model. The proposed system is expected to play a vital role in the medical sector.
In Preparation
First author
Edge AI · Industrial inference
EdgeVision: Quantization-Aware Inference for Industrial Edge
Md. Salauddin Sarker, et al.
Positioned for Industrial Edge Inference / IDAA Forum Track
Abstract: This research implements post-training 8-bit integer quantization schemas for industrial visual models. By adapting low-power ARM architecture nodes, the framework compresses standard CNN parameter blocks, successfully achieving sub-10ms edge inference latencies while protecting overall leaf classification fidelity.
Selected Implementations
Research &
Research &
Case Studies.
99.42%
Rice-leaf disease accuracy
Attention CNN + Grad-CAM · 42 FPS on edge
98.29%
Parkinson's detection
Weighted ensemble + XAI · B.Sc. thesis
94.74%
Student-retake prediction
Soft-voting + SHAP/LIME · IEEE ICCIT
RMSE 0.30
Multi-country forecasting
Honest OLS/ARIMA baseline · 61 yrs data
Case Study #01 — Deep Learning + XAI
Rice Leaf Disease Detection
Building an accurate and interpretable classifier for sustainable agri-diagnostics.
The Problem
Need for reliable, transparent identification of leaf pathogens in field conditions.
The Data
Multi-class leaf image dataset with diverse environmental backgrounds.
Approach: Transfer learning (VGG16/ResNet50/EfficientNetB0) + spatial attention mechanisms.
Explainability: Grad-CAM heatmaps for decision transparency and feature verification.
Results: 99.42% overall classification accuracy; sustained 42 FPS on mobile edge platforms.
RESEARCH IMPACT (MSc READY)
Demonstrated that spatial attention maps can pinpoint pathogenic regions in leaf images better than standard CNNs, crucial for trustworthy field diagnostic deployment.
Case Study #02 — Time-Series Forecasting
GitHub Profile
Climatic Baseline Optimization
Evaluating OLS and ARIMA architectures across 61 years of meteorological data.
The Problem
Forecasting temperature and variables accurately across highly diverse meteorological zones.
The Data
61 years of multi-country historical temperature data (Bangladesh, KSA, Japan, Russia).
Approach: Hybrid ARIMA structures + OLS linear trends; tested stationarity and residuals.
Validation: OLS reached an RMSE of 0.30 for temperature forecasting in Bangladesh.
BASELINE DISCIPLINE
Proved that complex forecasting models need careful, standard OLS/ARIMA statistical baselines to evaluate whether true predictive novelty exists.
Case Study #03 — Enterprise Data Engineering
Industrial Churn Prediction
Connecting raw enterprise database transactions with custom predictive ML models.
The Problem
Aggregating scattered commercial flows into clean, structured data records ready for ML.
The Integration
PostgreSQL ETL pipelines combining Odoo Sales, inventory, and ledger databases.
Ecosystem: Integrated Odoo 17 ERP transaction workflows, optimizing relational queries.
Machine Learning: XGBoost and RandomForest classifiers evaluated using local SHAP importance.
Outcome: Reached 85% predictive accuracy on operational client retention flows.
Attribution Strength
SHAP Importance
Contract Tenure
0.31
Customer Support Tickets
0.24
Transaction Frequency
0.18
Avg Order Volume
0.12
Applied Engineering Proof
Professional Experience.
Industry data-systems work that demonstrates the engineering rigor and reproducibility behind my research.
Senior Principal Officer — Software Developer (ERP)
CurrentLeading enterprise ERP and software development: designing Odoo modules, relational data models, and reporting workflows that turn raw operational data into auditable, analysis-ready systems.
Odoo
PostgreSQL
Python
ERP Architecture
Executive — Software Developer (ERP)
Q Cosmetics Ltd · Dhaka, Bangladesh
Built the connective data layer between operations and decision-making: Odoo 17 ERP workflows, PostgreSQL ETL pipelines, and BI dashboards — converting raw transactional data into auditable, analysis-ready evidence.
35%
Faster order-cycle workflow via ERP rollout.
12×
Reporting cut from ~3 hours to ~15 minutes.
6+
Systems deployed across sales, inventory, purchase.
Odoo 17
PostgreSQL
ETL Pipelines
Python
BI Dashboards
Trajectory & Certifications
Verified Credentials.
SSC Science
GPA 5.00 / 5.00
Dhaka Board
HSC Science
GPA 5.00 / 5.00
Dhaka College
B.Sc. in EdTE
CGPA 3.84 / 4.00
UFTB, Bangladesh · 2019-2024
Thesis: Explainable AI for Parkinson's disease prediction (weighted ensemble, 98.29%).
AI
Statistics
Robotics
MSc Track
Fall 2026
Seeking funded thesis supervision
Verified Certifications List
Click each credential to switch the high-resolution evidence scan on the right screen.
Document Evidence
Distinction & Engagement
Leadership & Community.
Academic honours, national innovation recognition, and elected leadership across engineering and language communities.
Dean's Award (2022 & 2024)
University of Frontier Technology (UFTB)
Recognised for top-tier academic performance across two non-consecutive cycles.
National STEAM Olympiad 2023
9th Nationally · Campus Ambassador
Selected campus leader for the national innovation drive; ranked top 10 nationally on technical pitch.
UFTB STEAM Club
Led technical mentorship for 200+ members and organised engineering hackathons and innovation drives.
UFTB Robotics Club
Coordinated large-scale robotics competitions and cross-discipline engineering teams.
UFTB Language Club
Drove multilingual academic exchange and Japanese (N5) learning initiatives for global readiness.
Faculty Endorsements
Academic Referees.
Official recommendations and academic credentials verified by university professors.
Farhana Islam
Assistant Professor & Chairman
Dept. of Educational Technology & Engineering
farhana0001@uftb.ac.bd
University of Frontier Technology, Bangladesh
Aditya Rajbongshi
Assistant Professor
Dept. of Educational Technology & Engineering
aditya0001@uftb.ac.bd
University of Frontier Technology, Bangladesh
For Prospective Supervisors
Considering a Fall 2026 student?
If your group works on Explainable AI, Trustworthy ML, time-series, or applied AI for agriculture, health, or education — I am funded-MSc ready and can contribute from day one.
Run honest baselines
Reproducible experiments with rigorous OLS/ARIMA and ML baselines before claiming novelty.
Build XAI evaluation
SHAP/LIME/Grad-CAM pipelines that test whether explanations are faithful, not just plausible.
Write & document
Publication-ready writing and clean, supervisor-friendly code — proven across 3 IEEE/Springer papers.
Inquiry & Collaboration
Get in Touch.
Open to MSc supervision, research collaboration, and inquiries regarding my scientific publications and deployments. Let's connect.
Phone / WhatsApp
+880 1521-260707
Location
Dhaka, Bangladesh
Message Prepared
Your email client has been opened with the inquiry prefilled. Click send inside your mail app!