Artificial intelligence · Machine learning · Explainability Turin, Italy

ArmanHomafar

Master’s student in Computer Engineering, specializing in AI at Politecnico di Torino.

Building intelligent systems
that can be understood.

I research explainable AI, applied machine learning, and computer vision for engineering and health.

01 / Research

From prediction
to understanding.

My research interests bring together artificial intelligence, machine learning, engineering applications, and model explainability.

01

Explainable AI

Interpreting predictive models with SHAP, examining feature relationships, and studying explanation faithfulness.

02

Applied machine learning

Predictive modeling for engineering systems and health, from industrial processes to voice-based classification.

03

Computer vision

Exploring vision foundation models for dense semantic correspondence and cross-dataset generalization.

02 / Publications

Published research.

Download bibliography

5 journal articles

2024

Engineering

CatBoost-SHAP for modeling industrial operational flotation variables – A “conscious lab” approach

S. Chehreh Chelgani, A. Homafar, H. Nasiri, M. Rezaei laksar

Minerals Engineering, 213, 108754.

Explainable machine learning for modeling industrial flotation variables and their interactions.

Read paper
BibTeX
@article{chelgani2024catboost,
  author = {Chehreh Chelgani, Saeed and Homafar, Arman and Nasiri, Hamid and Rezaei laksar, Mojtaba},
  title = {{CatBoost-SHAP for modeling industrial operational flotation variables – A “conscious lab” approach}},
  journal = {Minerals Engineering},
  year = {2024},
  volume = {213},
  pages = {108754},
  doi = {10.1016/j.mineng.2024.108754},
  url = {https://doi.org/10.1016/j.mineng.2024.108754}
}
2024

Health & medicine

Diagnosis of Parkinson’s disease based on voice signals using SHAP and hard voting ensemble method

P. Ghaheri, H. Nasiri, A. Shateri, A. Homafar

Computer Methods in Biomechanics and Biomedical Engineering, 27(13), 1858–1874.

First published online: 28 September 2023.

Voice-based Parkinson’s disease classification using a hard-voting ensemble and SHAP explanations.

Read paper
BibTeX
@article{ghaheri2024parkinsons,
  author = {Ghaheri, Paria and Nasiri, Hamid and Shateri, Ahmadreza and Homafar, Arman},
  title = {{Diagnosis of Parkinson’s disease based on voice signals using SHAP and hard voting ensemble method}},
  journal = {Computer Methods in Biomechanics and Biomedical Engineering},
  year = {2024},
  volume = {27},
  number = {13},
  pages = {1858--1874},
  doi = {10.1080/10255842.2023.2263125},
  url = {https://doi.org/10.1080/10255842.2023.2263125},
  note = {First published online: 28 September 2023.}
}
2023

Engineering

Modeling operational cement rotary kiln variables with explainable artificial intelligence methods – a “conscious lab” development

R. Fatahi, H. Nasiri, A. Homafar, R. Khosravi, H. Siavoshi, S. Chehreh Chelgani

Particulate Science and Technology, 41(5), 715–724.

First published online: 20 October 2022.

SHAP-XGBoost modeling of industrial rotary kiln variables, including feed rate and induced draft fan current.

Read paper
BibTeX
@article{fatahi2023kiln,
  author = {Fatahi, Rasoul and Nasiri, Hamid and Homafar, Arman and Khosravi, Rasoul and Siavoshi, Hossein and Chehreh Chelgani, Saeed},
  title = {{Modeling operational cement rotary kiln variables with explainable artificial intelligence methods – a “conscious lab” development}},
  journal = {Particulate Science and Technology},
  year = {2023},
  volume = {41},
  number = {5},
  pages = {715--724},
  doi = {10.1080/02726351.2022.2135470},
  url = {https://doi.org/10.1080/02726351.2022.2135470},
  note = {First published online: 20 October 2022.}
}
2022

Engineering

Modeling coking coal indexes by SHAP-XGBoost: Explainable artificial intelligence method

A. Homafar, H. Nasiri, S. Chehreh Chelgani

Fuel Communications, 13, 100078.

SHAP explanations of an XGBoost model relating coal properties to coking quality indexes.

Read paper
BibTeX
@article{homafar2022coking,
  author = {Homafar, A. and Nasiri, H. and Chehreh Chelgani, S.},
  title = {{Modeling coking coal indexes by SHAP-XGBoost: Explainable artificial intelligence method}},
  journal = {Fuel Communications},
  year = {2022},
  volume = {13},
  pages = {100078},
  doi = {10.1016/j.jfueco.2022.100078},
  url = {https://doi.org/10.1016/j.jfueco.2022.100078}
}
2021

Engineering

Prediction of uniaxial compressive strength and modulus of elasticity for Travertine samples using an explainable artificial intelligence

H. Nasiri, A. Homafar, S. Chehreh Chelgani

Results in Geophysical Sciences, 8, 100034.

An explainable SHAP-XGBoost approach to predicting mechanical properties of Travertine samples.

Read paper
BibTeX
@article{nasiri2021travertine,
  author = {Nasiri, H. and Homafar, A. and Chehreh Chelgani, S.},
  title = {{Prediction of uniaxial compressive strength and modulus of elasticity for Travertine samples using an explainable artificial intelligence}},
  journal = {Results in Geophysical Sciences},
  year = {2021},
  volume = {8},
  pages = {100034},
  doi = {10.1016/j.ringps.2021.100034},
  url = {https://doi.org/10.1016/j.ringps.2021.100034}
}

Recent projects

2026
Machine learning · Health

Explainable Parkinson’s Disease Classification from Voice Signals

Revised and extended our published voice-classification study with subject-wise evaluation, a validation-tuned soft-voting ensemble, and enhanced explainability through SHAP faithfulness and concept-level intervention analyses.

Subject-wise evaluationSHAPSoft voting
Related published study
Computer vision · Representation learning

Semantic Correspondence with Vision Foundation Models

Developed a dense correspondence pipeline comparing DINOv2, DINOv3, and SAM, with transformer fine-tuning and window soft-argmax decoding on SPair-71k. Evaluated cross-dataset generalization on PF-Pascal and PF-Willow.

DINOv2 / DINOv3SAMSPair-71k
Research project · 2026

04 / Background

Education & experience.

View full CV

Education

Fall 2024 — Present

Master of Computer Engineering

Politecnico di Torino Turin, Italy

Current cumulative GPA: 23.87 / 30

Selected coursework
  • Data Science and Database Technology29 / 30
  • Artificial Intelligence in Medicine26 / 30
  • Computer Architecture27 / 30

Fall 2019 — Summer 2023

Bachelor of Computer Engineering

Semnan University Semnan, Iran

GPA: 3.52 / 4
Last two years: 3.68 / 4

Awards & honours

  • Among the top 3% of Computer Engineering students at Semnan University.
  • Among the top 2% of university entrance exam participants in Iran.
  • Tuition waiver for the bachelor’s degree at Semnan University, 2019.

Teaching

Teaching Assistant

Artificial Intelligence
Advanced Programming
Logic Circuit
Programming in C++

Industry experience

Tapesh Company Internship

Implemented C++ and Python software for practical engineering tasks and organized database content for efficient access and processing.

Certificates

Machine Learning
Stanford University

Python Programming · Web Design · Image Processing · Adobe Illustrator
Tehran Institute of Technology

Technical toolkit

Programming

Python, C++, C, Java, MATLAB

Methods

Machine learning, deep learning, reinforcement learning, computer vision, image processing, explainable AI

Frameworks & libraries

PyTorch, TensorFlow, scikit-learn, SHAP, LIME, Pandas, NumPy

05 / Contact

Let’s discuss
research.

I am interested in PhD opportunities and research collaborations in explainable AI, machine learning, and their applications.

arman.homafar@ieee.org