Mehrdad Rafiepour's Image
Mehrdad Rafiepour

 mehrdad.rafie.p (at) gmail

Socials:     

HomeSkillsArticlesCVWorkTimeline
About Me

My research focus is on natural language processing, with a particular emphasis on dialogue systems, large language models and NLP for social good. During my master's, I explored several NLP subfields including dialogue systems and text to query.  My thesis focused on natural language understanding through joint intent-detection and slot-filling, resulting in "CTRAN", the current SOTA result in the field for slot-filling.

I also have five years of experience as a self-employed programmer, developing mobile apps with nationwide impact. My passion for accessibility and social impact drives my work, exemplified by projects like Hamyar, a Farsi-language smartphone interface for the visually impaired. These experiences inspire my ongoing focus on natural language processing to empower the less fortunate.

I am currently seeking PhD positions starting in 2025! If you know of any faculty specializing in my research interests below who are looking for students, please reach out via email at mehrdad.rafie.p (at) gmail. I am also open to productive collaborations. If you share my research interests, please get in touch.

Research Interests

🔹 Dialogue systems
🔹 Large Language Models
🔹 Natural Language Processing for Social Good
🔹 Alterations of Transformer Architecture in Textual Data
🔹 Natural Language Understanding
🔹 Text to SQL

Education

  M.Sc. in Computer Engineering, University of Kashan

Thesis:

Proposing a Model for Natural Language Understanding Using Deep Neural Networks

 Sep 2019 - Feb 2023

  B.Sc. in Computer Engineering, University of Qom

Thesis:

Designing and Implementing the Assistant Program for the Visually Impaired in Android

 Sep 2013 - Feb 2018
News
I successfully defended my M.Sc. Thesis at University of Kashan! Feb 2023
CTRAN's article is accepted by prestigious Engineering Applications of Artificial Intelligence journal. Sep 2022
I gave an oral presentation at the 2nd national conference of informatics entitled "Distinguishing dense networks from pseudo-tree networks for link prediction based on homogeneity and heterogeneity criteria" Dec 2021
My article on an appropriate criteria for detecting the proper link-prediction method has been accepted by the 2nd national conference of informatics Oct 2021
Mehrdad Rafiepour's Image

Mehrdad Rafiepour

HomeSkillsPublicationsDownload CVCareer TimelineWork Experience

Socials: 

Email: mehrdad.rafie.p (at) gmail

Designed by M. Rafiepour. Licenced under GPL.