Moein
Shahiki Tash
Professor at Universidad Panamericana
AI Researcher · NLP · Social Media Mining · Computational Social Science
Open to Collaboration
I am a Ph.D. in Computer Science (Natural Language Processing) from the Centro de Investigación en Computación (CIC), Instituto Politécnico Nacional (IPN) in Mexico City, where I conducted research at the prestigious Gelbukh NLP & AI Research Group.
My research focused on Social Media Mining, exploring how computational methods can uncover hidden patterns in human online communication. I specialized in a range of NLP tasks including social support detection, hate speech and hope speech detection, sentiment analysis, language identification, and fintech-related NLP.
Passionate about the intersection of technology, language, and data science, I build AI systems that understand and improve online human interactions. I have published in top-tier journals and conferences, and served as Main Organizer of the Social Support Detection (SSD) Shared Task at IberLEF 2026.
I currently teach graduate-level NLP courses at Panamericana University, CDMX, and bring extensive industry experience from my work as an ML Engineer and Python Developer.
Automatically identifying emotional, informational, and tangible support in online communities using multilingual NLP models and social media datasets.
Detecting harmful content and counter-narratives on social platforms using deep learning, transformer models, and psycholinguistic features.
Analyzing cryptocurrency discourse on social media platforms using SenticNet, sentiment analysis, and predictive market emotion models.
Word-level language identification in code-mixed texts (Kannada-English, Dravidian languages) using traditional ML and transformer architectures.
Developing robust sentiment classifiers for multilingual and code-mixed social media content across diverse linguistic and cultural contexts.
Extracting behavioral insights and communication patterns from large-scale social media corpora using computational methods and NLP.
Built a conversational AI chatbot that ingests content from URLs and answers user queries using retrieval-augmented generation. Deployed via Streamlit for interactive use.
Created and curated a bilingual social support detection dataset in English and Spanish sourced from social media, used for training and benchmarking NLP models.
Analyzed sentiment and psycholinguistic features of cryptocurrency discourse on X (Twitter), correlating emotional trends with market price movements using SenticNet.
Organizing the Social Support Detection Shared Task at IberLEF 2026 — including task design, dataset curation, evaluation framework, and coordinating international participants.
Deep learning pipelines for binary and multiclass classification of hate and hope speech in social media text, covering Dravidian, Spanish, and English languages.
Designed and deployed production NLP models for various text classification and sequence labeling tasks using TensorFlow and Scikit-learn during ML Engineering tenure.
I'm always open to academic collaborations, research discussions, and opportunities. Feel free to reach out through any of the channels below.