Research Overview
My research bridges two worlds: the deep, formal study of language structure and the applied challenges of multilingual technology. On one side, I explore the architecture of grammar — the primitive relations that underlie complex predicates, morphology, and meaning. On the other, I build systems for social media analysis, narrative and event extraction, scientometrics, and language learning, especially for Persian and other low-resource languages. Across these efforts runs a single goal: to understand how language works at its core and to develop AI systems that respect its complexity, cultural richness, and human purpose.
​
This broader research direction is deeply rooted in the intellectual influence of Jean-Roger Vergnaud. His view of grammar as a system built from abstract primitives, graph-like relations, and event structure resonates strongly with my work on complex predicates, morphology, and meaning. Although his ideas were often ahead of their time, they continue to shape the questions I ask: what is the underlying architecture of language, and how can it illuminate the future of linguistically grounded AI? In many ways, my work humbly extends the theoretical path he opened but never fully had the chance to complete.​
Computational Model of Language
My current research centers on building formal models of grammar that integrate insights from generative syntax, Distributed Morphology, and semantic theory. Drawing on Jean-Roger Vergnaud’s graph-theoretic approach to the primitive elements of meaning, I explore how linguistic systems represent complex predicates, event structure, and morphological processes. This work aims to develop a precise, computationally explicit account of grammar — one that captures the architecture of natural language independently of, but ultimately relevant to, today’s AI systems.
Social Media Analytics
My research in social media analytics explores how language is used across platforms to express emotion, ideology, and social tension. Focusing on both English and Persian content, I’ve developed NLP tools to detect hate speech, toxicity, sentiment, and emerging topics in blogs and Twitter data. I also analyze how information spreads across social networks, with applications in countering online radicalization and supporting public well-being.
Narrative Analytics
I develop systems that can automatically extract and interpret the structure of narratives—who did what, when, where, and why. This research supports timeline generation, event detection, and narrative understanding across domains such as media, healthcare, and criminal justice. From tracking life events in documents to identifying health risks or analyzing media bias, narrative analytics helps uncover deeper patterns in unstructured data.
Scientometrics & Research Analytics
In the field of scientometrics, I apply computational and network analysis techniques to study scientific research itself. My work examines how ideas evolve, how researchers connect, and how influence flows through academic networks. Using bibliometrics, knowledge graphs, and NLP, I uncover trends in fields like linguistics and synthetic biology to support science policy, funding decisions, and innovation mapping.
Persian NLP
With decades of experience in Persian computational linguistics, I’ve contributed foundational resources and tools for processing Persian text, including segmentation, morphology, parsing, and sentiment analysis. My work also bridges NLP with social media, ideology detection, and language teaching. Current projects include building Persian narrative analysis systems and AI-powered tools to support learners and teachers in the classroom.
Complex Predicates & Syntactic Theory
I investigate how verbs and verb-like constructions convey complex meanings in Persian and Armenian. My research focuses on light verbs, morphologically derived predicates, and causative constructions—analyzing how syntax and semantics interact in these systems. This work contributes to linguistic theory and supports better language processing tools for the analysis of Complex Predicates.
Pedagogical Linguistics & AI for Language Learning
Blending formal linguistics, AI, and language pedagogy, I design data-driven tools and curricula for teaching Persian. My approach centers on integrating real-world texts, NLP-based feedback, and linguistic theory into language instruction—especially for heritage speakers. I aim to build intelligent educational systems that help learners grasp deeper structural and cultural patterns of the language.