Corina Dima

University of Tübingen, Alexandru Ioan Cuza University

Papers

Total Citations

H-Index

About

Corina Dima is a computational linguist whose research bridges deep learning and cognitive modeling to unravel the complexities of language. Her most influential work, "Automatic Noun Compound Interpretation using Deep Neural Networks and Word Embeddings" (2015, 33 citations), pioneered the use of neural network classifiers and word embeddings to automatically identify semantic relations in English noun compounds—a foundational task for natural language understanding. This contribution demonstrated how distributed representations can capture nuanced meaning in compound structures, advancing both computational semantics and practical NLP applications. Dima also explored the evolutionary emergence of linguistic structures in "Why Would a Robot Make Use of Pronouns?" (2009), investigating how pronominal anaphora might arise in artificial agents through cognitive simulation. Her work sits at the intersection of machine learning, cognitive science, and linguistics, offering insights into how machines can interpret and generate human-like language. By combining empirical rigor with theoretical curiosity, Dima has helped shape modern approaches to semantic interpretation, making her research essential reading for students and researchers working on compositionality, representation learning, and the computational modeling of language.