Mehdi Mirzapour


pdf bib
Introducing RezoJDM16k: a French KnowledgeGraph DataSet for Link Prediction
Mehdi Mirzapour | Waleed Ragheb | Mohammad Javad Saeedizade | Kevin Cousot | Helene Jacquenet | Lawrence Carbon | Mathieu Lafourcade
Proceedings of the Thirteenth Language Resources and Evaluation Conference

Knowledge graphs applications, in industry and academia, motivate substantial research directions towards large-scale information extraction from various types of resources. Nowadays, most of the available knowledge graphs are either in English or multilingual. In this paper, we introduce RezoJDM16k, a French knowledge graph dataset based on RezoJDM. With 16k nodes, 832k triplets, and 53 relation types, RezoJDM16k can be employed in many NLP downstream tasks for the French language such as machine translation, question-answering, and recommendation systems. Moreover, we provide strong knowledge graph embedding baselines that are used in link prediction tasks for future benchmarking. Compared to the state-of-the-art English knowledge graph datasets used in link prediction, RezoJDM16k shows a similar promising predictive behavior.


pdf bib
Quantifier Scoping and Semantic Preferences
Davide Catta | Mehdi Mirzapour
Proceedings of the Computing Natural Language Inference Workshop

pdf bib
Finding Missing Categories in Incomplete Utterances
Mehdi Mirzapour
Actes des 24ème Conférence sur le Traitement Automatique des Langues Naturelles. 19es REncontres jeunes Chercheurs en Informatique pour le TAL (RECITAL 2017)

Finding Missing Categories in Incomplete Utterances This paper introduces an efficient algorithm (O(n4 )) for finding a missing category in an incomplete utterance by using unification technique as when learning categorial grammars, and dynamic programming as in Cocke–Younger–Kasami algorithm. Using syntax/semantic interface of categorial grammar, this work can be used for deriving possible semantic readings of an incomplete utterance. The paper illustrates the problem with running examples.