IMAGACT is a cross-linguistic ontology of action, in which action concepts are represented trough Prototypes (3D animations or brief films). The interface IMAGACT4ALL allows mother tongue informants to assign verbs of their language to each prototype and has been used to implement languages belonging to different families. The Ontology specifies the range of different actions which may fall in the extension of each action verb and the set of verbs which can identify each entry, ensuring an adequate translation to action verbs, which show high ambiguity and cross-linguistic semantic variability. A large initiative for the implementation of various Indian languages (Hindi, Urdu, Sanskrit, Bengali, Odia, Assamese, Magahi, Manipuri, Tamil) was undertaken. The paper sketches the status of the work, whose main achievement is the full implementation of Hindi/Urdu and focus on “taking events”, that are very relevant in ordinary communication, but feature strong differences in lexical encoding cross-languages. Hindi requires 7 different verbs to cover the actions extended by the general verb take. The main translator लेना(lenA) is also a general verb, but its application has specific semantic boundaries. The paper specifies how features are induced from prototypes, exploiting IMAGACT for the semantic interpretation of Hindi verbs.
@InProceedings{MONEGLIA18.10, author = {Massimo Moneglia ,Alessandro Panunzi and Lorenzo Gregori}, title = {“Taking Events” in Hindi. A Case Study from the Annotation of Indian Languages in IMAGACT}, booktitle = {Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018)}, year = {2018}, month = {may}, date = {7-12}, location = {Miyazaki, Japan}, editor = {Girish Nath Jha and Kalika Bali and Sobha L and Atul
Kr. Ojha}, publisher = {European Language Resources Association (ELRA)}, address = {Paris, France}, isbn = {979-10-95546-09-2}, language = {english} }
Summary of the paper