| Resumen |
The development of corpus in general represents an arduous task due to the analysis and labeling processes, as well as the elements to consider in its development. For its part, the study of literal language and non-literal language in its various literary figures represents an important area of study for Natural Language Processing. The study of this phenomenon in Spanish has been limited by the lack of corpora available for its study as well as the lexical and semantic complexity of the language. In accordance with the above, this project proposes a labeling algorithm for non-literal and literal language, likewise reviews the points to consider regarding design and experimentation and presents results from the CESS-ESP corpus. |
