Our project

The lexeme we chose to work on was "cochlear implant".
Our main goal was to obtain a linguistic and statistic view of a variety of opinions about the cochlear implant in three of the most spoken languages in the world: English, Spanish and French.
In order to do this, we wrote a bash script allowing us to extract the relevant information for the analysis, to convert them so we could manipulate and classify it as an ordered table.
We also used the necessary IT tools to do a linguistic and statistical analysis, present the results and interpret them.
As part of our Master NLP course, we also had to record our progress and difficulties chronologically in a blog, like a logbook. This blog is available online by clicking the corresponding section in the top menu.

The debate that inspired all

A cochlear implant is an electronic implant that is intended to provide a sense of hearing to people with certain severe deafness.
This raised and still raises a fundamental debate for the deaf and hard-of-hearing community. While supporters of the cochlear implant see it as a way to integrate into the community with hearing, opponents of this medical procedure see it as an expensive surgery, dangerous for the patient, without guarantee of optimal functioning and ostracizing in both communities (with hearing and deaf). For more details on this we recommend " Les sourds, c'est comme ça" from the anthropologist Yves Delaporte and " Psychologie de la surdité" from Benoit Virole.

This also leads to a difficulty. While this may seem trivial for people unaccustomed to text extraction, having not one but two words in our lexeme poses a series of additional challenges that we gladly accepted. Among these difficulties we found: the use of the plural (in two or one word of the term), the omission of a word, the acronyms of the term, the order of the words in each language, etcetera.

Our team

Our script

To see our script clic HERE.