Here's the version of the abstract which is in English:
Associated with preconceived images and full of prejudices, romance is a consumption product which embedies a social and cultural mirroring. In France for several decades, romance has gained an heterogeneous but loyal readership/audience. What are the motives of those novels which are said to be all similar? What do they represent for the readership? How are they perceived by those who read them? How are they read, apprehended and interpreted? These are some of the questions we try to give an answer to in this multidisciplinary work. The aims of this thesis are several : - On a 61-novel corpus: studying diachronically and morphologically the covers, the characters, the narrative patterns; analysing the lexical evolution on a digitized corpus of two million words thanks to different Natural Language Processing softwares ; - Studying reception, first by interviewing the readers, then by exploring a corpus constituted by messages posted on a discussion forum devoted to romance, applying Natural Language Processing methodologies. This research enabled to unveil the mecanisms involved in the evolution of romance since World War II, and to observe the different modes of reading and reception of these novels
Here's the version of the abstract which is in English: