This paper provides readers with a three-page summary of Reyes' Ph.D. thesis in Computer Science, whose main objective is to demonstrate how two figurative devices, i.e. humor and irony could be automatically detected on the basis of linguistic-based patterns. According to the author, the automatic recognition of figurative language is problematic in as far as it involves aspects that range from pronunciation, going through the lexical and syntactic realms, to conceptualization. One single technique or algorithm will thus fail to provide a computational account of figurative language use. Section 3 offers a succinct description of the structure and contents of each thesis chapter, to then present the two models of figurative language put forward, i.e. HRM (Humor Recognition Model) and IDM (Irony Detection Model). Both models are argued to go beyond surface elements to extract different kinds of patterns from a text: from lexicon to pragmatics. At the end of the paper, the major findings are listed, among which we may highlight the following: (i) through the representation of humor and irony in terms of their conceptual use, the model can efficiently capture the most salient feature of each figurative device; (ii) figurative language representation is achieved by analyzing the linguistic system as an integral structure which depends on grammatical rules as well as on cognitive, experiential, and social contexts; (iii) the study of non-prototypical examples allowed the author to establish the aforementioned models which were supported by general patterns employed by speakers to effectively communicate figurative intents.