Semantic Predicates in the Business Language

• Published in: Formalizing Natural Languages with NooJ and Its Natural Language Processing Applications Vol. 811; pp. 108 - 116
• Main Authors: della Volpe, Maddalena, Elia, Annibale, Esposito, Francesca
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 01.01.2018; Springer International Publishing
• Series: Communications in Computer and Information Science
• Subjects: Business language; Natural language processing; NooJ application; Semantic predicates; Text Mining
• ISBN: 9783319734194; 3319734199
• ISSN: 1865-0929; 1865-0937
• DOI: 10.1007/978-3-319-73420-0_9

In recent years, the interest in the use of language for business has grown. It is recognized that the hidden persuasive linguistic potential improves the company’s positioning in the public consciousness. The language of the business world is multifarious: we try to identify its features and behaviour, considering the evolution that it has faced primarily with the globalization of markets. Business activities are so complex that they require the application of several disciplines at the same time and therefore the use of specific languages and technical terminology. In order to reach an efficient analysis of business language, this study explores the role of semantic predicates constructed from lexical and the syntactic structures in which they are placed within business communication contexts. From the point of view of LG framework, a set of lexical-syntactic structures defines the value of semantic predicates, while the arguments selected by each semantic predicate are given the value of actants, subjects included. The features of each verb are expressed by the application of the rules of co-occurrence and selection restriction, through which verbs select semantically their arguments to construct acceptable simple sentences. In this way, the entries belonging to electronic dictionaries should be classified presuming their similarity and proximity. Even if the list of semantic tags is not simply identifiable, grammars could be built for single sets of semantic predicates. LG descriptions assign correlated predicates and arguments by applying electronic dictionaries of Italian. Using NooJ environment and Italian linguistic resources to automatically processing natural language, we will process a corpus of business documents. We will show and describe the syntactic structures, semantic and syntactic properties of predicates, in order to build formal grammar for business language.