LINGUISTIC DATABASE FOR AUTOMATIC GENERATION SYSTEM OF ENGLISH ADVERTISING TEXTS

Abstract

The article deals with the linguistic database for the system of automatic generation of English advertising texts on cosmetics and perfumery. The database for such a system includes two main blocks: automatic dictionary (that contains semantic and morphological information for each word), and semantic-syntactical formulas of the texts in a special formal language SEMSINT. The database is built on the result of the analysis of 30 English advertising texts on cosmetics and perfumery. First, each word was given a unique code. For example, N stands for nouns, A – for adjectives, V – for verbs, etc. Then all the lexicon of the analyzed texts was distributed into different semantic categories. According to this semantic classification each word was given a special semantic code. For example, the record N01 that is attributed to the word «lip» in the dictionary means that this word refers to nouns of the semantic category «part of a human’s body».The second block of the database includes the semantic-syntactical formulas of the analyzed advertising texts written in a special formal language SEMSINT. The author gives a brief description of this language, presenting its essence and structure. Also, an example of one formalized advertising text in SEMSINT is provided.

Keywords

• automatic generation of english advertising texts
• linguistic database
• automatic dictionary
• semanticsyntactical formal language