Automatic Text Generation (ATG) is a Natural Language Processing (NLP) task that aims at writing acceptable and grammatical written text exploiting machine-representation systems, such as for instance knowledge bases, taxonomies and ontologies. In this sense, it is possible to state that an ATG system works like a translator that converts data into a natural-language written representation. The methods to produce the final texts may differ from those used by compilers, due to the inherent expressivity of natural languages. ATG is not a recent discipline, even if commercial ATG technology has only recently become widely available. Today, many software environments cope with ATG, as Text Spinner, DKB Lettere, or textOmatic*Composer, to mention just some of them. As a discipline strictly connected to NLP, ATG should be based strongly on morph-syntax formalization and semantic predicate use. However, in some cases it seems possible to avoid these steps. A simple example of ATG not involving the use of morph-syntactic and semantic rules may be the generation of texts using only simple alphabetic letters. This method can prove itself useful when the text to gener-ate is somehow generic in terms of semantics, and fix in terms of syntax. For in-stance, it can be used to generate a letter to a consumer stating that a credit card spending limit has been reached, or also to generate receipts from an ATM machine, or Social Media notifications. However, in theory and practice the automatic generation of more complex texts can only be based on a complete system of Natural Language Formalization (NLF), as for instance Maurice Gross’ Lexicon-Grammar. Therefore, in order to build an ATG procedure for novel plots, in this paper we will use both Lexicon-Grammar theoretical and practical framework and Max Silberztein’s NooJ NLP Environment , which as it is well known are in a strict connection. Starting from Gross’ definition of semantic predicates [1] and from the NooJ paraphrase generation routine [5,6], our aim will be to write automatically the basic plot of a novel. While achieving our aim, we will take into due account that the novel is a kind of writing difficult to define formally [7], and that the automatic writing of a novel plot is probably one of the most complex challenges an ATG routine can choose to tackle. Finally, in carrying out our research, we will also make extensive references to Text Linguistics (TL) and its theoretical and practical contact points with Formal Linguis-tics (FL).

Automatic Text Generation: How to Write the Plot of a Novel with NooJ

Mario Monteleone
2020

Abstract

Automatic Text Generation (ATG) is a Natural Language Processing (NLP) task that aims at writing acceptable and grammatical written text exploiting machine-representation systems, such as for instance knowledge bases, taxonomies and ontologies. In this sense, it is possible to state that an ATG system works like a translator that converts data into a natural-language written representation. The methods to produce the final texts may differ from those used by compilers, due to the inherent expressivity of natural languages. ATG is not a recent discipline, even if commercial ATG technology has only recently become widely available. Today, many software environments cope with ATG, as Text Spinner, DKB Lettere, or textOmatic*Composer, to mention just some of them. As a discipline strictly connected to NLP, ATG should be based strongly on morph-syntax formalization and semantic predicate use. However, in some cases it seems possible to avoid these steps. A simple example of ATG not involving the use of morph-syntactic and semantic rules may be the generation of texts using only simple alphabetic letters. This method can prove itself useful when the text to gener-ate is somehow generic in terms of semantics, and fix in terms of syntax. For in-stance, it can be used to generate a letter to a consumer stating that a credit card spending limit has been reached, or also to generate receipts from an ATM machine, or Social Media notifications. However, in theory and practice the automatic generation of more complex texts can only be based on a complete system of Natural Language Formalization (NLF), as for instance Maurice Gross’ Lexicon-Grammar. Therefore, in order to build an ATG procedure for novel plots, in this paper we will use both Lexicon-Grammar theoretical and practical framework and Max Silberztein’s NooJ NLP Environment , which as it is well known are in a strict connection. Starting from Gross’ definition of semantic predicates [1] and from the NooJ paraphrase generation routine [5,6], our aim will be to write automatically the basic plot of a novel. While achieving our aim, we will take into due account that the novel is a kind of writing difficult to define formally [7], and that the automatic writing of a novel plot is probably one of the most complex challenges an ATG routine can choose to tackle. Finally, in carrying out our research, we will also make extensive references to Text Linguistics (TL) and its theoretical and practical contact points with Formal Linguis-tics (FL).
978-3-030-38833-1
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11386/4733683
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact