Computational modelling of poetry generation

2004

Poetry is a unique artifact of the human language faculty, with its defining feature being a strong unity between content and form. Contrary to the opinion that the automatic generation of poetry is a relatively easy task, we argue that it is in fact an extremely difficult task that requires intelligence, world and linguistic knowledge, and creativity. We propose a model of poetry generation as a state space search problem, where a goal state is a text that satisfies the three properties of meaningfulness, grammaticality, and poeticness.

Connection Science, 2016

Most poetry-generation systems apply opportunistic approaches where algorithmic procedures are applied to explore the conceptual space defined by a given knowledge resource in search of solutions that might be aesthetically valuable. Aesthetical value is assumed to arise from compliance to a given poetic form-such as rhyme or metrical regularity-or from evidence of semantic relations between the words in the resulting poems that can be interpreted as rhetorical tropes-such as similes, analogies, or metaphors. This approach tends to fix a priori the aesthetic parameters of the results, and imposes no constraints on the message to be conveyed. The present paper describes an attempt to initiate a shift in this balance, introducing means for constraining the output to certain topics and allowing a looser mechanism for constraining form. This goal arose as a result of the need to produce poems for a themed collection commissioned to be included in a book. The solution adopted explores an approach to creativity where the goals are not solely aesthetic and where the results may be surprising in their poetic form. An existing computer poet, originally developed to produce poems in a given form but with no specific constraints on their content, is put to the task of producing a set of poems with explicit restrictions on content, and allowing for an exploration of poetic form. Alternative generation methods are devised to overcome the difficulties, and the various insights arising from these new methods and the impact they have on the set of resulting poems are discussed in terms of their potential contribution to better poetry-generation systems.

ASPERA is a case based reasoning application that generates poetry versions of texts provided by the user. The cases on its case base consist of a sentence of prose (used as retrieval key) associated with a corresponding poem fragment (used as starting point for the solution). Each case includes information about how the words in the prose text are related to the words in the poetry solution. The system applies a generation process guided by metrical rules to adapt its best matching cases to the sentences in the text provided by the user. ASPERA is a refinement of an earlier system and its casebase is currently being extended. It is hoped that soon it will have enough coverage for the system to become fully operational. The issue of whether such as system should be creative (surprising the user with poems not explicitly determined by his proposal) or faithful (keeping as close as possible to the proposal given) is discussed with a view to including in the system a facility to control its behaviour in this respect.

2000

In this paper we describe a flexible approach to natural language generation that employs a stochastic hillclimbing search algorithm and an integrated architecture. We then discuss the benefits of this approach over existing, informative, goal-driven generation systems. We choose the generation of poetry as our research task domain, as it is a prime example of natural language that demands the degree of flexibility provided by our approach. Finally, we report and discuss results of our preliminary implementation work.

We present a system to generate poems based on the information extracted from input text such as blog posts. Our design uses the blackboard architecture , in which independent specialized modules cooperate during the generation process by sharing a common workspace known as the blackboard. Each module is responsible for a particular task while generating poetry. Our implementation incorporates modules that retrieve information from the input text, generate new ideas, or select the best partial solutions. These distinct modules (experts) are implemented as diverse computational units that make use of lexical resources, grammar models, sentiment-analyzing tools, and language-processing algorithms. A control module is responsible for scheduling actions on the blackboard. We argue that the blackboard architecture is a promising way of simulating creative processes because of its flexibility and compliance with the Global Workspace Theory of mind. The main contribution of this work is the design and prototype implementation of an extensible platform for a poetry-generating system that may be further extended by incorporating new experts as well as some existing poetry-generating systems as parts of the blackboard architecture. We claim that this design provides a powerful tool for combining many of the existing efforts in the domain of automatic poetry generation.

Proceedings of the …, 2012

We describe a corpus-based poetry generation system which uses templates to construct poems according to given constraints on rhyme, meter, stress, sentiment, word frequency and word similarity. Moreover, the software constructs a mood for the day by analysing newspaper articles; uses this to determine both an article to base a poem on and a template for the poem; creates an aesthetic based on relevance to the article, lyricism, sentiment and flamboyancy; searches for an instantiation of the template which maximises the aesthetic; and provides a commentary for the whole process to add value to the creative act. We describe the processes behind this approach, present some experimental results which helped in fine tuning, and provide some illustrative poems and commentaries. We argue that this is the first poetry system which generates examples, forms concepts, invents aesthetics and frames its work, and so can be assessed favourably with respect to the FACE model for comparing creative systems.

2009

Poetry is one of the most interesting and complex natural language generation (NLG) systems, because a text needs to simultaneously satisfy three properties, to be considered a poem; namely poeticness (poetic structure), grammaticality (grammatical structure and syntax) and meaningfulness (semantic content). In this thesis, we discuss the development and implementation of an autonomous system, capable of generating unique yet meaningful poetry, that harnesses the advantages of Constraint Handling Rules (CHR). This is realized through the implementation of a reasoner, which generates poems, that satisfy poeticness, grammaticality and meaningfulness, based on a customized lexicon. In the proposed system, a poem is generated by incrementally selecting its words, through a step-wise pruning of the lexicon by the reasoner. This is done based on the constraints, that represent poeticness, grammaticality and meaningfulness. The developed approach proves, that CHR can be used to develop a h...

Academia Letters, 2021

ToTransformer, Bot or Not are just some of them. Most of the above sites and projects are based on one of the branches of AI called machine learning. According to machine learning, the right algorithms can help a computer selfimprove as it acquires more experience. 1 Thus, a computer can be "trained" to produce what is required of it, in this instance poems, based on a body of data (e.g. the poems of an existing poet) and an algorithm that explains to it how it can use them (which would include elements such as the number of stanzas or lines in a stanza; this is the template-based approach) 2 or just by putting together random words based on typical requirements, e.g. the desired number of syllables for rhythm (generate and test approach). Results have been found to range from simply bad to fairly good, in which latter case they have been produced by chance. But why poetry? Why have scientists who already work on AI speech synthesis chosen poetry out of all literary genres? Why not choose another short from, e.g. bonsai short stories (with a typical ceiling of around 600 words)? The answer lies in the characteristics of each genre. A short story, however small, must have a structure: characters, dialogues, plot. Poetry,

Proceedings of the 5th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2021), 2021

In this work, we design an end-to-end model for poetry generation based on conditioned recurrent neural network (RNN) language models whose goal is to learn stylistic features (poem length, sentiment, alliteration, and rhyming) from examples alone. We show this model successfully learns the ‘meaning' of length and sentiment, as we can control it to generate longer or shorter as well as more positive or more negative poems. However, the model does not grasp sound phenomena like alliteration and rhyming, but instead exploits low-level statistical cues. Possible reasons include the size of the training data, the relatively low frequency and difficulty of these sublexical phenomena as well as model biases. We show that more recent GPT-2 models also have problems learning sublexical phenomena such as rhyming from examples alone.

PoeTryMe is a poetry generation platform under development that intends to help the automatic generation of meaningful poetry according to a given semantics. It has a versatile architecture that provides a high level of customisation where the user can define features that go from the base semantics and sentence templates to the generation strategy and the poem configuration. A prototype using PoeTryMe was implemented to generate Portuguese poetry. The results are interesting but there is still a long way for improvement, so we devise ideas for future work.