Figure - uploaded by Mika Hämäläinen
Content may be subject to copyright.
Mean and sample standard deviation of evaluation scores, and number of noun-property pairs evaluated
Source publication
Many linguistic creativity applications rely heavily on knowledge of nouns and their properties. However, such knowledge sources are scarce and limited. We present a graph-based approach for expanding and weighting properties of nouns with given initial, non-weighted properties. In this paper, we focus on famous characters, either real or fictional...
Contexts in source publication
Context 1
... mean scores are given in Table 1. The strong properties have a mean score of 4.13, weak properties have 3.60 and random properties 2.18. ...Context 2
... Deviation of Scores Additional standard deviations of the scores (Table 1) can provide insight to the degree of agreement between judges. We can see that strong properties are typically more agreed on (have smaller standard deviation); however, in the case of real characters judges seem to have had slightly diverse opinions. ...Similar publications
An implicit and metaphorical identification of learning with work activity frames the language used to construct learning and the orientation of the teaching practices foregrounded in formal educational settings. Play practices-improvisations, simulations, creative writing, games-as exploratory and creative processes are sometimes present in the L2...
Citations
... The reviewed literature on story generation does not relate adequate stylistic resources with particular communicative goals. Nevertheless, there is already a relevant bulk of work on some stylistic resources that may be used in generated stories, especially those related to figurative language (e.g., [196][197][198][199]). We suggest that another important research direction is the study of the most effective stylistic resources for each possible communicative goal. ...
We present a literature survey on non-interactive computational story generation. The article starts with the presentation of requirements for creative systems, three types of models of creativity (computational, socio-cultural, and individual), and models of human creative writing. Then it reviews each class of story generation approach depending on the used technology: story-schemas, analogy, rules, planning, evolutionary algorithms, implicit knowledge learning, and explicit knowledge learning. Before the concluding section, the article analyses the contributions of the reviewed work to improve the quality of the generated stories. This analysis addresses the description of the story characters, the use of narrative knowledge including about character believability, and the possible lack of more comprehensive or more detailed knowledge or creativity models. Finally, the article presents concluding remarks in the form of suggestions of research topics that might have a significant impact on the advancement of the state of the art on autonomous non-interactive story generation systems. The article concludes that the autonomous generation and adoption of the main idea to be conveyed and the autonomous design of the creativity ensuring criteria are possibly two of most important topics for future research.
... The Non-Official Characterization (NOC) list (Veale 2016) which contains information about more than 1000 well-known characters (e.g. Donald Trump and Kim Jung-un) and their expanded stereotypical properties supplied by (Alnajjar et al. 2017) (e.g. Donald Trump: ...
Automated news generation has become a major interest for new agencies in the past. Oftentimes headlines for such automatically generated news articles are unimaginative as they have been generated with ready-made templates. We present a computationally creative approach for headline generation that can generate humorous versions of existing headlines. We evaluate our system with human judges and compare the results to human authored humorous titles. The headlines produced by the system are considered funny 36\% of the time by human evaluators.
... The Non-Official Characterization (NOC) list (Veale 2016) which contains information about more than 1000 well-known characters (e.g. Donald Trump and Kim Jung-un) and their expanded stereotypical properties supplied by (Alnajjar et al. 2017) (e.g. Donald Trump: ...
Automated news generation has become a major interest for new agencies in the past. Oftentimes headlines for such automatically generated news articles are unimaginative as they have been generated with ready-made templates. We present a computationally creative approach for headline generation that can generate humorous versions of existing headlines. We evaluate our system with human judges and compare the results to human authored humorous titles. The headlines produced by the system are considered funny 36% of the time by human evaluators.
... NOC-listan sisältämiä piirteitä on laajennettu automaattisesti datalähtöisin menetelmin (Alnajjar et al., 2017 Semanttista läheisyyttä mallintaaksemme rakennamme logaritmisen todennäköisyysmatriisin (log-likelihood) termien välisistä suhteista, joka perustuu termien yhteisesiintymiin korpuksessa käyttäen Meta4Meaning-mallia (Xiao et al., 2016). Kuvataksemme semanttista samankaltaisuutta käytämme word2vec-mallia (Mikolov et al., 2013) Vesi-Pokémoneille swime (huimaava, arkaainen merkitys) saa korkeat pisteet johtuen lähinnä siitä, että se on lähellä sanaa swim (uida). ...
Tässä artikkelissa kokeilemme erilaisia menetelmiä kuvaavien piirteiden tuottamiseksi 151:lle alkuperäiselle Pokémonille. Tuotamme eri menetelmillä sanavektorimalleja nettikorpuksen avulla, ja luokittelemme niillä automaattisesti englannin kielen adjektiiveja sen perusteella, kuinka ominaisia ne ovat tietylle Pokémonille. Kokeidemme perusteella voimme todeta, että sanavektorimalli toimii vain, jos se on tuotettu Pokémon-korpuksen avulla. Valmiiksi tuotetut mallit eivät pysty kuvaamaan Pokémoneja hyvin. Word2Vec-malli tuottaa parempia tuloksia kuin fastText-malli. Lisäksi kokeilemme laajentaa jokaisen Pokémonin piirteiden luetteloa automaattisesti. Mikään menetelmistä ei kuitenkaan toimi riittävällä tarkkuudella, ja eri semanttisissa malleissa on kaikissa huomattava määrä ongelmia.
... There has been an automated effort for expanding the properties recorded in the NOC list (Alnajjar et al., 2017). While this method is a step towards the desired direction in the sense that it does not require the nouns to exist in a massive corpus, it still relies on mined associations between adjectival properties and a hand annotated list of properties for famous characters in order to expand them further. ...
... Furthermore, we experiment with an existing method for expanding properties for the results of each method. The property expansion is based on the data and algorithm presented by Alnajjar et al. (2017). The method takes in a list of properties and produces an extended property list by using Thesaurus Rex data (Veale and Li, 2013). ...
We present different methods for obtaining descriptive properties automatically for the 151 original Pokémon. We train several different word embeddings models on a crawled Pokémon corpus, and use them to rank automatically English adjectives based on how characteristic they are to a given Pokémon. Based on our experiments, it is better to train a model with domain specific data than to use a pretrained model. Word2Vec produces less noise in the results than fastText model. Furthermore, we expand the list of properties for each Pokémon automatically. However, none of the methods is spot on and there is a considerable amount of noise in the different semantic models. Our models have been released on Zenodo.
... Figurative language is a form of indirect communication, where the meaning of a sentence cannot be derived by the individual semantic meanings of the words. Figurative language, and especially metaphors, have been widely studied in the field of natural language processing (NLP) from the point of view of detection, interpretation and generation (Rai and Chakraverty 2020;Veale and Li 2013;Galván et al. 2016;Alnajjar et al. 2017). ...
... Meta4meaning is capable of interpreting metaphors if the tenor and vehicle occur in the corpus with other words adequately. To overcome this obstacle, an extension to Meta4meaning has been suggested by Alnajjar et al. (2017) where (a few) adjectival properties of rare concepts such as famous proper nouns are automatically expanded and weighted using automatically mined links between adjectives, allowing the method to produce interpretations to metaphors like "Hillary Clinton is a cat". Moreover, Bar, Dershowitz, and Dankin (2018) extended Meta4meaning by testing out different modifications such as using syntactic dependencies to obtain collocations and applying clustering. ...
... As part of Paper III, we research the idea of applying a metaphor interpretation model (Meta4meaning) to pick metaphors generated by a simple metaphor generation method that produces metaphors from word associations obtained from the knowledge bases shared by Veale and Li (2013) and Alnajjar et al. (2017). The goal of employing the metaphor interpretation model is to identify apt metaphors that are more likely to result in the desired meaning. ...
Computational creativity has received a good amount of research interest in generating creative artefacts programmatically. At the same time, research has been conducted in computational aesthetics, which essentially tries to analyse creativity exhibited in art. This thesis aims to unite these two distinct lines of research in the context of natural language generation by building, from models for interpretation and generation, a cohesive whole that can assess its own generations.
I present a novel method for interpreting one of the most difficult rhetoric devices in the figurative use of language: metaphors. The method does not rely on hand-annotated data and it is purely data-driven. It obtains the state of the art results and is comparable to the interpretations given by humans. We show how a metaphor interpretation model can be used in generating metaphors and metaphorical expressions.
Furthermore, as a creative natural language generation task, we demonstrate assigning creative names to colours using an algorithmic approach that leverages a knowledge base of stereotypical associations for colours. Colour names produced by the approach were favoured by human judges to names given by humans 70% of the time.
A genetic algorithm-based method is elaborated for slogan generation. The use of a genetic algorithm makes it possible to model the generation of text while optimising multiple fitness functions, as part of the evolutionary process, to assess the aesthetic quality of the output. Our evaluation indicates that having multiple balanced aesthetics outperforms a single maximised aesthetic.
From an interplay of neural networks and the traditional AI approach of genetic algorithms, we present a symbiotic framework. This is called the master-apprentice framework. This makes it possible for the system to produce more diverse output as the neural network can learn from both the genetic algorithm and real people.
The master-apprentice framework emphasises a strong theoretical foundation for the creative problem one seeks to solve. From this theoretical foundation, a reasoned evaluation method can be derived. This thesis presents two different evaluation practices based on two different theories on computational creativity. This research is conducted in two distinct practical tasks: pun generation in English and poetry generation in Finnish.
... The method begins by retrieving nouns associated with the input property P using two resources: Thesaurus Rex (Veale and Li 2013) is used to obtain general nouns such as coffee or flower, while the resource by Alnajjar et al. (2017) provides human categories such as actor, lawyer, or politician. The former will be used for generating general metaphors and the latter for personifications. ...
In advertising, slogans are used to enhance the recall of the advertised product by consumers and to distinguish it from others in the market. Creating effective slogans is a resource-consuming task for humans. In this paper, we describe a novel method for automatically generating slogans, given a target concept (e.g., car) and an adjectival property to express (e.g., elegant) as input. Additionally, a key component in our approach is a novel method for generating nominal metaphors, using a metaphor interpretation model, to allow generating metaphorical slogans. The method for generating slogans extracts skeletons from existing slogans. It then fills a skeleton in with suitable words by utilizing multiple linguistic resources (such as a repository of grammatical relations, and semantic and language models) and genetic algorithms to optimize multiple objectives such as semantic relatedness, language correctness, and usage of rhetorical devices. We evaluate the metaphor and slogan generation methods by running crowdsourced surveys. On a five-point Likert scale, we ask online judges to evaluate whether the generated metaphors, along with three other metaphors generated using different methods, highlight the intended property. The slogan generation method is evaluated by asking crowdsourced judges to rate generated slogans from five perspectives: (1) how well is the slogan related to the topic, (2) how correct is the language of the slogan, (3) how metaphoric is the slogan, (4) how catchy, attractive, and memorable is it, and (5) how good is the slogan overall. Similarly, we evaluate existing expert-made slogans. Based on the evaluations, we analyze the method and provide insights regarding existing slogans. The empirical results indicate that our metaphor generation method is capable of producing apt metaphors. Regarding the slogan generator, the results suggest that the method has successfully produced at least one effective slogan for every evaluated input.
