Abstract This paper proposes a basic and general model of narrative structure representation for computer processing in order to provide a common basis for both generative and analytic processing of narratives. In particular, we address the following two aspects. First, we formalize a multidimensional framework to integrate the three structural aspects of a narrative: the story world, which is the background world structure; the... This paper proposes a basic and general model of narrative structure representation for computer processing in order to provide a common basis for both generative and analytic processing of narratives. In particular, we address the following two aspects. First, we formalize a multidimensional framework to integrate the three structural aspects of a narrative: the story world, which is the background world structure; the story, which represents chronologically organized events as the content of a narrative; and discourse, which represents the structure used for the expression of a narrative. Second, we propose a hierarchical graph model as a common mode of representing each structural dimension. This model organizes various types of structural elements (e.g., entities, events, and discourse units) by hierarchical grouping and network-like linkage. Then, we discuss the advantages and future challenges of the proposed model from the perspective of representational capacity by presenting several examples based on the experimental implementation of the model. Read More Read Less

Abstract In this study, we examined ideophones (onomatopoeias or sound symbolism) in taste descriptions of wine and sake using quantitative text analysis and morphological analysis to investigate how sound symbolic expressions function (word count of wine text: 201,294; word count of sake text: 50,147).We found the following co-occurrence tendencies for Japanese ideophones in the tasting corpora: ideophones are used to modify words for taste,... In this study, we examined ideophones (onomatopoeias or sound symbolism) in taste descriptions of wine and sake using quantitative text analysis and morphological analysis to investigate how sound symbolic expressions function (word count of wine text: 201,294; word count of sake text: 50,147).We found the following co-occurrence tendencies for Japanese ideophones in the tasting corpora: ideophones are used to modify words for taste, rather than words for brewing or grades of wines and sakes. More specifically, ideophones are used to express the sense of ‘appearing’ or ‘finishing’ of flavours (specifically in the Sake Corpus). Morphological analysis of these texts in comparison to the BCCWJ (The Balanced Corpus of Contemporary Written Japanese) revealed that double consonant patterns are a characteristic form in the tasting descriptions. Integrating these findings with those of the text analysis, we concluded that the ideophones are employed to describe breaking points, turning points, and changing processes of taste, rather than stable states. Therefore, ideophones can be distinguished from general adverbs, which mainly describe the states and manners of objects or events. Read More Read Less

Abstract Computational approaches to metaphor processing and identification are gaining popularity. In this paper, I will discuss the nature of a domain from a Cognitive Linguistic perspective, the kinds of domains used in sample computational approaches, and how they arrive at them.... Computational approaches to metaphor processing and identification are gaining popularity. In this paper, I will discuss the nature of a domain from a Cognitive Linguistic perspective, the kinds of domains used in sample computational approaches, and how they arrive at them. Read More Read Less

Abstract Quantitative indicators of narrative analysis can be used to enhance the objectivity of story analysis. However, conventional narrative descriptions are not suitable for extracting the narrative functions of a story as a whole, nor have complex stories with parallel narrative structures been analyzed. With the ultimate goal of developing an automatic plot analysis method suitable for describing parallel storylines and punchlines, this paper... Quantitative indicators of narrative analysis can be used to enhance the objectivity of story analysis. However, conventional narrative descriptions are not suitable for extracting the narrative functions of a story as a whole, nor have complex stories with parallel narrative structures been analyzed. With the ultimate goal of developing an automatic plot analysis method suitable for describing parallel storylines and punchlines, this paper proposes a prototype algorithm for automatic plot structure extraction focusing on agent vocabulary and behavior expressions. In the proposed prototype algorithm, categories and dictionaries for identifying agents and behaviors in story texts are constructed in Japanese and algorithms for extracting propositions and divisions of scenes are implemented based on the resulting agent and behavior vocabulary and expressions. Further, omitted agents are reproduced based on knowledge about agents and general rules of discourse. This enables the processing of fundamental data about story structure and the categorization of behaviors and scene divisions. Read More Read Less