Automatic story generation has a long-standing tradition in the field of Artificial Intelligence. The ability to create stories on demand holds great potential for entertainment and education. For example, modern computer games are becoming more immersive, containing multiple story lines and hundreds of characters. This has substantially increased the amount of work required to produce each game. However, by allowing the game to write its own story line, it can remain engaging to the player whilst shifting the burden of writing away from the game’s developers. In education, intelligent tutoring systems can potentially provide students with instant feedback and suggestions of how to write their own stories. Although several approaches have been introduced in the past (e.g., story grammars, story schema and autonomous agents), they all rely heavily on handwritten resources. Which places severe limitations on its scalability and usage. In this thesis we will motivate a new approach to story generation which takes its inspiration from recent research in Natural Language Generation. Whose result is an interactive data-driven system for the generation of children’s stories. One of the key features of this system is that it is end-to-end, realising the various components of the generation pipeline stochastically. Knowledge relating to the generation and planning of stories is leveraged automatically from corpora and reformulated into new stories to be presented to the user. We will also show that story generation can be viewed as a search task, operating over a large number of stories that can be generated from knowledge inherent in a corpus. Using trainable scoring functions, our system can search the story space using different document level criteria. In this thesis we focus on two of these, namely, coherence and interest. We will also present two major paradigms for generation through search, (a) generate and rank, and (b) genetic algorithms. We show the effects on perceived story interest, fluency and coherence that result from these approaches. In addition, we show how the explicit use of plots induced from the corpus can be used to guide the generation process, providing a heuristically motivated starting point for story search. We motivate extensions to the system and show that additional modules can be used to improve the quality of the generated stories and overall scalability. Finally we highlight the current strengths and limitations of our approach and discuss possible future approaches to this field of research.

This research examines how the contents and structure of a story may be enriched by computational means. A review of pertinent semantic theory and previous work on the structural analysis of folktales is presented. Merits and limitations of several content-generation systems are discussed. The research develops three mechanisms - elaboration, interpolation, and continuity fixes - to enhance story content, address issues of rigid structure, and fix problems with the logical progression of a story. Elaboration works by adding or modifying information contained within a story to provide detailed descriptions of an event. Interpolation works by adding detail between high-level story elements dictated by a story grammar. Both methods search for appropriate semantic functions contained in a lexicon. Rules are developed to ensure that the selection of functions is consistent with the context of the story. Control strategies for both mechanisms are proposed that restrict the quantity and content of candidate functions. Finally, a method of checking and correcting inconsistencies in story continuity is proposed. Continuity checks are performed using semantic threads that connect an object or character to a sequence of events. Unexplained changes in state or location are fixed with interpolation. The mechanisms are demonstrated with simple examples drawn from folktales, and the effectiveness of each is discussed. While the thesis focuses on folktales, it forms the basis for further work on the generation of more complex stories in the greater realm of fiction. / Thesis (Ph.D, Computing) -- Queen's University, 2010-10-12 11:24:33.536

No description available.

Digital media has created more opportunities for stories to be interactive, allowing the user to participate in plot events and even change the direction of a story. Interactivity can make stories more engaging however if the plot can change as a result of user input then the story becomes non-linear. The resulting system is known as a potential narrative because a narrative is only presented through user interaction. Non-linear stories require extra considerations and content compared to a linear story of similar length. Computer generated content for video games has grown as a field and this raises the possibility of using computer assistance for the creation and management of interactive stories. This thesis explores the paradigm required for both computers and developers to understand potential narratives. The success of a potential narrative requires the same coherency and credibility that a regular story does, in addition to a new layer known as player agency. Coherency refers to the logical causal progression of a plot, credibility is the verisimilitude of the presented story world and player agency is how satisfying the interactive elements are. These three criteria are the guiding principles for the creation of an intuitive potential narrative model. We present a state transition model for potential narratives as well as a prototype using relational programming to generate coherent traversals through a manually authored potential narrative. / Thesis / Master of Applied Science (MASc) / Writing stories for video games and interactive fiction is a time consuming and complex affair. The resulting web of possible events and pathways through the non-linear story is known as a potential narrative. Computer assisted tools for creating potential narratives could cut down on production costs and open the door for stories that are able to adapt to the player in real-time. In order for these generated stories to be viable they must be coherent, credible and allow for player agency. These criteria describe whether stories make logical sense and are satisfying for the user to interact with. Taking that into account we present a model for coherent interactive stories which keeps track of the story world and the events that could happen within it. As a proof of concept we implement this model with a prototype to generate traversals through a manually authored potential narrative.

This project examined the story generations of five children with language impairment (LI) between the ages of 5;11 and 10;1 across the treatment sessions of a narrative-based intervention program designed to improve social communication. These stories were analyzed to find whether the participants would approach the task by describing the stimulus pictures or if their stories would reflect an episodic structure containing cause and effect relationships. Additionally, the stories were analyzed for inclusion of emotion words to discern the participants' awareness of the characters' emotions. There was a high degree of variability in the participants' performance; however, the majority of the stories were composed of picture descriptions, and most of the participants generated short stories with few episodic elements in response to the probes over the course of treatment. In terms of emotion word use, two of the participants increased their use of emotion words in later sessions. Overall the participants' performance on the story generation probe did not reflect their performance in other treatment tasks including shared book readings, story enactments, and journal writing. This was likely due to their disinterest or fatigue in the story generation task and stimuli, as well as their continued need for the clinician modeling that was present in the other treatment tasks. Future research is needed to determine effective treatments that help school-aged children with LI recognize goal-directed behavior and emotional content in stories.

Stories are complex linguistic constructions through which we share our interpretations of the social world. The ability to comprehend and produce stories is referred to as narrative competence. Narrative competence is rooted in social interactions in the preschool years that foster a sense of story structure and familiarity with story language. It has been shown to play a critical role in reading and writing achievement in the elementary school years since the language of literacy, like the language of storytelling, is predominantly decontextualized. The mastery of decontextualized language poses a significant challenge for children who enter kindergarten with little previous exposure to the majority language. The storytelling ability of these children was the focus of the present study. The study's primary aim was to examine second language narrative competence across the kindergarten years by comparing the fictional stories generated by a cross-section of EL1 and ELL junior and senior kindergarten children in response to a wordless picture book from the point of view of macrostructure (story structure), microstructure (story language), and the use of evaluative language. Grade and gender differences across and within language groups were also considered. A second aim of the study was to examine the relationship between narrative competence and receptive vocabulary and between narrative competence and print-based emergent reading skill. Overall, the results suggested that the ELL children's narratives were comparable to those of their EL1 counterparts with respect to most measures of microstructure, and with respect to macrostructure and evaluative language use. The one clear language-based difference favouring the EL1 children related to morpho-syntactic quality. Age-related differences were obtained on most measures and the results suggested parallel developmental trajectories across language groups. Gender was found to play a more prominent role in ELL than EL1 narrative performance. Few aspects of narrative were predicted by receptive vocabulary, suggesting dissociation between word- and discourse-level skills, particularly among the ELL children. On the other hand, emergent literacy scores predicted several aspects of microstructure, macrostructure and evaluative language use. The study provides evidence that various aspects of narrative competence might be differentially related to vocabulary and emergent literacy skills in ELL and EL1 kindergarten children.

Stories are complex linguistic constructions through which we share our interpretations of the social world. The ability to comprehend and produce stories is referred to as narrative competence. Narrative competence is rooted in social interactions in the preschool years that foster a sense of story structure and familiarity with story language. It has been shown to play a critical role in reading and writing achievement in the elementary school years since the language of literacy, like the language of storytelling, is predominantly decontextualized. The mastery of decontextualized language poses a significant challenge for children who enter kindergarten with little previous exposure to the majority language. The storytelling ability of these children was the focus of the present study. The study's primary aim was to examine second language narrative competence across the kindergarten years by comparing the fictional stories generated by a cross-section of EL1 and ELL junior and senior kindergarten children in response to a wordless picture book from the point of view of macrostructure (story structure), microstructure (story language), and the use of evaluative language. Grade and gender differences across and within language groups were also considered. A second aim of the study was to examine the relationship between narrative competence and receptive vocabulary and between narrative competence and print-based emergent reading skill. Overall, the results suggested that the ELL children's narratives were comparable to those of their EL1 counterparts with respect to most measures of microstructure, and with respect to macrostructure and evaluative language use. The one clear language-based difference favouring the EL1 children related to morpho-syntactic quality. Age-related differences were obtained on most measures and the results suggested parallel developmental trajectories across language groups. Gender was found to play a more prominent role in ELL than EL1 narrative performance. Few aspects of narrative were predicted by receptive vocabulary, suggesting dissociation between word- and discourse-level skills, particularly among the ELL children. On the other hand, emergent literacy scores predicted several aspects of microstructure, macrostructure and evaluative language use. The study provides evidence that various aspects of narrative competence might be differentially related to vocabulary and emergent literacy skills in ELL and EL1 kindergarten children.

Narrative Intelligence is the ability to craft, tell, understand, and respond appropriately to narratives. It has been proposed as a vital component of machines aiming to understand human activities or to communicate effectively with humans. However, most existing systems purported to demonstrate Narrative Intelligence rely on manually authored knowledge structures that require extensive expert labor. These systems are constrained to operate in a few domains where knowledge has been provided. This dissertation investigates the learning of knowledge structures to support Narrative Intelligence in any domain. I propose and build a system that, from an corpus of simple exemplar stories, learns complex knowledge structures that subsequently enable the creation, telling, and understanding of narratives. The knowledge representation balances the complexity of learning and the richness of narrative applications, so that we can (1) learn the knowledge robustly in the presence of noise, (2) generate a large variety of highly coherent stories, (3) tell them in recognizably different narration styles and (4) understand stories efficiently. The accuracy and effectiveness of the system have been verified by a series of user studies and computational experiments. As a result, the system is able to demonstrate Narrative Intelligence in any domain where we can collect a small number of exemplar stories. This dissertation is the first step toward scaling computational narrative intelligence to meet the challenges of the real world.

[pt] Este trabalho visa explorar os requisitos de software, descritos na forma de casos de uso, como instrumento capaz de apoiar a geração e execução automática de testes funcionais, com o intuito de verificar automaticamente se o resultado obtido nos testes gerados e executados estão em conformidade com o especificado. O trabalho consiste em utilizar um processo e ferramenta para documentar casos de uso e automaticamente gerar e executar scripts de teste para verificar o comportamento funcional de aplicações web. As informações do caso de uso, em especial os fluxos de eventos (principal e alternativos), devem ser estruturados obedecendo um modelo de comportamento para que seja possível armazenar os dados e utilizá-los como entrada na integração com a ferramenta de testes. Neste trabalho, utilizou-se a ferramenta Selenium para a automação da interação com o navegador. A avaliação do esforço deu-se através da aplicação do processo em sistemas reais e através de comparação com outras técnicas aplicadas nos mesmos sistemas. / [en] This work aims at exploring the software requirements, described in the form of use cases, as an instrument to support the automatic generation and execution of functional tests, in order to automatically check if the results obtained in the tests generated and executed are in accordance with specified. The establishes a process and a tool for documenting use cases and automatically generating and executing test scripts that verify the behavior of web applications. The content of the use case, especially the flow of events (main and alternative), is structured in accordance to a "behavior model" that stores the test data and generates input for a browser testing tool. In this work, we have used the Selenium tool to automate the interaction with the browser. The assessment of our approach involved the application of the process and generating tool in real systems, comparing the results with other techniques applied in the same systems.

The design of autonomous agents capable of performing a given goal in a 3D domain continues to be a challenge for computer animated story generation systems. We present a novel prototype which consists of a 3D engine and a planner for a simple virtual world. We incorporate the 2D planner into the 3D engine to provide 3D animations. Based on the plan, the 3D world is created and the objects are positioned. Then the plan is linearized into simpler actions for object animation and rendered via the 3D engine. We use JINNI3D as the engine and WARPLAN-C as the planner for the above-mentioned prototype. The user can interact with the system using a simple natural language interface. The interface consists of a shallow parser, which is capable of identifying a set of predefined basic commands. The command given by the user is considered as the goal for the planner. The resulting plan is created and rendered in 3D. The overall system is comparable to a character based interactive story generation system except that it is limited to the predefined 3D environment.