A key challenge in computational narrative is story generation. In this paper we focus on analogy-based generation, and, specifically, how to generate both and text using analogy. We present a dual representation formalism where human-understandable (composed of English sentences) computer-understandable (consisting graph) are linked together order natural language by have implemented our technique the Riu interactive system.

Analogy-based Story Generation (ASG) is a relatively under-explored approach for story generation and computational narrative. In this paper, we present the SAM (Story Analogies through Mapping) algorithm as our attempt to expand scope complexity of stories generated by ASG. Comparing with existing work prior work, there are two main contributions SAM: it employs 1) analogical reasoning both at specific content general domain knowledge levels, 2) temporal about (phase) structure in order...

Within interactive narrative research, agency is largely considered in terms of a player's autonomy game, defined as theoretical agency. Rather than whether or not the player feels they have agency, their perceived An effective needs to provide level that satisfies desires and must do without compromising its own structure. Researchers frequently turn techniques for increasing accomplish this. This paper proposes an approach categorize explore which affected requiring more less

In this paper we propose a method for automatically assigning narrative roles to characters in stories. To achieve goal our proposal is combine natural language processing techniques with domain knowledge extracted from Propp's morphology of the folktale. We use matrix that encodes representing interactions between character roles.

Computational analogy offers a promising direction to algorithmically generating stories, key challenge in computational narrative. Since methods are very sensitive the story representation being used, this paper focuses on for analogy-based generation. Specifically, we analyze existing formalisms and propose new approach based cognitive semantics theory of force dynamics. Finally, present results our interactive narrative system, Riu, illustrate utility proposal.

In this paper, we present our method for automatically extracting narrative information of characters and their roles from natural language stories. corpus 15 unannotated folk tales, Voz system identifies 87% the in stories correctly assigns 68% character roles. To better understand sources error system, an analytical methodology to study how is introduced by different modules it propagates through pipeline. This allows us identify bottleneck with largest impact on final error, which might...

Computational narrative is a complex and interesting domain for exploring AI techniques that algorithmically analyze, understand, most importantly, generate stories. This paper studies the importance of knowledge in story generation, particularly analogy-based generation (ASG). Based on construct container case-based reasoning, we present theoretical framework incorporating ASG. We complement with empirical results our existing system Riu.

This paper proposes a Trajectory Based Control (TBC) method for connected automated traffic. The goal is to ensure absolute transit priority without heavy sacrifices from the general To achieve goal, Bus Priority Lane (BPL) designed. BPL open traffic if needed. implement BPL, proposed TBC controller would control whether and how accesses so that sacrifice of minimized. evaluated through comparison against non-control baseline ensures with Dedicated (DBL). Sensitivity analysis conducted under...

Originating from the premise that Generative Adversarial Networks (GANs) enrich creative processes rather than diluting them, we describe an ongoing PhD project proposes to study GANs in a co-creative context. By asking How can be applied co-creation, and doing so, how they contribute fashion design processes? sets out investigate GAN applications further develop them for specific application area of design. We do so by drawing on field mixed-initiative co-creation. Combined with technical...

A new method for automatic volume determination of loose engineering deposits (LEDs) using point clouds collected by terrestrial laser scanning is proposed. The starts with a problem lacking bottom surface when LEDs, assuming that the flat, spatial plane generated based on fitting to surface. Then sample cloud projected this fitted construct subface clouds, which, together initial data, are stitched into complete data good closure, integrity, and accuracy. 3D model established introducing...

Cooperative signal control in the large-scale urban road network has always been a difficult task due to non-stationary problem. This study proposes an asynchronous cooperative framework for enhance stability of practice. The timing plans can be optimized parallel fashion framework. Signal controllers are able respond and adapt environment change timely through queue prediction based on information from neighbor intersections. Additionally, rolling-horizon strategy is introduced improve...

In this paper we present a survey of two the main areas research within field computational narrative, namely narrative analysis and generation. We argue that there is gap between these lines work propose taxonomy models used each. outline potential mappings with goal bridging alleviating authorial bottleneck problem occurring when authoring content for generation systems. Finally discuss related in direction report on our work-in-progress towards an end-to-end system to bridge gap.