Say Anything: A Massively Collaborative Open Domain Story Writing Companion

Abstract

Abstract.Interactive storytelling is an interesting cross-disciplinary area that has importance in research as well as entertainment. In this paper we explore a new area of interactive storytelling that blurs the line between traditional interactive fiction and collaborative writing. We present a system where the user and computer take turns in writing sentences of a fictional narrative. Sentences contributed by the computer are selected from a collection of millions of stories extracted from Internet weblogs. By leveraging the large amounts of personal narrative content available on the web, we show that even with a simple approach our system can produce compelling stories with our users.

A Data-Driven Case-Based Reasoning Approach to Interactive Storytelling

“…Although Swanson & Gordon [16] showed that even a simple technique like the one described in the previous section can be quite effective, many of its problems should be readily apparent. The first thing to notice is a limitation of the IR mechanism itself.…”

“…The style of game-play and system architecture described in this paper follows the Say Anything system developed by Swanson & Gordon [16]. Similar to their system, our system is a textual, turn-based game in which a human participant and a computer agent alternate writing sentences of a developing narrative.…”

“…Traditionally, these case libraries have been built by hand, however this will not provide us with the depth of content we been arguing for. Instead, we follow a similar approach to Swanson & Gordon [16] and mine the social web for personal stories written by ordinary people on their weblogs. We roughly define a personal story to be a first person description of temporal and causally connected sequence of events that the author actively participated in.…”

“…We roughly define a personal story to be a first person description of temporal and causally connected sequence of events that the author actively participated in. For a more complete definition see [16].…”

“…Swanson & Gordon [16] first proposed a large-scale data driven approach that decouples the authoring process from the narrative decision process by completely separating the content authoring from the mechanics of the narrative engine. Instead of using a relatively small knowledge-base of highly structured rules and content, they leveraged the massive amounts of natural language personal stories on the Web.…”