Santiago Ontañón1, Josep Valls-Vargas1 and Jichen Zhu2

1Computer Science, 2Digital MediaDrexel University

Bridging the Gap Between Computational Narrative and Natural Language Processing

February 4th 2017 – AAAI W17: What's Next for AI in Games?

Introduction

Narratology

Artificial Intelligence

Natural Language Processing

2

ComputationalNarrative

Motivation

3

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Joe Bear was hungry. He asked Irving Bird where some honey was. Irving refused to tell him so Joe offered to bring him a worm if he’d tell him where some honey was. Irving agreed. But Joe didn’t know where any worms were, so he asked Irving, who refused to say.

Tale-spin [Meehan 1976], ASD [Riedl 2011], Opiate [Fairclough 2007]

Authorial Bottleneck Problem

4

Opiate [Fairclough 2007]

Narrative FunctionSequences

Characters, Attitudes, …

Locations, Props, …

• Input required by OPIATE

Motivation

5

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

Story Workbench [Finlayson 2011], Scheherazade [Elson 2012]

Motivation

6

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

ProppASM [Finlayson 2011], Social Networks [Elson 2010]

Motivation

7

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

How can we bridge the “gap” in computational narrative in order to solve the authorial bottleneck problem?

PCG & Narrative Systems

Content & Experiences

Open Problems

• How to model narrative?

8

• “Gap” in Computational Narrative• Authorial bottleneck

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Open Problems

• How to model narrative?• How to adapt and reuse general purpose NLP/IE?

9

• “Gap” in Computational Narrative• Authorial bottleneck

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Open Problems

• How to model narrative?• How to adapt and reuse general purpose NLP/IE?• How would an author use a NL interface to a CN system?

10

• “Gap” in Computational Narrative• Authorial bottleneck

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Open Problems

• How to model narrative?• How to adapt and reuse general purpose NLP/IE?• How would an author use a NL interface to a CN system?• How to evaluate narrative IE systems?

11

• “Gap” in Computational Narrative• Authorial bottleneck

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Conclusions

12

Narrative Comp. Models of Narrative

New Content

Natural Language Processing

Procedural Content

Generation

PCG & Narrative Systems

Content & Experiences

Thanks

13

Santiago Ontañón1, Josep Valls-Vargas1 and Jichen Zhu2

1Computer Science, 2Digital MediaDrexel University

Bridging the Gap Between Computational Narrative and Natural Language Processing

February 4th 2017 – AAAI W17: What's Next for AI in Games?

Backup Slides

15

16

Neural all the things!

Authorial Bottleneck Problem

17

Opiate [Fairclough 2007]

Narrative FunctionSequences

Characters, Attitudes, …

Locations, Props, …

• Input required by OPIATE

Once upon a time, Bonji ran into Lili, Mimo and Bibi, three friends who lived in a hut. In a field nearby lived Snomm who had a Magic Mirror. Past the field and further into the woods lived Blobar. In the other side of the woods there was a little town where Sergeant Lip and Corporal Foot lived. They stole the Magic Mirror. [...]

Authorial Bottleneck Problem

18

Opiate [Fairclough 2007]

Narrative FunctionSequences

Characters, Attitudes, …

Locations, Props, …

• Input required by OPIATE

Once upon a time, Bonji ran into Lili, Mimo and Bibi, three friends who lived in a hut. In a field nearby lived Snomm who had a Magic Mirror. Past the field and further into the woods lived Blobar. In the other side of the woods there was a little town where Sergeant Lip and Corporal Foot lived. They stole the Magic Mirror. [...]

VerbExtraction

Mention Extraction

Feature-Vector Assembly

RoleIdentification

Coreference Resolution

Character Identification

External Knowledge Examples Examples

FunctionIdentification

Domain Knowledge

Natural LanguagePreprocessing

Automated Narrative Information Extraction

• Voz

19

Story Graphs

20

Evaluation of IE PipelinesResults

21

Coreference Voting for Roles