Journal article
Error Analysis in an Automated Narrative Information Extraction Pipeline
IEEE transactions on computational intelligence and AI in games, v 9(4), pp 342-353
01 Dec 2017
Abstract
In this paper, we present our method for automatically extracting narrative information of characters and their narrative roles from natural language stories. In our corpus of 15 unannotated folk tales, our Voz system identifies 87% of the characters in the stories and correctly assigns 68% of the character roles. To better understand the sources of error in our system, we present an analytical methodology to study how the error is introduced by different modules and how it propagates through the pipeline. This methodology allows us to identify the bottleneck with the largest impact on the final error, which might be different from the module with the largest individual error in isolation. Our methodology can be applied to a wide variety of similar information extraction pipelines.
Metrics
Details
- Title
- Error Analysis in an Automated Narrative Information Extraction Pipeline
- Creators
- Josep Valls-Vargas - Drexel UniversityJichen Zhu - Drexel UniversitySantiago Ontanon - Drexel University
- Publication Details
- IEEE transactions on computational intelligence and AI in games, v 9(4), pp 342-353
- Publisher
- IEEE
- Number of pages
- 12
- Resource Type
- Journal article
- Language
- English
- Academic Unit
- Digital Media; Computer Science
- Web of Science ID
- WOS:000418422900003
- Scopus ID
- 2-s2.0-85044404650
- Other Identifier
- 991019168525004721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Software Engineering