2013 | OriginalPaper | Chapter
Critiquing Text Analysis in Social Modeling: Best Practices, Limitations, and New Frontiers
Author : Peter A. Chew
Published in: Social Computing, Behavioral-Cultural Modeling and Prediction
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Natural language processing (NLP) is an important contributor to the field of social modeling. Language is a social artifact; it is how people express opinions, persuade, or convey what they believe is important. It is thus rightly recognized that computational tools can automate at least some of the analytical work of reading an ever-increasing volume of textual data, reducing time and costs. Language is also, however, a complex and variegated system, creating a challenge for social modelers. In this paper, we contend NLP’s full potential is commonly not being exploited, leading to unnecessary work and lower-quality results, and that social modelers using NLP should understand at a high level what NLP problems are, and are not, solved. Our findings have implications for both the practice and validation of NLP in the social modeling community.