Modeling Humor Within Text: Data Mining and Visualization Strategies for Automated Joke Detection

Presenter Information

Andrew Smigaj

Document Type

Oral Presentation

Campus where you would like to present

SURC 137B

Start Date

21-5-2015

End Date

21-5-2015

Keywords

Computational Humor, Natural Language Processing, Data Visualization

Abstract

The goal of this project was to investigate the use of data mining and visualization as an approach to modeling humor within text. In particular, we developed algorithmic and automated approaches to visualizing and detecting belief shifts as they occur as intelligent agents parse meaning from garden path jokes. Garden path jokes can occur when a reader's initial interpretation of an ambiguous text turns out to be incorrect, leading them down the wrong path to a semantic dead end. Given new information, semantic incongruities arise that require resolution, often triggering a humorous response. For both humans and computers, parsing of meaning requires an ontology describing what type of things exist in the world and how they are connected, as well as methods for establishing belief given uncertainty and ambiguity. One major aim of this project has been to explore automated methods for identifying what things exist in this world and how they are related, using the world wide web as a massive corpus of natural language data for knowledge discovery. The methodology and tools resulting from this project offer a new approach to testing and generating hypothesis related to theories of humor, as well as many other incongruity-based linguistic phenomena.

Faculty Mentor(s)

Boris Kovalerchuk

Department/Program

Computer Science

Additional Mentoring Department

Computer Science

This document is currently not available here.

Share

COinS
 
May 21st, 10:40 AM May 21st, 11:00 AM

Modeling Humor Within Text: Data Mining and Visualization Strategies for Automated Joke Detection

SURC 137B

The goal of this project was to investigate the use of data mining and visualization as an approach to modeling humor within text. In particular, we developed algorithmic and automated approaches to visualizing and detecting belief shifts as they occur as intelligent agents parse meaning from garden path jokes. Garden path jokes can occur when a reader's initial interpretation of an ambiguous text turns out to be incorrect, leading them down the wrong path to a semantic dead end. Given new information, semantic incongruities arise that require resolution, often triggering a humorous response. For both humans and computers, parsing of meaning requires an ontology describing what type of things exist in the world and how they are connected, as well as methods for establishing belief given uncertainty and ambiguity. One major aim of this project has been to explore automated methods for identifying what things exist in this world and how they are related, using the world wide web as a massive corpus of natural language data for knowledge discovery. The methodology and tools resulting from this project offer a new approach to testing and generating hypothesis related to theories of humor, as well as many other incongruity-based linguistic phenomena.