Return to search

Future talk : the race to build a bot that gabs like a human / Race to build a bot that gabs like a human

Thesis: S.M. in Science Writing, Massachusetts Institute of Technology, Department of Comparative Media Studies/Writing, 2019 / Cataloged from PDF version of thesis. Page 21 blank. / Includes bibliographical references. / Gunrock is a chatbot designed in the likeness of an average 29-year-old woman living in Seattle. Fourteen students from the University of California, Davis, spent spring and summer of 2018 designing and testing the bot. At the end of summer, Gunrock placed first in the 2018 Amazon Alexa Prize, a competition that challenges students to build the best "socialbot," a computer program that talks out loud and engages in "fun, high-quality conversations on popular societal topics." Although Gunrock is rudimentary compared to the conversational ability of a real person, she is also cutting-edge and a predecessor of more advanced systems. Gunrock pulls information from many sources, including Reddit and Twitter comments. As chatbots like Gunrock become more prevalent, their designers must make important decision to determine what chatbots say, which in turn has influence on the user. / by Madeleine Turner. / S.M. in Science Writing / S.M.inScienceWriting Massachusetts Institute of Technology, Department of Comparative Media Studies/Writing

Identiferoai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/123787
Date January 2019
CreatorsTurner, Madeleine(Madeleine Renee)
ContributorsThomas Levinson., Massachusetts Institute of Technology. Department of Comparative Media Studies., Massachusetts Institute of Technology. Graduate Program in Science Writing., Massachusetts Institute of Technology. Department of Comparative Media Studies, Massachusetts Institute of Technology. Graduate Program in Science Writing, Massachusetts Institute of Technology. Graduate Program in Science Writing
PublisherMassachusetts Institute of Technology
Source SetsM.I.T. Theses and Dissertation
LanguageEnglish
Detected LanguageEnglish
TypeThesis
Format21 pages ;, application/pdf
RightsMIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission., http://dspace.mit.edu/handle/1721.1/7582

Page generated in 0.0015 seconds