Towards Building Intelligent Revision Assistant for Argumentative Writings

Type PhD Dissertation Defense
Start Date April 7, 2017 03:00 PM
End Date April 7, 2017 05:00 PM
Location Eli Lily's room
Organizer Name
Speaker Name Fan Zhang
Speaker Affiliation Ph.D. Candidate, University of Pittsburgh
Abstract The current intelligent writing assistance tools (e.g. Grammarly, Turnitin, etc.) typically works by locating the problems of essays for users(grammar, spelling, argument, etc.) and providing solution suggestions. These tools only focus on providing feedback to a single draft, while ignoring the feedback to the author's changes between drafts (revision). This work argues that it is also important to provide feedback to the authors' revision, as such information can not only improve the quality of the writing but also improve the rewriting skill of the authors. Thus, it is necessary to build an intelligent assistant that focus on providing feedback to revisions.

This thesis presents works from three perspectives towards the building of such assistant: 1) the study on the revision' impact on writings, which includes the development of a sentence-level revision schema, the annotation of corpora based on the schema and the data analysis on the created corpora; 2) the development of algorithms for automatic revision identification, which includes the automatic extraction of the revised content and the automatic classification of revision types; 3) the user study on the effects of an intelligent revision assistant: which includes the development of a prototype intelligent revision assistant ArgRewrite and the analysis of user study results on the tool.
Committee Advisor: Diane Litman