Publication Date

8-2020

Date of Final Oral Examination (Defense)

7-6-2020

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Computer Science

Department

Computer Science

Major Advisor

Jerry Fails, Ph.D.

Advisor

Casey Kennington, Ph.D.

Advisor

Maria Soledad Pera, Ph.D.

Advisor

Katherine Wright, Ph.D.

Abstract

Children commonly use software applications such as search engines and word processors in the classroom environment. However, a major barrier to using these programs successfully is the ability of children to type and spell effectively. While many programs make use of spellcheckers to provide spelling corrections to their users, they are designed for more traditional users (i.e., adults) and have proven inadequate for children. The aims of this work is twofold: first, to address the types of spelling errors children make by researching, developing, and evaluating algorithms to generate and rank candidate spelling suggestions; and second, to evaluate the impact interactive elements have on children's spellchecking behaviors seeking to improve their user experience.

Motivated by children's phonological strategies to spell, a phonetic encoding strategy is used to map words and misspellings to phonetic keys to effectively and efficiently provide spelling correction candidates. Machine learning methods, including Learning to Rank, are used to rank candidates effectively and reveal the importance of phonetic features. Experimental results show this method is able to more accurately provide and rank spelling corrections when handling misspellings generated by children in both essay writing and web search settings when compared to state-of-the-art baselines. The design of an interactive spellchecker reveals children's propensity towards visual and audio cues. A study on visual and audio cues show an influence on children's selection habits and a positive impact on assisting children in selecting correct spelling suggestions.

DOI

10.18122/td/1721/boisestate

Share

COinS