A Minimal Rule Generalization Algorithm for Multi-Part Phonological rule Learning

Date
2017
Journal Title
Journal ISSN
Volume Title
Publisher
Producer
Director
Performer
Choreographer
Costume Designer
Music
Videographer
Lighting Designer
Set Designer
Crew Member
Funder
Rehearsal Director
Concert Coordinator
Moderator
Panelist
Alternative Title
Department
Swarthmore College. Dept. of Linguistics
Type
Thesis (B.A.)
Original Format
Running Time
File Format
Place of Publication
Date Span
Copyright Date
Award
Language
en_US
Note
Table of Contents
Terms of Use
Full copyright to this work is retained by the student author. It may only be used for non-commercial, research, and educational purposes. All other uses are restricted.
Rights Holder
Access Restrictions
Terms of Use
Tripod URL
Identifier
Abstract
In the late 1990's, Adam Albright and Bruce Hayes worked together on an algoritlnn that learned morphologic patterns from input data, intending to use this as a model for how humans learn morphological and phonemic changes across languages as well. Though the system was effective, it had some shortcomings - the algorithm was not designed to handle some more complicated relationships between morphology and phonology. One such relationship is known as a derived environment, created when the addition of some morpheme generates a phonological pattern deemed unacceptable in that given language. When analyzing a word pair that contained the result of one of these environments, the algorithm reads the resulting change as an original part of the affix, without regard for underlying forms. Here, I examine the structure of the original algorithm and layout a method for how future modifications can implement a system to account for theoretical derived environments. This requires splitting the area of change into a substructure that breaks down into two segments: a section that originally is introduced by the affix, and a section that, given a feature-matched phone in the root, likely was transformed after the morpheme was added. To give an example in Polish, for the root word [vag] (,weight'), the result after applying a verb-denoting suffix (here, the suffix [iiJ!) is [va3iij] ('to weigh'). Where the origioal algorithm would read [3iij] as the entire suffix, the modified version breaks it into the 'original' suffix [iij] and the segment [3], the latter of which maps onto [g] in the original word. From this, a rule describing this particular phonological change is generated in addition to a rule describing the affixation.
Description
Subjects
Citation
Collections