Entry Tepper:2010:IMU from talip.bib
Last update: Sun Oct 15 02:55:04 MDT 2017
Top |
Symbols |
Numbers |
Math |
A |
B |
C |
D |
E |
F |
G |
H |
I |
J |
K |
L |
M |
N |
O |
P |
Q |
R |
S |
T |
U |
V |
W |
X |
Y |
Z
BibTeX entry
@Article{Tepper:2010:IMU,
author = "Michael Tepper and Fei Xia",
title = "Inducing Morphemes Using Light Knowledge",
journal = j-TALIP,
volume = "9",
number = "1",
pages = "3:1--3:??",
month = mar,
year = "2010",
CODEN = "????",
DOI = "https://doi.org/10.1145/1731035.1731038",
ISSN = "1530-0226 (print), 1558-3430 (electronic)",
ISSN-L = "1530-0226",
bibdate = "Mon Mar 29 15:34:01 MDT 2010",
bibsource = "http://portal.acm.org/;
http://www.math.utah.edu/pub/tex/bib/talip.bib",
abstract = "Allomorphic variation, or form variation among morphs
with the same meaning, is a stumbling block to
morphological induction (MI). To address this problem,
we present a hybrid approach that uses a small amount
of linguistic knowledge in the form of orthographic
rewrite rules to help refine an existing MI-produced
segmentation. Using rules, we derive underlying
analyses of morphs---generalized with respect to
contextual spelling differences---from an existing
surface morph segmentation, and from these we learn a
morpheme-level segmentation. To learn morphemes, we
have extended the Morfessor segmentation algorithm
[Creutz and Lagus 2004; 2005; 2006] by using rules to
infer possible underlying analyses from surface
segmentations. A segmentation produced by Morfessor
Categories-MAP Software v. 0.9.2 is used as input to
our procedure and as a baseline that we evaluate
against. To suggest analyses for our procedure, a set
of language-specific orthographic rules is needed. Our
procedure has yielded promising improvements for
English and Turkish over the baseline approach when
tested on the Morpho Challenge 2005 and 2007 style
evaluations. On the Morpho Challenge 2007 test
evaluation, we report gains over the current best
unsupervised contestant for Turkish, where our
technique shows a 2.5\% absolute {\em F\/} -score
improvement.",
acknowledgement = ack-nhfb,
articleno = "3",
fjournal = "ACM Transactions on Asian Language Information
Processing",
journal-URL = "http://portal.acm.org/browse_dl.cfm?&idx=J820",
keywords = "allomorphy; computational linguistics; machine
learning; Morphological induction",
}
Related entries
- absolute,
8(3)10
- address,
6(3)11,
7(1)2,
7(1)3,
8(2)6,
8(4)15,
8(4)18,
8(4)19,
10(3)13,
10(4)21,
11(2)6,
11(4)13,
11(4)17,
13(1)1,
13(2)8,
13(4)17
- against,
6(1)z-1
- algorithm,
1(4)281,
5(2)165,
6(3)11,
6(4)2,
6(4)3,
7(1)3,
7(2)7,
7(3)8,
7(3)10,
7(4)12,
8(1)4,
8(3)12,
9(1)4,
10(2)7,
10(2)9,
10(2)10,
10(4)19,
11(1)3,
11(2)4,
11(2)7,
11(3)9,
11(3)11,
11(4)14,
12(1)1,
12(1)2,
12(2)6,
13(1)4,
13(3)13,
13(4)18
- amount,
7(3)8,
8(4)17,
9(4)15,
10(2)7,
10(2)9,
10(4)21,
13(1)3,
13(3)14
- analyses,
10(1)2,
10(2)10,
11(4)18
- baseline,
7(1)1,
7(2)5,
7(2)6,
7(3)8,
8(1)4,
8(3)10,
9(1)4,
9(2)6,
9(2)7,
9(3)12,
11(4)13,
11(4)14
- best,
5(2)89,
5(3)183,
7(2)7,
7(4)13,
8(2)6,
9(2)7,
9(3)11,
9(3)12,
11(3)8,
11(4)13,
12(1)2,
12(2)5,
12(2)7,
12(3)9,
12(4)14,
13(1)4,
13(4)16
- challenge,
4(3)243,
6(3)10,
6(4)3,
8(1)4,
8(4)14,
8(4)18,
10(3)12,
13(1)1,
13(1)4,
13(2)8,
13(3)12,
13(4)17
- computational,
7(3)9,
10(1)1,
12(2)5,
13(2)10
- contextual,
1(4)297
- Creutz,
13(3)14
- current,
8(4)14,
9(2)5,
10(1)6,
13(1)2
- difference,
7(3)10,
8(4)18,
9(2)5,
9(3)11,
10(1)2,
10(4)17,
12(1)1,
12(3)11,
13(2)10,
13(4)16
- English,
2(3)245,
4(2)135,
5(2)89,
5(2)121,
5(3)245,
6(2)6,
6(2)7,
6(3)11,
6(4)2,
7(1)1,
7(4)11,
8(2)9,
8(4)15,
8(4)16,
8(4)17,
9(1)1,
9(2)7,
9(3)12,
9(4)14,
9(4)15,
10(1)2,
10(1)4,
10(2)8,
10(3)14,
10(3)15,
10(4)17,
11(2)4,
11(2)5,
11(2)6,
11(3)8,
11(3)11,
12(2)5,
12(3)12,
12(4)14,
12(4)17,
13(1)1,
13(4)16
- evaluate,
5(2)121,
7(2)5,
7(2)7,
8(2)8,
9(3)11,
11(4)18,
12(1)2,
12(3)10,
13(2)7
- evaluation,
3(4)243,
5(2)165,
6(1)z-2,
7(2)7,
7(4)13,
8(2)6,
8(3)12,
8(4)15,
8(4)16,
9(2)6,
9(3)10,
9(3)11,
9(3)12,
9(4)14,
10(3)15,
11(2)5,
11(4)13,
11(4)18,
12(2)6,
12(4)15,
13(1)1
- existing,
6(2)8,
6(4)2,
8(1)2,
8(2)7,
9(2)5,
9(4)13,
10(3)16,
11(2)4,
11(2)7,
11(4)18,
12(2)5,
12(4)17,
13(1)2,
13(2)8
- extended,
5(3)183,
7(3)10,
7(4)13,
8(4)15,
8(4)18,
11(2)6,
13(4)16
- form,
6(3)9,
7(1)3,
8(1)3,
8(4)18,
9(3)12,
9(4)13,
10(2)8,
11(2)7,
11(4)13,
12(1)4,
12(3)11
- generalized,
2(3)193,
11(2)4,
13(4)16
- has,
5(2)89,
5(2)121,
6(4)2,
6(4)3,
7(3)10,
7(4)11,
7(4)12,
7(4)13,
8(1)2,
8(3)10,
8(4)16,
8(4)17,
8(4)18,
9(1)1,
9(1)2,
9(2)5,
9(4)15,
10(1)4,
10(1)5,
10(4)17,
11(1)2,
11(3)9,
12(1)1,
12(1)2,
12(4)15,
13(1)1,
13(2)6,
13(2)7,
13(2)8,
13(3)12,
13(4)18
- have,
5(2)121,
6(3)10,
6(4)2,
7(2)7,
7(3)8,
7(4)11,
7(4)12,
7(4)13,
8(2)8,
8(2)9,
8(4)14,
8(4)16,
8(4)17,
8(4)18,
9(1)1,
9(2)7,
9(4)15,
10(1)5,
10(2)7,
10(4)19,
10(4)20,
10(4)21,
11(1)2,
11(2)6,
11(3)8,
12(1)3,
12(2)6,
13(1)3,
13(2)6,
13(2)7,
13(2)9,
13(3)11,
13(3)12
- help,
5(2)146,
6(2)6,
7(3)8,
8(4)17,
8(4)18,
9(2)7,
10(2)10,
11(3)9,
12(1)3,
12(4)15
- hybrid,
3(2)113,
7(2)5,
7(4)13,
11(3)11,
13(3)14
- improvement,
4(3)280,
5(2)121,
6(2)7,
7(1)2,
7(1)3,
7(3)8,
7(3)10,
8(1)4,
8(3)10,
8(4)15,
8(4)16,
9(3)11,
10(2)8,
11(4)17,
11(4)18,
12(1)1,
12(2)7,
12(3)11,
13(3)12,
13(3)14,
13(4)16,
13(4)17
- induction,
2(3)290,
9(1)4,
11(3)9
- infer,
11(1)3,
11(3)9
- input,
5(2)165,
8(2)7,
9(1)1,
9(2)6,
10(1)6,
10(4)20,
12(1)2,
12(1)4,
12(2)6,
13(1)4,
13(2)8,
13(3)12,
13(4)17
- knowledge,
4(4)435,
5(1)4,
5(1)74,
5(2)121,
5(2)146,
7(3)9,
7(4)12,
8(1)4,
11(1)3,
11(2)4,
11(4)13,
11(4)15,
11(4)16,
11(4)17,
11(4)18,
13(4)16
- Lagus,
13(3)14
- language-specific,
5(2)121,
9(3)12
- learn,
7(1)1,
12(1)3,
13(1)2,
13(4)16
- learning,
3(2)159,
5(1)61,
5(2)121,
5(4)413,
6(2)6,
6(4)1,
6(4)3,
7(2)7,
7(3)9,
8(3)10,
8(4)15,
9(2)5,
10(1)5,
10(2)10,
10(3)16,
10(4)20,
10(4)21,
11(1)3,
11(4)14,
11(4)16,
12(1)1,
12(1)3,
12(2)5,
12(2)7,
12(4)15,
13(1)2,
13(1)3,
13(2)9,
13(4)16,
13(4)17
- light,
9(3)11
- linguistic,
2(2)101,
6(3)10,
7(3)9,
7(4)11,
7(4)13,
8(4)15,
8(4)17,
9(1)2,
9(2)5,
10(1)1,
10(2)8,
11(3)10,
11(4)16,
11(4)18,
12(1)2,
12(3)11,
13(3)11
- machine,
4(1)18,
4(4)377,
5(2)89,
5(3)185,
6(2)6,
6(4)2,
6(4)3,
7(1)1,
7(2)5,
7(2)7,
7(3)9,
7(3)10,
8(1)4,
8(2)5,
8(2)6,
8(2)7,
8(2)8,
8(2)9,
8(3)10,
8(4)15,
9(4)13,
10(1)2,
10(1)5,
10(3)16,
10(4)18,
10(4)20,
11(1)2,
11(3)8,
11(4)14,
11(4)16,
12(3)9,
12(3)12,
12(4)14,
12(4)16,
12(4)17,
13(1)2,
13(3)11,
13(4)16,
13(4)17
- meaning,
3(1)33,
7(4)11,
9(2)5,
9(2)7,
11(4)14,
12(3)11,
12(4)16
- Morfessor,
13(3)14
- morphemes,
6(3)9
- morphological,
6(4)2,
7(3)9,
8(4)16,
9(4)15,
10(1)4,
11(3)9,
13(2)9,
13(3)14
- needed,
13(2)10
- orthographic,
10(1)3
- possible,
7(2)6,
8(4)16,
8(4)18,
9(1)1,
10(1)4,
10(3)15,
10(4)19,
12(1)2
- present,
5(2)89,
5(2)165,
6(2)7,
6(3)10,
6(4)2,
7(1)2,
7(1)3,
7(2)7,
7(3)9,
7(4)11,
7(4)13,
8(1)3,
8(2)6,
8(2)7,
8(2)8,
8(3)10,
8(4)14,
8(4)16,
8(4)17,
8(4)18,
8(4)19,
9(1)1,
9(1)2,
9(2)6,
9(4)14,
10(1)4,
10(1)6,
10(2)7,
10(3)14,
10(4)18,
10(4)19,
11(1)2,
11(1)3,
11(2)4,
11(2)5,
11(2)6,
11(3)10,
11(4)13,
12(1)3,
12(2)5,
12(3)9,
12(3)11,
12(4)15,
13(2)8,
13(4)16
- problem,
6(2)7,
6(3)9,
6(3)11,
6(4)1,
7(1)2,
7(2)7,
7(3)10,
8(1)2,
8(2)9,
8(3)10,
8(4)19,
9(1)1,
9(2)5,
9(4)13,
10(1)2,
10(1)4,
10(3)14,
10(3)16,
10(4)21,
11(3)8,
11(3)11,
11(4)17,
11(4)18,
12(1)2,
12(1)3,
12(2)7,
12(3)10,
12(3)12,
12(4)16,
13(2)8,
13(3)14,
13(4)17
- procedure,
6(3)10,
9(3)11
- produced,
8(1)4
- promising,
5(2)165,
8(1)2,
12(1)3,
12(2)6
- refine,
6(2)8,
10(1)2
- report,
8(2)7,
8(4)16,
10(2)9,
10(2)10,
13(2)7,
13(3)11
- respect,
7(3)9,
8(4)15,
11(2)6
- rule,
5(2)165,
7(1)1,
7(4)12,
9(4)14,
10(2)8,
11(1)3,
11(3)8,
11(4)15,
12(3)11,
13(4)16
- same,
7(4)11,
9(2)6,
10(2)10,
11(2)6,
12(4)16,
13(2)6
- score,
7(2)7,
7(4)12,
8(2)7,
8(3)10,
9(2)6,
10(4)18,
11(2)6,
12(3)9,
12(4)17,
13(1)2,
13(3)13,
13(4)16
- segmentation,
8(2)7,
8(4)16,
9(1)2,
9(2)5,
9(4)15,
11(2)7,
12(1)2,
12(1)4,
12(4)16,
13(2)9
- set,
1(3)269,
5(2)121,
6(1)z,
7(1)3,
7(3)8,
7(4)11,
7(4)13,
8(3)10,
8(3)12,
8(4)15,
9(1)1,
9(2)5,
10(1)4,
10(2)8,
10(4)20,
11(2)5,
11(2)7,
11(3)10,
11(3)11,
11(4)13,
11(4)14,
12(1)2,
12(1)4,
12(3)9,
13(2)8,
13(2)9,
13(3)12,
13(3)13,
13(4)17
- show,
5(2)89,
5(2)146,
7(1)1,
7(1)2,
7(1)3,
7(4)11,
7(4)12,
7(4)13,
8(1)4,
8(2)7,
8(2)9,
8(3)12,
8(4)16,
8(4)17,
9(1)1,
9(1)2,
9(2)5,
9(2)6,
9(2)7,
9(3)11,
9(3)12,
9(4)14,
10(1)3,
10(3)15,
11(2)4,
11(2)5,
11(2)7,
11(3)8,
11(3)11,
11(4)14,
11(4)15,
11(4)17,
11(4)18,
12(1)2,
12(1)4,
12(2)5,
12(2)7,
12(3)9,
12(3)10,
12(3)11,
12(4)15,
12(4)16,
13(1)3,
13(2)6,
13(2)7,
13(2)9,
13(3)14
- small,
6(3)11,
7(3)9,
7(4)11,
8(2)9,
8(4)17,
10(4)21,
12(2)7
- specific, language-,
5(2)121,
9(3)12
- spelling,
10(1)3,
13(2)8
- suggest,
9(3)11,
11(4)14,
13(2)10,
13(3)11
- surface,
7(4)12,
8(2)6,
9(2)6,
11(4)15,
12(3)11
- technique,
2(1)63,
3(4)243,
5(2)89,
6(1)z,
6(2)7,
6(2)8,
6(3)9,
6(4)2,
6(4)3,
7(2)5,
7(2)7,
7(3)8,
7(4)12,
8(4)18,
9(2)5,
10(1)4,
10(2)9,
10(2)10,
10(4)21,
11(1)1,
12(2)6,
12(3)9,
12(3)11,
12(4)14,
13(2)8,
13(3)12,
13(3)13,
13(4)18
- test,
6(2)6,
7(2)5,
9(3)10,
9(3)11,
11(4)13,
12(3)9,
13(1)4,
13(2)7,
13(3)13,
13(4)17
- tested,
6(3)11,
11(2)7,
12(1)4,
13(2)6,
13(4)16
- underlying,
8(1)2,
8(3)12,
10(2)9,
10(3)12
- unsupervised,
6(2)6,
7(3)9,
8(1)3,
9(1)4,
9(2)7,
10(2)7,
11(3)9,
13(3)12
- use,
4(2)159,
5(2)89,
5(2)146,
6(2)8,
6(3)11,
7(2)6,
7(3)9,
7(4)11,
7(4)12,
8(1)3,
8(2)9,
8(3)10,
8(3)11,
9(1)1,
9(3)11,
10(1)3,
10(1)4,
11(1)1,
11(2)6,
11(2)7,
11(3)8,
11(3)10,
11(4)14,
11(4)18,
12(1)1,
12(2)6,
12(3)9,
12(3)10,
13(2)6,
13(2)9,
13(2)10,
13(3)12
- used,
5(2)89,
5(2)146,
7(1)3,
7(2)6,
7(2)7,
7(3)9,
7(4)12,
7(4)13,
8(3)10,
8(4)17,
9(1)1,
9(2)6,
9(3)10,
10(1)2,
10(1)6,
10(2)7,
10(2)8,
10(3)12,
10(3)13,
10(4)20,
11(1)2,
11(1)3,
11(3)10,
11(4)13,
11(4)14,
12(2)5,
12(3)9,
12(3)11,
12(3)12,
13(2)6,
13(3)11
- variation,
8(4)19,
12(2)6,
12(4)14,
13(3)12,
13(4)16
- when,
5(2)89,
6(3)9,
6(3)11,
7(2)6,
7(4)11,
7(4)13,
8(1)3,
8(4)16,
8(4)17,
9(2)5,
9(3)11,
9(3)12,
9(4)13,
11(2)4,
12(1)2,
12(4)14,
13(2)8,
13(3)12
- where,
7(3)8,
7(3)9,
7(3)10,
8(2)7,
8(2)8,
9(2)7,
9(4)13,
10(4)18,
11(2)4,
11(4)16,
12(1)3,
12(4)14
- yielded,
13(4)16