An Unsupervised Method for Uncovering Morphological Chains

NLPIR SEMINAR Y2018#9
自然语言处理与信息检索共享平台e/xu%K ^lG

INTRO自然语言处理与信息检索共享平台dN VV#p8M!i

        In the new semester, our Lab, Web Search Mining and Security Lab, plans to hold an academic seminar every Wednesdays, and each time a keynote speaker will share understanding of papers published in recent years with you.

;Khn
`FD1Nx0

 自然语言处理与信息检索共享平台$UG9n(u;|

Arrangement自然语言处理与信息检索共享平台?L/oq3\&Lq,B

        This week’s seminar is organized as follows:自然语言处理与信息检索共享平台+C0R]:?’xs9p

        1. The seminar time is 1.pm, Wed, at Zhongguancun Technology Park ,Building 5, 1306.自然语言处理与信息检索共享平台 oW:z[*A%{!o9P_u

        2. The lecturer is Yaofei Yang, the paper’s titles are An Unsupervised Method for Uncovering Morphological Chains and From Segmentation to Analyses A Probabilistic Model for Unsupervised Morphology Induction.自然语言处理与信息检索共享平台#TG&X(R|LN
f

        3. The seminar will be hosted Wang Gang.自然语言处理与信息检索共享平台?Irve/SPf

        4. Attachment is the paper of this seminar, please download in advance

0~
X!t dG/Y F0

 

[M&~,j\Sm.V0

        Everyone interested in this topic is welcomed to join us. the following is the abstract for this week’s paper自然语言处理与信息检索共享平台m*_,sN_u#{

 自然语言处理与信息检索共享平台(\p+o xB#U]

An Unsupervised Method for Uncovering Morphological Chains自然语言处理与信息检索共享平台&Go2wq:n

Karthik Narasimhan               Regina Barzilay            Tommi Jaakkola

c?Q/ee#e}0

Abstract

#On9?*]*A0

Most state-of-the-art systems today produce morphological analysis based only on orthographic patterns. In contrast, we propose a model for unsupervised morphological analysis that integrates orthographic and semantic views of words. We model word formation in terms of morphological chains, from base words to the observed words, breaking the chains into parent-child relations. We use log-linear models with morpheme and wordlevel features to predict possible parents, including their modifications, for each word. The limited set of candidate parents for each word render contrastive estimation feasible. Our model consistently matches or outperforms five state-of-the-art systems on Arabic, English and Turkish.

M;j
H0b;Y0

 自然语言处理与信息检索共享平台Jj_+DiS4FBTA

From Segmentation to Analyses:

`x]$?*di*h U0

A Probabilistic Model for Unsupervised Morphology Induction

A0Q
j?{\{4zp0

Toms Bergmanis           Sharon Goldwater

2Szv{NX/M0

Abstract

? f1\l
y-d8b t0

        A major motivation for unsupervised morphological analysis is to reduce the sparse data problem in under-resourced languages. Most previous work focuses on segmenting surface forms into their constituent morphs (e.g., taking: tak +ing), but surface form segmentation does not solve the sparse data problem as the analyses of take and taking are not connected to each other. We extend the MorphoChains system (Narasimhan et al., 2015) to provide morphological analyses that can abstract over spelling differences in functionally similar morphs. These analyses are not required to use all the orthographic material of a word (stopping: stop +ing), nor are they limited to only that material (acidified: acid +ify +ed). On average across six typologically varied languages our system has a similar or better F-score on EMMA (a measure of underlying morpheme accuracy) than three strong baselines; moreover, the total number of distinct morphemes identified by our system is on average 12.8% lower than for Morfessor (Virpioja et al., 2013), a state-of-the-art surface segmentation system.

(LY:v8e$F%jm
whi
]0

 

#[ o3O!z$RJ SA0

 

u’u:{9Xq p0u0

 

wTR$J/~5C0

 自然语言处理与信息检索共享平台”c’Ia%SLQ:P&^z

You May Also Like

About the Author: nlpir

发表回复