evevant
relevant to from
their own
infomation
these
websites.The inquiries
series of keywords,rather than naturallanguage fromat.In
a
f.act,users may be more accustomed to using naturallanguage to describe most
cases,users
problem.In
need the exact answer to the probIem,rather than is
a
a
series of ralated 6eld of natural
websites.QA
technology
Very
a
hot
research direction
in
the
language processing.It combines try
to
large Variety of NLP technologies.In this paper,we for Chinese
an
inVestigate
some
technoIogies
QA
systems.In
the
field
of
information retrieVal,
semantic
simiIarity plays
important role,to improve the
accuracy of semantic sim“arity has important theoretical and practical signi6cance. It’s more difficult f|or conlputers to process Chinese than to processing of word
Western
the
letters in the and
can
segmentation.Word
segmentation
is
foundation
precondition of Chinese sentense similarity conlputing,the accuracy of lhe result be greatly improVed when adopting more improVed Chinese word
e艏cient
arithmetic.In this paper,a kind of
on
segmentation method based
on
double.arraV trie and the and contrast of
strategy to eliminate the ambiguity is put f-orward common Chinese word segmentation
the analysis
an
arithmetic.We
propose
imporoved method
about dictionary mechanisms,segmentation steps and ambiguity,the integrality and accuracy of word segmentation will be enhanced. In Chinese information processing,sentence sim订arity computing is widely used in the
area
of information retrieval,machine translation,automatic question?answerjng,
a
text mining and etc.It’s
as a
question of much essential and imponant that people study
a
hotspot and difficulty f.0r
long time.In this paper,on the basis of analysis and
contrast of existing sentense similarity computing methods,a new sentense similarity
computing method is put forward which make and semantic method based
on
use
of TF.IDF method based with the word
on
VSM
Hownet
combind
segmentation
arithmetic which mentioned above.The rea“zation of Chinese text word segmentation and similarity computing with computer system is put fbrward and carry through
Ⅲ
硕士学位论文
many
testing.Question-answering
retrieVal syatem is tested
as
example to Validate the
use
method that used.Expefimental results show that aRer making word segmentation aIgorithm,the time and
of the presented
space utilization efnciency has been
greatly improVed and proposed methods of sentence similarity effect is good.
Key
Words:
Natural language
processing;
Question
answering;
Hownet;Word
segmentation;Semantic similarity
IV
口碑营销
http://info.3dou.com/
基于HouNet多特征结合的句子相似度计算
插图索引
图3.1以“中"字开头的部分Trie树结构…………………………………………………?19
图3.2分词流程………………………………………………………………………22
图4.1树状的义原层次结构…………………………………………………………“28 图5.1用于查找候选问题集的结构………………………………………………