ABSTRACT:
We
identify relation completion (RC) as one recurring problem that is central to
the success of novel big data applications
such
as Entity Reconstruction and Data Enrichment. Given a semantic relation R, RC
attempts at linking entity pairs between two entity lists under the relation R.
To accomplish the RC goals, we propose to formulate search queries for each
query entity a based on some auxiliary information, so that to detect its
target entity b from the set of retrieved documents. For instance, a
pattern-based method (PaRE) uses extracted patterns as the auxiliary
information in formulating search queries. However, high-quality patterns may decrease
the probability of finding suitable target entities. As an alternative, we
propose CoRE method that uses context terms learned surrounding the expression
of a relation as the auxiliary information in formulating queries. The
experimental results based on several
real-world
web data collections demonstrate that CoRE reaches a much higher accuracy than
PaRE for the purpose of RC.
EXISTING SYSTEM:
This
data is typically unstructured and naturally lacks any binding information
(i.e., foreign keys). Linking this data clearly goes beyond the capabilities of
current data integration systems. This motivated novel frameworks that
incorporate information extraction (IE) tasks such as named entity recognition (NER)
and relation extraction (RE). Those frameworks have been used to enable some of
the emerging data linking applications such as entity reconstruction and data
enrichment.
DISADVANTAGES OF
EXISTING SYSTEM:
v
The number of retrieved documents is
expected to be prohibitively large and in turn, processing them incurs a large
overhead.
v
Those documents would include
significant amount of noise, which might eventually lead to a wrong.
PROPOSED SYSTEM:
Our goal is to formulate effective
and efficient search queries based on RE methods. In general, given some
semantic relation R (e.g., (Lecturer, University)), general RE tasks target at
obtaining relation instances of the relation R from free text.
we propose a novel Context-Aware Relation Extraction method (CoRE), which is particularly designed for the RC task.
we propose a novel Context-Aware Relation Extraction method (CoRE), which is particularly designed for the RC task.
CoRE
recognizes and exploits the particular context of an RC task. Towards this,
instead of representing a relation in the form of strict high-quality patterns,
CoRE uses context terms, which we call relation-context terms (RelTerms). For example
in Fig. 1b, CoRE searches the web for documents that contain each of the seed
instance pairs and from those documents it learns some RelTerms such as
“department” and “faculty”. Based on those RelTerms, CoRE can formulate a query
such as “Bob Brown + (department OR faculty)” for the incomplete instance (Bob
Brown, ?). From the returned documents, we can then obtain “UIUC” as the target
entity.
ADVANTAGES OF PROPOSED
SYSTEM:
v
It allows more flexibility in
formulating the search queries based on context terms instead of patterns.
v
It seamlessly allows including any query
entity as one of the context terms, which further improves the chances of
finding a matching target rather than lowering it.
SYSTEM ARCHITECTURE:
 |
SYSTEM CONFIGURATION:-
HARDWARE REQUIREMENTS:-
ü Processor - Pentium
–IV
ü Speed - 1.1 Ghz
ü RAM - 512 MB(min)
ü Hard
Disk - 40 GB
ü Key
Board - Standard Windows Keyboard
ü Mouse - Two or Three Button Mouse
ü Monitor - LCD/LED
SOFTWARE
REQUIREMENTS:
•
Operating system : Windows XP.
•
Coding Language : .Net
•
Data Base : SQL
Server 2005
•
Tool : VISUAL STUDIO 2008.
REFERENCE:
Zhixu
Li, Mohamed A. Sharaf, Laurianne Sitbon, Xiaoyong Du, and Xiaofang Zhou, “CoRE: A CONTEXT-AWARE RELATION EXTRACTION
METHOD FOR RELATION COMPLETION” IEEE TRANSACTIONS ON KNOWLEDGE AND DATA
ENGINEERING, VOL. 26, NO. 4, APRIL 2014
No comments:
Post a Comment