For Project Enquiry +91 9962 067 067

EXISTING SYSTEM

PROPOSED SYSTEM

EXISTING CONCEPT:-

Wikipedia has become a popular data source in hyponym relation extraction research. Several such studies adopted the syntactic-pattern-based methods or textural feature- based machine learning methods to extract hyponym relations from Wikipedia.

These methods rely mainly on features extracted from the text content of Wikipedia. When shifting to a new domain, these methods require new syntactic patterns to be learned or new training samples to be manually constructed, which usually entail high labor costs.

PROPOSED CONCEPT:-

In this approach to develop Wikipedia with relation extraction approach without navigating different domains. It has two approaches to check our mechanism first, manually labeled data and automatic labeled data sets of different domains.

We found that the Wikipedia navigation box, a type of structural information in Wikipedia article pages, contains plenty of domain- specific hyponym relations that can be automatically extracted. This type of structural information is suitable for automatic labeling of training data with weakly supervised learning because of its high quality.

EXISTING TECHNIQUE:-

Information Extraction.

PROPOSED TECHNIQUE:-

Motif based hyponym relation Extraction.

TECHNIQUE DEFINITION:-

Considerable knowledge is necessary to accurately extract these tuples from a broad range of text. Existing techniques obtain it in ways ranging from direct knowledge-based encoding to supervised learning to self-supervised learning.

TECHNIQUE DEFINITION:-

Automatically construct motif-based features from the WAG of a domain. Extract structural information from Wikipedia and heuristically label the training set of the domain based on the extracted structural information.

DRAWBACKS:-

Labor cost is high for generating manual labeling.

Domain independency.

ADVANTAGES:-

Labor cost is low for generating automatic labeling.

Domain dependency.