This is a Context based stemming or semantic stemming. It even goes beyond stemming hence it is named as Personalized derivative. User can even customize the output based on its domain by integrating ontologies .
2. What is PTD?
Semantic approach for Stemming
•Capabilities to include Ontology and Semantic dictionary like wordnet
Extends the meaning of stemming to hypernym and user defined root word
Includes linguistic and morphological stemming provided by Porter-Stemmer and K-
Stem
4. Architecture Flow Diagram
Extract
Sentences
Lemmatized each word in the
Sentences
Named Entity Extraction ,
Tokenization
Annotation
For Each
Sentences
For Each Tokens
If
Term
= N-E
Word Sense Disambiguation
for each word in a sentence
Right sense of each word ,
POS Identification
Tagging
If
POS
=
NN
For Each
Tokens
YES
YES
NO
Find Hypernym/Root word
NO (Till all terms
are processed
once)
Knowledge Data
Sources
Thesaurus
Dictionary,
Ontology
Create
Exclusion
List
K-Stemmer
If Root word
not found
Documents