An efficient information retrieval ontology system based indexing for context

IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308
_______________________________________________________________________________________
Volume: 04 Issue: 05 | May-2015, Available @ http://www.ijret.org 437
AN EFFICIENT INFORMATION RETRIEVAL ONTOLOGY SYSTEM
BASED INDEXING FOR CONTEXT
G.Krishna Raju1
, Padmanabham2
, A.Govardhan3
1
CS Department, Matrusri PG Studies, Saidabad, Hyderabad, India
2
Dean, Bharat Engineering College, Ibrahimpatnam, Telengana, India
3
Director, School Of IT, JNTUH, Hyderabad, Telangana, India
Abstract
Many of the research or development projects are constructed and vast type of artifacts are released such as article, patent, report of
research, conference papers, journal papers, experimental data and so on. The searching of the particular context through the
keywords from the repository is not an easy task because the earliest system the problem of huge recalls with low precision. This
paper challenges to construct a search algorithm based on the ontology to retrieve the relevant contexts. Ontology's are great
knowledge of retrieving the context. In this paper, we utilize the WordNet ontology to retrieve the relevant contexts from the document
repository. It is very difficult to retrieve the relevant context in its original format since we use the pre-processing step, which helps to
retrieve context. The pre-processing step includes two major steps first one is stop word removal and the second one is stemming
process. The outcome of the pre-processing step is indexing consist of important keywords and their corresponding keywords. When
the user enter the keyword to the system, the ontology makes the several steps to make the refine keywords. Finally, the refine
keywords are matched with index and relevant contexts are retrieved. The experimentation process is carried out with the help of
different set of contexts to achieve the results and the performance analysis of the proposed approach is estimated by the evaluation
metrics like precision, recall and F-measure.
Keywords— Ontologies; WordNet; contexts; stemming; indexing.
-----------------------------------------------------------------------***-------------------------------------------------------------------
1. INTRODUCTION
Information Retrieval (IR) deals with the retrieval of all
contexts, which contain information relevant to any information
need expressed by any user’s query. The methodological rule
given in literature is to begin an evaluation by analyzing what is
the objective of the system, process or service to be evaluated
[1] [2]. It is assessed that to what extent the object of evaluation
attains the defined goals. Therefore, it is necessary to identify
the goals of the system, and measures of goal attainment and
criteria for achieving goals [3] [4].
An Information Retrieval System (IRS) consists of a software
program that facilitates a user in finding the information the
user needs [5]. IR provides the contexts that satisfy their needs.
IRS has to extract the key words from the contexts and assign
weights for each keyword. Recently, however, researchers have
undertaken the task of understanding the human, or user, role in
IR [6] [7]. The basic assumption behind these efforts is that we
cannot design effective IR systems without some knowledge of
how users interact with them. Therefore, the research that
studies users in the process of directly consulting an IR system
is called interactive information retrieval (IIR) [8] [9].
Query efficiency must be ensured to find out whether the
queries are running fast. Query Effectiveness also affects the
IRS since the retrieved result set must be relevant [10].
Research in IR includes modeling, context classification and
categorization, system’s architecture, user interfaces, data
visualization, filtering, languages, etc. A global perspective
holds that all of the factors that influence and interact with a
user, such as search intermediary, IR system, and texts, should
be considered in IR research [11] [12]. The design variables put
forth by Ingwersen show the wide-ranging influence of factors
such as social environment, IR system, information objects,
intermediary, and user [13].
The main assumption is that context does not change in time.
However, this assumption is unlikely. Consider Relevance
Feedback (RF) technique, the idea behind RF is that the first
retrieval operation can be considered as an “initial query
formulation” [14]. Some initially retrieved items are examined
for relevance, and then the automatic modification of the query
can be performed by the system by using the feedback collected
from the user for instance adding keywords, selecting and
marking contexts [15]. The modified query can be considered a
“refinement” of the initial query. A possible solution is the
adoption of techniques, which are transparent to the user that is
“implicit”. Implicit Relevance Feedback (IRF) techniques [16]
[17] can use different contextual features collected during the
interaction between the user and the system in order to suggest
query expansion terms, retrieve new search results, or

_______________________________________________________________________________________
dynamically reorder existing results. One of the difficulties with
this kind of techniques is the need of combining different
sources of evidence, i.e. different contextual features [19]. The
complexity of these approaches is one of the reasons for
investigating the problem in a principled way that is for the
adoption of a model-based development [19]. One of the
benefits of this approach is that all the assumptions are made
explicit: this is crucial in modeling context in order to
understand which elements of the context are actually
considered, and in which way the relationship between such
elements is modeled [20].
2. RELATED RESEARCHES: A REVIEW
Despite a plenty of works available in the literature, a handful
of significant research works are reviewed here. Xuehua Shen et
al. [21] have proposed a method for retrieval models and
systems that the retrieval decision was made based solely on the
query and document collection; information about the actual
user and search context. In this proposed method, they studied
how to exploit implicit feedback information, including
previous queries and click through information, to improve
retrieval accuracy in an interactive information retrieval setting.
They proposed context-sensitive retrieval algorithms based on
statistical language models to combine the preceding queries
and clicked context summaries with the current query for better
ranking of documents. They used the TREC AP data to create a
test collection with search context information, and
quantitatively evaluate our models using this test set.
Emanuele Di Buccio et al. [22] have proposed a technique for
an information retrieval (IR) system documents according to
their predicted relevance to a formulated query. In this proposed
method, for each user it is assumed one information need for
each query, one location where the user is, and no temporal
dimension. Exploiting the context in a way that does not require
a high user effort may be effective in IR as suggested. The high
number of factors to be considered by these techniques suggests
the adoption of a theoretical framework, which naturally
incorporates multiple sources of evidence. Moreover, the
information provided by the context might be a useful source of
evidence in order to personalize the results returned to the user.
Indeed, the information need arises and evolves in the present
and past context of the user. Since the context changes in time,
modeling the way in which the context evolves contributes to
achieve personalization.
Massimo Melucci et al. [23] have proposed a method for
Information retrieval for context model by vector space base
and its evolution was modeled by linear transformations from
one base to another. Each document or query can be associated
to a distinct base, which corresponds to one context. They
proposed to discover contexts from document, query or groups
or them. Linear algebra could do thus by employed in a
mathematical framework to process context, its evolution and
application.
Massimo Melucci et al. [24] have proposed Information
retrieval (IR) model based on vector spaces have been
investigated for a long time. Nevertheless, they have recently
attracted much research interest. In parallel, context has been
rediscovered as a crucial issue in information retrieval. This
article presents a principled approach to modeling context and
its role in ranking information objects using vector spaces. First,
the article outlines how a basis of a vector space naturally
represents context, both its properties and factors. Second, a
ranking function computes the probability of context in the
objects represented in a vector space, namely, the probability
that a contextual factor has affected the preparation of an object.
David Robins et al. [25] have introduced interactive information
retrieval systems. Interactive information retrieval may be
contrasted with the "system entered" view of information
retrieval in which changes to information retrieval system
variables are manipulated in isolation from users in laboratory
situations. In this proposed method, they elucidates current
models of interactive information retrieval, namely, the episodic
model, the stratified model, the interactive feedback and search
process model, and the global model of poly representation.
3. PROPOSED METHODOLOGY
Here we proposed a new IR method which is used for
recovering traceability links between code and documentation.
To access the large database, initially the database will be
partitioned by using Jensen-Shannon (JS) method. The JS will
be constructed by partitioning the database into smaller sizes.
WordNet is an online lexical database of English, developed
under the guidance of Miller at Princeton University. Here, a set
of cognitive synonyms called synsets, each representing a
different concept, are formed by grouping the nouns, verbs,
adjectives and adverbs. Synsets are created by using conceptual
semantic and lexical relations. WordNet can also be seen as
ontology for natural language terms. It has more than 100000
words, organized into taxonomic hierarchies. Nouns, verbs,
adjectives and adverbs are grouped into synonym sets (synsets).
The synsets are also grouped into senses i.e., diverse meanings
of the same word or concept. Same as the Open Directory, the
synset ids are altered when new versions of the ontology are
published, however a backward compatibility utility program is
used to map synsets between the versions.
4. PROPOSED APPROACH OF DESIGN AND
IMPLEMENTATION OF AN ONTOLOGY-BASED
CONTEXT RETRIEVAL
The aim of this proposed research is to design and develop an
approach for the context retrieval by combining keyword in
ontology platform. Initially, a user submits the keywords into
the system, the ontology operates with the keywords, and a list
of contexts is retrieved from the document repository. Initially
the system find out the possible synsets (set of synonyms) for

_______________________________________________________________________________________
the each keyword as the user entered. Subsequently the system
makes the possible combinations of all keywords from the
synsets. The neighborhoods are a set of words that are relevant
to the combinations. From the collection of neighborhoods, the
system count the frequency of the each keyword, if the keyword
is supported by the minimum support then the words go to the
representation table else the correspond word get neglected. If
no words are present after given the minimum _support then the
user, get the chance to provide the relevant keyword by the
keyword-refining schema. The distance measure helps to find
out the refined keywords from the representation. Finally, the
keywords are matched with the indexing and relevant context
are retrieved.
Fig 1: The architectural diagram of the proposed approach
4.1 Preprocessing
In the proposed approach, there are some complexities to deal
with the context in its original format since we have to do some
pre-processing techniques to make the context repository
prepare for our proposed method to retrieve the relevant
contexts based on keywords given by the user. The main
objective of this pre-processing is to obtain the important
keyword from the all contexts present in the database
repository. Finding of important keyword from the document
repository is not an easy task because each of the context
contains a vast amount of common words and branch words. In
order to remove those kinds of words from the context the
following methods are going to use in the pre-processing phase.
The pre-processing step mainly consists of three steps first one
is stop word removing, the second one is stemming algorithm,
third one is similarity measure.
4.1.1 Deletion of Stop Words
It is difficult to select keywords in contexts, which have a bulk
number of words. Picking the keywords among the huge
number of words in a context can be achieved through the stop
word removing. The general words (such as was, is, the) are
removed through stop word removing process in order to
extract the keywords from a context. Because of this procedure,
only important words are left as a residue. The major reason of
eliminating stop words is to conserve the system resources by
deleting those words that have little value for mining procedure.
The common words that are noticed as stop words consists of
function word and a few more (i.e. articles, conjunctions,
interjections, prepositions, pronouns). Stop words like “it”, “a”,
“can”, “an”, "and", "by", "for", "from", "of", "the", "to", "with"
are the common stop words.
4.1.2 Stemming Process
The stemming algorithm has filtered token, this token has the
branch words with the root word and this will help to find out
the documents, which contain the branch words of the root
words. For instance, if a query includes the word walk, the user
may desire documents that contain the word walks, walking or
walked. This process helps to reduce the need of memory space
while indexing process and it helps to make better finding out
the relevant documents.
4.2 Indexing
The document retrieval system prepares for retrieval by
indexing the documents and formulating the keywords,
resulting in document representations and keyword
representations respectively. Automatic indexing begins with
the important keywords, such as extracting all the words from a
text, followed by refinements in accordance with the conceptual
schema. After finishing the pre-processing process, the
documents contain only the keywords. The system calculates
the frequency of keyword in all of the documents. This process
gives the count of each keyword. To select the important
keywords, the user sets the threshold value. The keyword is
preferred as an important keyword when the following
condition 1 is satisfied. If the count value of the keyword is
greater than the threshold value then that keyword considered as
important keyword.
ThKCntKimp ii  )()(
(1)
Here, the indexing is done with the aid of the inverted indexing
method so that the matching process can be done easily. The
index   ii DKimpI ,
consists of important keywords
)( iKimp
and their corresponded documents iD
.
Key Words
Pre-
processing
Removal
and
stemmin
g
process
Similarity
measure
Keyword
Refining
Scheme
Relev
ant
Conte
xt
docu
ments
Ontologies
Process
Indexing
Process
Docum
ents

_______________________________________________________________________________________
4.2.1 Retrieval of Relevant Documents by Using the
Proposed Similarity Measure
The user gives the keyword to ontology to retrieve the relevant
document, in order to achieve that goal; our proposed similarity
measure uses the following approaches. Synset iS : which helps
to find out the relevant synonyms of the keywords as the user
entered from the ontology.
A). Combination: the combination of keywords helps to find
out the relevant neighborhood from the ontology.
B). Neighborhood: which helps to find out the nearest words of
the combinations, which made in the previous section.
C). Representation: which helps to find out the relevant
keywords from the collection of nearest words of the
combination. If the frequency of the nearest word is greater than
the min-support then that word considered as relevant keyword.
If there is no words having the frequency greater than min-
support then the user has the chance to give the relevant
keyword through a keyword-refining schema. Each of the
keywords from the representation is matched with the
neighborhoods to find out the distance between them. The
combination of the keywords selects as a refined keyword one
who having the maximum distance among them. Finally, the
keywords are matched with the index and relevant documents
are retrieved from it.
4.3.1 Synsets
Synset is a set of synonyms of the keyword that are given by the
user with the aim of retrieving the relevant document from the
repository. User given the set of keywords }{ iKK 
where
ni 1 are the keywords to the ontology, the synset of each
keyword
  OKKKS kijiii  ,
where j and k represents
the synset of the keyword iK
, the synset of the keyword iK
is
get from the ontologyO.
4.3.2 Combination of all Keywords from the Synset
Each of synset have set of keywords each of them are combined
with the other keywords for finding the efficient neighborhood
for the keyword. The combination of the keyword helps to find
out the accurate document through the neighborhood. Every
keywords of the each synset combines with other
keywords
 kiji
m
ji KKC ,
where
kji ,, represents the
keywords in the synsets and m represents the identity of
combination.
4.3.3 Neighborhood
The neighborhoods are nearest words of the keyword present in
the above of combinations. This will help to improve the quality
of the keywords. The neighborhoods are represented for each
combination
  ONNC m
q
m
pm  , qp NandN
are contains the
set of nearest words
    OKNKN m
qi
m
q
m
pi
m
p  ,
of the
keyword kiji KandK
respectively. Likewise, all of the above
combinations have the neighborhoods from the ontology.
4.3.4 Representation
Representation is used to find out the important keywords from
the neighborhoods of all combinations. Each keyword
    OKNKN m
qi
m
q
m
pi
m
p  ,
of the neighborhood of
qp NandN
has the count value. The keywords are removed
when their count value less than the min-support that is given
by the user. The
representation
  supmin_|  m
pipip KCntRR
,
  supmin_|  m
qiqiq KCntRR
. Finally
qp RandR
have the set of
keywords
   qiqpip KRKR  ,
. If the representation
qp RandR
has no keywords after given the value of
supmin_ the given by the user, subsequently the user has the
chance to provide the relevant keywords through the keyword
refining schema.
Pseudo Code
INPUT: keywords, iK ;
OUTPUT: Relevant Documents DR
ASSUMPTIONS:
)( iKCnt  Count of Keyword )( iK
Th  Threshold
iS
Synsets of keyword )( iK
Th  Ontology
mC  Combination of Synset and Keywords
 m
q
m
p NN ,  Neighborhoods of Combination mC
  m
qi
m
pi KK ,  Set of keywords belongs to Neighborhoods
 m
q
m
p NN ,
pR  Representation
  ii DKimpI ,  Indexed documents (  iKimp refer to
important keyword and iD refers to corresponding
documents)
mD  Distance measure
DR  Relevant Document

_______________________________________________________________________________________
Pseudo code:
Begin
Step1: Pre-processing of documents by stop word removal
and stemming process
Count the Keywords in a set of document )( iKCnt
 ThKCntif i )(
 ii KimpasdenotedK )(
documentsfromremovedK
else
i
Step2:   ii DKimpIindexConstruct ,
Step3: }{ iKKkeywordGet 
Step4:   OKKKSObtain kijiii  ,
Step 5:  kiji
m
ji KKCObtain ,
Step6:
tobelongsodsNeighborhoGet
  ONNC m
q
m
pm  ,
qp RandRationrepreesentCompute
  supmin_|  m
pipip KCntRR
  supmin_|  m
qiqiq KCntRR
Step 7: Compute Distance q
m
qp
m
pm RNRND  
Select m
jiC Which has max Value of mD
Step 8: Get the Relevant Key Words From  kiji
m
ji KKC ,
kiji KK , are match with   ii DKimpI ,
Step 9: Get Relevant Document   iii DKimpIfromRD ,
End
4.3.5 Finding of the Refined Keywords from the
Representation
The keywords from     OKNKN m
qi
m
q
m
pi
m
p  , are
comparable with the representative    qiqpip KNKR  , to
find out the distance between them. The distance calculation is
done by the following equation 2.
q
m
qp
m
pm RNRND  
This distance calculation is done for all the combinations
mC
and we get the set of distance measure from that we choose
the distance value mD
which having the maximum value. The
corresponding combination is extracted with the help of
m
q
m
p NandN
since the neighborhoods are a subset of the
combination. The combination has the set of keywords
 kiji
m
ji KKC ,
that are considered as refined keyword.
4.3.6 Finding of the Relevant Document
The refined keywords kiji KK , are matched to the index
  ii DKimpI , if the keywords are same then the
corresponding relevant documents DR are retrieved.
5. RESULTS AND DISCUSSION
The results obtained from the experimentation of the proposed
cross ontology-based similarity measure for bio-document
retrieval system is presented in this section. We have
implemented our proposed bio-document retrieval system using
Java (JDK 1.6). The dataset utilized in our experimental results
are bio-medical documents obtained from the PubMed
database.
5.1 Evaluation Metrics
An evaluation metric is used to evaluate the effectiveness of
document retrieval systems and to justify theoretical and
practical developments of these systems. It consists of a set of
measures that follow a common underlying evaluation
methodology. Some of the metrics that we have chosen for our
evaluation purpose are Recall, Precision and the F-measure.
Precision,
documentsretrieved
documentreleventdocumentrelevant
P
}{}{ 

Recall,
documentsrelevent
documentretrieveddocumentrelevant
R
}{}{ 

F- Measure,
2
( )
PR
F
P R


As suggested by the above equations in the field of Document
retrieval, Precision is the fraction of retrieved documents that
are relevant to the search, Recall is the fraction of the
documents that are relevant to the query that are successfully
retrieved and the F-measure that combines precision and recall
is the harmonic mean of precision and recall.

_______________________________________________________________________________________
5.2 Performance Analysis
The performance of the proposed document retrieval system is
evaluated based on the input query keywords to the WordNet
ontology using the proposed similarity measure. Here, we have
utilized four query keywords and the corresponding refined
keywords are extracted from the WordNet ontology using the
proposed similarity measure. We have analyzed our proposed
system with query keywords with the refined keyword. The
table 2 lists the obtained values for the evaluation measures
with query keywords and the relevant keywords. It reveals that
the proposed system works fine in the similarity measure
process.
Table 1. Refined keywords for the input query keyword
Query keyword Max-
distanceQuery Keyword Refined Word
Software processing development software 7
Sequential pattern Sequential structure 14
Computer Graphics
host computer
Graphics 16
Digital Image
process Process 18
5.3 Performance Analysis using Evaluation Metrics
The performance of the proposed document retrieval system is
evaluated based on the input query keywords to the WordNet
ontology using the Precision, recall and F-measure. Here, we
have utilized four query keywords and the corresponding
documents are obtained from the document repository. We have
analyzed our proposed system with different keywords with the
relevant and retrieved documents. The table 2 lists the obtained
values for the evaluation measures with different keywords and
the relevant documents as 20. It reveals that the proposed
system works fine in the document retrieving process.
Table 2. Precision, Recall and F-measure for different keywords
Query keyword
Relevant
documents
Retrieved
documents
Precision Recall
F-
measure
Software processing
Development software,
software documentation 7 10 1 0.8 0.8888
Sequential pattern Sequential structure 9 10
0.9677 0.8 0.9836
Computer Graphics
host computer, computer
graphics, host 10 19 0.8569 0.8 0.8957
Digital Image process
Process
8 10 0.8 0.6 0. 6153
6. CONCLUSION
In this paper, we have presented the design and implementation
of ontology based document retrieval approach. At first, the set
of keywords is extracted from the documents as the outcomes
of the pre-processing steps. The indexing process makes the
important keyword and their corresponding documents. The
refined keywords are extracted by the proposed similarity
measure after the user given the input keyword to the system.
The refined keywords are matched with the index and
corresponding documents are retrieved. Finally, the refine
keywords are matched with index and relevant documents are
retrieved. The experimentation process is carried out with the
help of different set of documents to achieve the results and the
performance analysis of the proposed approach is estimated by
the evaluation metrics like precision, recall and F-measure.
REFERENCES
[1]. Carla Teixeira Lopes “Context Features and their use in
Information Retrieval”, the preceding of 3rd Symposium on
Future Directions in Information Access (FDIA), pp. 36-42,
[2]. Abdelkrim Bouramoul, Mohamed-Khireddine Kholladi,
and Bich-Lien Doan “Using Context to Improve the Evaluation
of Information Retrieval Systems”, International Journal of
Database Management Systems, Vol.3, No.2, pp. 22-39, 2011.
[3]. Ali Bahrami, Jun Yuan, Paul R. Smart and Nigel R.
Shadbolt, “Context Aware Information Retrieval for Enhanced
Situation Awareness”, IEEE Military Communications
Conference, Orlando, FL, USA, pp. 1-6, 2007.
[4]. Peter D. Turney and Patrick Pantel “From Frequency to
Meaning: Vector Space Models of Semantics”, Journal of
Artificial Intelligence Research, Vol. 37, pp. 141-188, 2010.
[5]. Castells P., Fernandez M. and Vallet D. “An Adaptation of
the Vector-Space Model for Ontology-Based Information
Retrieval”, IEEE Transactions on Knowledge and Data
Engineering, Vol. 19, No. 2, pp. 261 – 272, 2007.
[6]. Michael W. Berry, Zlatko Drmac, and Elizabeth R. Jessup,
“Matrices, Vector Spaces, and Information Retrieval”, Society
for Industrial and Applied Mathematics, Vol. 41, No. 2, pp.
335–362, 1999.
[7]. Gerard Salton and Christopher Buckley “Term-weighting
approaches in automatic text retrieval”, Information Processing
& Management, Vol. 24, No. 5, pp. 513–523, 1988.

_______________________________________________________________________________________
[8]. Vijay V. Raghavan and S. K. M. Wong “A Critical
Analysis of Vector Space Model for Information Retrieval”,
Journal of the American Society for Information Science, Vol.
37, No. 5, pp. 279-287, 1986.
[9]. Faloutsos, Christos, Oard and Douglas W. “A Survey of
Information Retrieval and Filtering Methods”, Technical
Reports of the Computer Science Department, 1998.
[10]. Robert M. Losee “Learning syntactic rules and tags with
genetic algorithms for information retrieval and filtering: An
empirical basis for grammatical rules”, Information Processing
& Management, Vol. 32, No. 2, pp. 185–197, 1996.
[11]. Lee D.L., Huei Chuang and Seamons K. “Document
ranking and the vector-space model”, IEEE Software, Vol. 14,
No. 2, pp. 67-75, 1997.
[12]. S. K. M Wong and Y. Y Yao “A probabilistic inference
model for information retrieval”, Information Systems, Vol.
16, No. 3, pp. 301–321, 1999.
[13]. Marko Balabanovic and Yoav Shoham, “Learning
Information Retrieval Agents: Experiments with Automated
Web Browning” AAAI Technical Report SS, pp. 13-18, 2008.
[14]. Osinski, S.; Weiss, D., "A concept-driven algorithm for
clustering search results," IEEE Intelligent Systems, Vol. 20,
No. 3, pp. 48-54, 2005.
[15]. Hayes, J. H.; Dekhtyar, A. and Osborne J., "Improving
requirements tracing via information retrieval," Proceedings
11th IEEE International Requirements Engineering
Conference, pp.138-147, 2003.
[16]. Holger Billhardt, Daniel Borrajo and Victor Maojo “A
context vector model for information retrieval”, Journal of the
American Society for Information Science and Technology,
Vol. 53, No. 3, pp. 236–249, 2002
[17]. Bellegarda J. R. “Latent semantic mapping [information
retrieval]”, IEEE Signal Processing Magazine, Vol. 22, No. 5,
pp. 70-80, 2005.
[18]. Oliveto R., Gethers M., Poshyvanyk D. and De Lucia A.,
"On the Equivalence of Information Retrieval Methods for
Automated Traceability Link Recovery," IEEE 18th
International Conference on Program Comprehension (ICPC),
pp. 68-71, 2010.
[19]. Jean Véronis “HyperLex: lexical cartography for
information retrieval”, Computer Speech & Language, Vol. 18,
No. 3, pp. 223–252, 2004.
[20]. Praveen Pathak; Gordon, M.; Fan, W., "Effective
information retrieval using genetic algorithms based matching
functions adaptation," Proceedings of the 33rd Annual Hawaii
International Conference on System Sciences, Vol. 1, pp. 1-8,
2000.
[21]. Xuehua Shen, Bin Tan and Cheng Xiang Zhai “Context-
Sensitive Information Retrieval Using Implicit Feedback”, In
proceedings of the 28th annual international ACM SIGIR
conference on Research and development in information
retrieval, pp. 43-50, ACM New York, NY, USA, 2005.
[22]. Emanuele Di Buccio “Modeling the Evolution of Context
in Information Retrieval”, the 2nd BCS-IRSG Symposium on
Future Directions in Information Access, pp. 6-12, 2008.
[23]. Massimo Melucci “Context Modeling and Discovery
using Vector Space Bases” Proceedings of the 14th ACM
international conference on Information and knowledge
management, ACM New York, NY, USA, pp. 808-815, 2005
[24]. Massimo Melucci “A Basis for Information Retrieval in
Context”, ACM Transactions on Information Systems, Vol. 26,
No. 3, 2008.
[25]. David Robins “Interactive Information Retrieval: Context
and Basic Notions”, Informing Science Journal, Vol. 3, pp. 57-
62, 2000.

An efficient information retrieval ontology system based indexing for context

More Related Content

What's hot (20)

Similar to An efficient information retrieval ontology system based indexing for context (20)

More from eSAT Journals (20)

Recently uploaded (20)

An efficient information retrieval ontology system based indexing for context