Comparison comprises of the most important task of retrieving

Comparison of Information Retrieval Models

 

Abstract

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Information
retrieval is a developing field of computer science which is based upon the
storage of data and retrieving them upon the request of the user. It comprises
of the most important task of retrieving relevant document rendering to the demanded
inquiry. For this task effective and operational retrieve models have been prepared
and proposed. The survey paper shows on some of these information retrieval
models. These models have been made for different datasets and purposes. A
healthy association among these models is also shown

Keywords: Information retrieval, retrieval
models.

 

Introduction

The electronic
form of information is available in a great amount and its strength is continuously
increasing. Without any information retrieval, handling information would be
impossible. Researchers start paying attention on how to obtain or extract relevant
information when the size of data increases. Primarily much of the information
retrieval technology was related to experimentation and trial error. Managing
the increasing amount of textual information available in electronic form
efficiently and effectively is very crucial. Different retrieval models have
been formed based upon different terminologies to cope and extracting the
information.

 

Information
is stored in the form of documents more often. The purpose mainly associated
for these retrieval systems is to find the needed information. The information
retrieval system is a software program which is capable of storing and managing
information in the form of documents, often textual documents but most preferable
a multimedia. The system helps the users to find information which is required
by them. A perfect retrieval system retrieves only the significant documents
but basically it is not likely as significance is mostly depended on the individual
opinion of the user.         

Basic
patterns of models:

Almost
every retrieval model includes following basic steps:

Document
Content representation
Query
representation
Query
and collection comparison

Representation of
results

 

Figure 1 Information Retrieval Process (Hiemstra, November 2009)

       Many models represent documents in
indexed form as it is efficient approach. The efficient approach of indexed
form is used for the representation of many models. For the purpose of indexing
purposes, different algorithms are used and developed since better the data is
stored, it will be retrieved in a more efficient manner.

The next
important step is the Query Formulation. Data seraching through the key words
or phrases are mostly used by the users. The query must be presented in the
same form in order to search these phrases. Indexing can be prepared by different ways depending
upon the content representation of both the documents in the collection and the
user query. (Cerulo, 2004) (Hiemstra, November
2009)

The outcomes of the retrieval system depends on its
associated algorithm therefore it regulates the precision of the system. The better
results are obtained when the comparison is better. As the outcome of this
comparison, a list of documents is obtained which can be relevant or
irrelevant. The main
purpose of a retrieval model is to measure the degree of relevance of a document
according to the given query. (Paik, August 13,
2015)

The comparison
of relevant document or irrelevant document shows that the rank of relevant
document is higher and they are shown at the topmost of the list to minimize
user time and efforts devoted in the search of documents.

 The paper is divided in different sections
where each section explains different models & their results with their significance
and limitations.

Retrieval
Models

Exact match models

 

This model
labels the documents as relevant or irrelevant. It is also known as Boolean Model, the earliest and the
easiest model to retrieve documents. It uses logical functions in the query to
retrieve the required data. George Boole’s mathematical logic operators are
combined with query terms and their respective documents to form new sets of
documents. There are three basic operators AND (logical product) OR (logical
sum) and NOT (logical difference)
(Ricardo Baeza-Yates, 2009). The resultant of AND operator is a set of
documents smaller than or equal to the document sets of any of the terms. OR
operator results in a document set that is bigger than or equal to the document
sets of single terms.

 

The documents are labelled as relevant or irrelevant by this
model. It is also identified as Boolean Model which is the earliest and the
easiest most model to retrieve. Logical functions are used in the query for
retrieving the data which is required. The
mathematical logic operators of George Boolean are combined with query terms
and their corresponding documents to custom new sets of documents. There are
three basic operators AND (logical product) OR (logical sum) and NOT (logical
difference) (Ricardo Baeza-Yates,
2009). The resultant of AND operator is a set of documents smaller than
or equal to the document sets of any of the terms. OR operator results in a
document set which is bigger or equal to the document sets of single terms.

Sense of
control over the system is given by Boolean Model. The clear distinguishing is
done between the relevant or irrelevant documents clearly if the query is found
to be accurate. This model is not supposed to rank any document as the degree
of relevance is totally ignored. This model moreover retrieved the document or
not and this causes the frustration for the end user.

 

Region models

The extension
of the Boolean model which aims about arbitrary parts of the textual data which
is called as segments, extents or regions. A region can be word, a phrase, a
text element like a title or it can be a complete document. Start position and
an end position identify the regions. Region systems are not restricted to
retrieving documents. The region models does not show a great impact on the
information retrieval research community and also not on the development of new
retrieval systems. The reason of that is the region models do not explain in any
way that how to rank the search results. In fact, most region models are not apprehensive
with ranking at all; one might say they like the relational model are actually
data models instead of information retrieval models. (Mihajlovi´)

 

Ranking Models

Important data may be skipped by Boolean Model as ranking
mechanism is not supported. Therefore a need was felt to introduce ranking
algorithms in retrieval system. The outcomes are ranked with regards to the occurrence
oif terms in the queries. Some
ranking algorithms be contingent only on the link structure of the documents whereas
some use a combination of both that they use document content along with the link
structure to allot a rank value for a given document. (Gupta, 2013)

Similarity
measures/coefficient

Document
sets and query are used and compared and the documents with more similarities
are returned back to the user. There are many methods for the users to measure
the similarities which are cosine similarity, , tf-idf etc.

Cosine
similarity

The
cosine similarity compute the angles between the vectors in n dimensional
space. The cosine similarity in d documents and d’ is given by :

( d * d’ ) / | d | * | d’ |

 

The performance of retrieval vector base model can be improved by
utilizing user-supplied information of those documents that are relevant to the
query in question. (Kita, oct 1 , 2000)

It
has been described that the vector space model for information retrieval
provides a guide to the users which are more similar and have more significance
by calculating the angle between query and the terms or the documents. Here
documents are represented as term-vectors (Vaibhav
Kant Singh, 2015).

d = (t1,
t2, t3………tn)

Where ti
=1