Scribd
Upload
16 views65 pages

Information Retrieval

Uploaded by

Vikram Karthik
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views65 pages

Information Retrieval

Uploaded by

Vikram Karthik
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 65

lOMoARcPSD|21906275

Information Retrieval

Information Technology (University of Delhi)

Studocu is not sponsored or endorsed by any college or university


Downloaded by Karthik ([email protected])
lOMoARcPSD|21906275

INFORMATION RETRIEVAL

IR- retrieving relevant documents as per user needs.


IRS should be user-friendly, memory efficient, and retrieve efficiently.

Objectives of IR:
1. Representation
2. Storage and Organization(if the IRS is memory efficient, we can retrieve more
information at a faster rate)
3. Access

IRS are designed keeping the 3 Vs in mind:


velocity(the rate at which data is increasing in the network)
variety(multiple types of data) and Veracity(heterogeneity)
Volume

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Difference between info retr and data retrIEVAL

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Retrieval process:
User generates query
After generation passed to next stage
System query generated
Crawling
Index the documents
Ranking

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

MODELLING

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Why ranking?= to measure relevance between user query and document query
problem(point 3) of uncertainty: a single word with multiple meanings. eg- if we enter
jaguar, we don't know if the user is looking for the car or the animal

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

APPLICATIONS---

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Conditional probability classifier

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

example(case study)

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Semanticity is lost- no meaning after elimination of stop words

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Term document matrix:

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

clustering text categorization

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Probabilistic model

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Responsible to capture vagueness by degree of membership function

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

NEURAL NETWORK MODEL

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

NEW CHAPTER

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

precision= relevant retrieved / retrieved


It tells that how much our model correctly classifies
positives cases out of all actual positive cases​.
recall= relevant retrieved/ relevant
It tells that how much the model correctly predicts the
positive cases out of the cases which the model predicts
positive.

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Reference collection=standard database

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Summary table stats- accuracy


Downloaded by Karthik ([email protected])
lOMoARcPSD|21906275

Inex- exclusively for xml retrieval


Downloaded by Karthik ([email protected])
lOMoARcPSD|21906275

Newsgroup around 20k docs

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])


lOMoARcPSD|21906275

Downloaded by Karthik ([email protected])

You might also like