International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-9 Issue-6, April 2020
2263
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number: F3879049620/2020©BEIESP
DOI: 10.35940/ijitee.F3879.049620
Handwritten Digit Recognition by using Pattern
Recognition & Consensus Clustering
SubbaRao Gogulamudi, Vital Kumar Pinnela, Lakshmi Sai Tejaswi Pathuri, RamTeja Borra
Abstract: In Big Data, Pattern Recognition and Consensus
Clustering procedures have developing significance to the
scholastic and expert networks. Today there is an extraordinary
worry for ordering the information, as information in wrong
classification implies incorrect data, which thus results wastage
of resources and hurting the association. Example
acknowledgment (PR) helps in maintaining a strategic distance
from poor order of information by recognizing the right structure
of information in dataset. Perceiving an example is the
computerized procedure of finding the specific match and
regularities of information, which is firmly identified with
Artificial Intelligence and Machine Learning. PR goes about as
an essential advance to give bunching since it examinations the
structure and vector estimation of every character in dataset.
Accord Clustering (CC) additionally called as bunching
gatherings, assumes a critical job in arranging and keep up in
any sort of information. This is a strategy that joins different
bunching answers forget steady, precise and novel outcomes.
Right now, actualize PR and CC strategies; we use MNIST
dataset which is an enormous database of transcribed digits that
is regularly utilized for preparing different frameworks in the
field of Machine Learning.
Keywords: Consensus Clustering, Pattern recognition,
MNIST Dataset, Handwritten digit recognition.
I. INTRODUCTION
A. Patter Recognition(PR)
Today’s digital world is filled with Patterns and these pat-
terns can either be observed physically or it can be derived
mathematically with the help of certain algorithms. Some of
the examples of pattern are colours on clothes, speech pat-
tern, alphanumerics in data, etc. In computer science, pattern
is referred by the value of its vector features. In a simple PR
application [11], initially the raw data is processed then the
resultant data is converted into a machine understandable
format. This involves two major phases of operation i.e.,
classification of patterns and clustering the patterns. In clas-
sification phase, each pattern is assigned an appropriate la-
bel based on the abstraction and is controlled by supervised
learning. In clustering, the pattern with the same label is
partitioned and categorized under one group and is con-
trolled by unsupervised learning.
Revised Manuscript Received on April 09, 2020.
SubbaRao Gogulamudi,CSE Department, Koneru Lakshmaiah
Education Foundation, Green Fields, Vaddeswaram, Andhra Pradesh.
Vital Kumar Pinnela,CSE Department, Koneru Lakshmaiah Education
Foundation, Green Fields, Vaddeswaram, Andhra Pradesh.
Lakshmi Sai Tejaswi Pathuri,CSE Department, Koneru Lakshmaiah
Education Foundation, Green Fields, Vaddeswaram, Andhra Pradesh.
RamTeja Borra,CSE Department, Koneru Lakshmaiah Education
Foundation, Green Fields, Vaddeswaram, Andhra Pradesh.
B. B.MNIST Dataset
The MNIST dataset (Modified National Institute of
Standards and Technology database) is an enormous
database of manually written digits that is ordinarily utilized
for preparing frameworks. The MNIST database is a subset
of an a lot bigger dataset known as the NIST Special
Database [9]. This dataset contains both transcribed
numerals and letters. It speaks to an a lot bigger and
increasingly broad grouping task, alongside the chance of
including progressively complex assignments, for example,
semantic understandings through words translation. The
openness of this MNIST dataset has unquestionably added
to its far reaching use. The entire dataset is nearly little
(when contrasted with numerous ongoing benchmarking
dataset), allowed to access and use, and is encoded and put
away in altogether direct way. The encoding doesn't depend
on complex stockpiling structure, compressions, or any
information design. Hence, it is made extremely simple to
access and this dataset can be incorporated from any stage
or through any programming dialects. The NIST dataset, by
complexity to the MNIST, has stayed hard to access and
use. Coming about to the greater expense and the
accessibility of storage when it was gathered, the NIST
dataset was initially put away in an effective and
conservative way. In spite of the fact that source code to get
to the information is given, it was exceptionally testing to
use on some ongoing processing stages. Thus, the NIST
have as of late discharged a second release of the NIST
dataset. Be that as it may, the encoding of that datasetstays
in the first configuration from which MNIST was extricated.
The portrayal of the MNIST dataset is appeared in figure
Fig. 1. MNIST Dataset
C. Consensus Clustering(CC)
As we all know, Clustering defines the process of gathering
a specific set of data related to some category and aggregat-
ing them with respect to their characteristics. Whereas Con-
sensus Clustering (CC) also
known as aggregated clustering,