You are on page 1of 4

A REVIEW ON DATA MINING TECHNIQUES AND

THEIR APPLICATIONS

SAI NARENDRA VARMA SURYA VENKATA NAGA SUBA AMRUTHA PAVAN KUMAR POLA Mr. VISWANATH REDDY
UPPALAPATI PRAVEEN MUSTI
COMPUTER SCIENCE COMPUTER SCIENCE COMPUTER SCIENCE
COMPUTER SCIENCE COMPUTER SCIENCE AND ENGINEERING AND ENGINEERING AND ENGINEERING
AND ENGINEERING AND ENGINEERING
NADIMPALLI NADIMPALLI NADIMPALLI
NADIMPALLI NADIMPALLI SATYANARAYANA RAJU SATYANARAYANA RAJU SATYANARAYANA RAJU
SATYANARAYANA RAJU SATYANARAYANA RAJU INSTITUTE OF INSTITUTE OF INSTITUTE OF
INSTITUTE OF INSTITUTE OF TECHNOLOGY TECHNOLOGY TECHNOLOGY
TECHNOLOGY TECHNOLOGY

sainarendravarmavarma msvpraveen2@gmail.co amruthha.varada23@g pavanpola1712@gmail.


m co
@gmail.com mail.com

ABSTRACT INTRODUCTION

Data mining is a process which finds The process of extraction of useful


useful patterns from large amount of past and information and patterns from huge amount of
present data. It is used in different fields like data from various domains. The research in
science, Engineering, Health, Business etc... databases and information technology has given
The paper discusses few of the data mining rise to an approach to store and analyze this
techniques and some of the organizations which precious data for further decision making. It is
have adapted data mining technology to improve also called as knowledge discovery process (KDP),
and enhance their businesses and organizations knowledge mining from selected data to find
to found excellent results and shows the Data relationships or patterns.
mining scope in future.

Keywords

Datamining, Knowledge data discovery,


techniques, Applications.

Fig: KDD [1]

1
2) Graph-based data

3) Sequential data
LITERATURE SURVEY
DATA MINING TECHNIQUES:
A study of Datamining techniques and their Based on the type of the task the Datamining
applications techniques are applied.
By this process, many companies got profits on Predictive tasks provide the results of future
their respective domains. it Increases efficiency of queries based on past data.
marketing campaigns and also increases the Classification, Regression and Outer detection
cross-selling to existing customers are predictive data mining techniques.
[1] Soft map Company Ltd. Tokyo, Association rules, Sequential patterns and
Page views increased 67% per month after the prediction are few most commonly used data
recommendation engine went live. mining techniques.
Profits tripled in 2001, as sales increased 18 Classification:
percent versus the same period in the previous Classification is the most commonly applied data
year. mining technique, which invokes a set of pre-
[2] Standard Life Mutual Financial Services classified examples to develop a model that can
Companies classify the population of records at large.
Achieved, with the model, a nine times greater We use Decision Trees, Bayesian Classifiers,
response than that achieved by the control group. Neural Networks, K-Nearest Neighbors, Support
Secured $47 million worth of mortgage Vector Machines, Linear Regression, Logistic
application revenue. Regression, as classifiers in this technique.
[3] Shenandoah Life insurance company United Clustering:
States.
Reduced the time required to issue certain policies By using clustering techniques, we can further

by 20 %. identify dense and sparse regions in object space

Improved underwriting and employee and can discover overall distribution

performance review processes.


pattern and correlations among data attributes.
We use different clustering methods for different
[4] FBTO Dutch Insurance Company
applications. Some methods are Partitioning
Decreased mailing costs by 35 %.
Method, Grid-Based Method, Density-based
Increased conversion rates by 40 %.
Method, Model-Based Method, Hierarchical
Method, Constraint-based Method.
METHODOLOGY
In the process of data mining, choosing a dataset Outlier Detection:

from a huge repository is the primary thing.


Outlier detection detects and excludes outliers
Datasets are divided into three types:
from the data set. Some outlier detection methods
1) Record data are Z-Score, DBSCAN, Isolation Forest, Linear

2
Regression Models. Fraud detection, Intrusion Classification, Association Rule mining,
detection, Medical and health outlier detection, Regression in data mining is used to predict
Fraud detection of Insurance claim are the product development time and cost, the
applications of outlier detection. relationship between product architecture,
APPLICATIONS OF DATA MINING: customer needs, dependencies among tasks etc.
Data mining is applied vastly in many Data mining tools used in this field are Rapid
organizations. miner, Data melt, Board, Weka.
[1] Bioinformatics:
Bioinformatics is the collection of various [3] Criminal Investigation:
methods to manage, store and study biological Criminal analysis includes detecting crimes and
data using computers. the data mining tools used criminal’s relationships with these crimes. From
in bioinformatics are BLAST (Basic Local different crimes like cyber-crimes, violent crimes,
Alignment Search Tool), FASTA, CS-BLAST for fraud detection, drug offences, we get high
finding sequence alignment, Gen-Scan, Gene- volumes of criminal datasets. Data mining is
Mark for gene finding, P-fam, BLOCKS, Pro- utilized in this field for applications like counter-
Dom for protein analysis. terrorism activities, crime matching, crime trends,
etc. Data mining tools used in this field are Weka,
[2] Manufacturing-Engineering: H2o, Orange etc. are field.
Manufacturing enterprise contains data related to
its company's products. Techniques like

CONCLUSION [2]. Dr. Gary Parker, vol 7, (2004), Data Mining:

Data mining has importance regarding finding Modules in emerging fields, CD-ROM.

the patterns, forecasting, discovery of knowledge [3]. Crisp-DM 1.0 Step by step Data Mining guide

in different business domains. Data mining from http://www.crisp-dm.org/CRISPWP-

techniques and algorithms such as classification, 0800.pdf.

clustering [4]. Customer Successes in your industry from

etc., helps in finding the patterns to decide upon http://www.spss.com/success/?

the future trends in businesses to grow. source=homepage&hpzone=nav_bar.

Now a days almost every field is digitalized these


[5]. https://www.allbusiness.com/Technology
days, and because of this, a
/computer-software-data-management/ 633425-
large volume of data is generated every day. Data
1.html, last retrieved on 15th Aug (2010).
mining plays a vital role in future-prediction.
plays a vital role in managing, analysing and [6]. http://www.kdnuggets.com/.Pu

extracting the
[7] Dr. M. Dhanabhakyam , Dr. M. Netravali ,
required information from these large databases.
―A Survey on Data Mining Algorithm for
REFERENCES Market Basket Analysis‖ in Global Journal of
[1]. Jiawei Han and Micheline Kamber (2006), Computer Science and Technology Volume 11
Data Mining Concepts and Techniques, published Issue 11 Version 1.0 July 2011, Publisher: Global
by Morgan Kauffman, 2nd ed.

3
Journals Inc. (USA) Online ISSN: 0975-4172 & [9] V.K. Jha, R.K. Singh ―Application of Data
Print ISSN: 0975-4350. Mining in Manufacturing Industry‖ in
International Journal of Information Sciences
[8] Stefano Lunardi, Jake Chen, ―Data Mining
and Application. ISSN 0974- 2255 Volume 3,
in Bioinformatics: Selected Papers from
Number 2 (2011), pp. 59-64.
BIOKDD‖ in IEEE/ACM Transactions on
[10] Brijendra Singh, Hemant Kumar Singh,
Computational Biology and Bioinformatics, Vol.
―Web Data Mining Research: A Survey‖ in
7, no. 2, April-June (2010)
IEEE International INTERNATIONAL
JOURNAL OF SCIENTIFIC &
TECHNOLOGY RESEARCH VOLUME 9,
ISSUE 02, FEBRUARY (2020) ISSN 2277-8616
3388 IJSTR©2020 www.ijstr.org Conference on
Computational Intelligence and Computing
Research, (2010).

You might also like