Professional Documents
Culture Documents
A Survey Paper
SN Saurav
BE/25047/13
BIT Mesra Jaipur
-------------------------------------------------------------------ABSTRACT--------------------------------------------------------------
Data mining has been used to uncover hidden patterns and relations to summarize the data in
ways to be useful and understandable in all types of businesses to make prediction for future
perspective. Medical data is consider most famous application to mine that data, Health care
industry produces enormous quantity of data that clutches complex information relating to
patients and their medical conditions. Data mining has an infinite potential to utilize healthcare
data more efficiently and effectually to predict different kind of disease. This paper features
various Data Mining techniques such as classification, clustering, association and also
highlights related work to analyse and predict human disease.
--------------------------------------------------------------------------------------------------------------------------------------------------
1. Introduction stay of patients in hospital, for medical diagnosis and
Data mining is also usually referred to as knowledge creating plan for active information system management.
discovery from Data (KDD). The purpose of data mining New and current technologies are used in medical field
is to mine useful information from huge databases or data to improve the medical services in cost effective manner.
warehouses The data made by the health organizations is
exact huge and difficult due to which it is hard to analyze 2. Classification of Data Mining System
the data in order to mark vital conclusion regarding
patient health. This data covers details regarding Data mining systems can be categorized
hospitals, patients, medical claims, treatment cost and according to various criteria as given below
etc. So, there is an essential to make a powerful tool for
analysing and extracting significant information from
1) Type of data sources mined
this complex data. The analysis of health data expands
2) Database involved
the healthcare by improving the presentation of patient
3) Kind of knowledge discovered
management jobs. The consequence of Data Mining 4) Mining techniques used
technologies are to make available welfares to healthcare
organization for grouping the patients having
related/similar type of diseases or health issues so that 3. Process of Data Mining
healthcare organization provides them active treatments.
Data mining process includes the following few
steps
The data made by the health organizations is
1) Data Cleaning It is used to remove
exact huge and difficult due to which it is hard to analyze
the data in order to mark vital conclusion regarding noise and inconsistent data.
patient health. This data covers details regarding 2) Data Integration It is used to
hospitals, patients, medical claims, treatment cost and combine multiple data sources.
etc. So, there is an essential to make a powerful tool for
analyzing and extracting significant information from 3) Data Selection It is used to retrieve
this complex data. The analysis of health data expands the relevant data from the database for
the healthcare by improving the presentation of patient analysis task.
management jobs. The consequence of Data Mining 4) Data Transformation It is used to
technologies are to make available welfares to healthcare transformed or consolidated data into
organization for grouping the patients having particular appropriate form for mining by
related/similar type of diseases or health issues so that performing summary or aggregation
healthcare organization provides them active treatments. operations.
It can also valuable for predicting the how many days of
In classification, make the software that can acquire how
5) Data Mining Here the intelligent
to classify the data items into groups. For instance, first
methods are applied in order to extract data apply classification in the application that given all
patterns. records of employees who left from the company;
6) Pattern Evaluation It is used to predict who will probably leave the company in a future
evaluate the data patterns. period. In hold those books in a way that readers can
take several books on a particular topic without trouble.
7) Knowledge Presentation Here the By using the clustering technique, can retain books that
knowledge is represented. have some kinds of similarities in one cluster or one
bookshelf and label it with a meaningful name. If
The following figure 1 shows the data mining process. readers need to take books in that topic, they would only
have to go to that bookshelf instead of looking for the
whole library.
4.4 Prediction
Figure 1: Process of Data Mining
Prediction is a wide topic and runs from predicting the
4. Data Mining techniques failure of components or machinery, to identifying fraud
and even the prediction of company profits. Used in
combination with the other data mining techniques,
There are enormous number of data mining prediction involves analyzing trends, classification,
techniques have been evolving and using in data pattern matching, and relation. By analyzing past events
mining projects recently. Some of the data mining
or instances, you can make a prediction about an
techniques are given below,
event.Using the credit card authorization, for example,
you might combine decision tree analysis of individual
4.1 Association past transactions with classification and historical
Association is one of the top - well known pattern matches to identify whether a transaction is
data mining techniques. In association, a pattern is fraudulent.
learned based on an association between items in the
similar transaction. Thats the purpose the association
4.5 Sequential Patterns
technique is also well-known as relation technique. The
association technique is used in marketplace basket
Sequential patterns analysis is one of data
analysis to classify a set of products that customers
mining technique that pursues to determine or recognize
regularly purchase together. Dealers are using
associated patterns, fixed events or fashions in
association technique to investigation buyers buying
transaction data over a business period. In sales, with
lifestyles. Based on ancient sale data, retailers might ancient transaction data, businesses can recognize a set
catch out that customers always buy jam when they buy
of items that customers buy together different times in
breads, and, therefore, they can put jam and breads a year. Then industries can use this information to
following to each other to save time for customer and mention customers buy it with better deals based ontheir
make steps to growth sale. purchasing regularity in the past.
4.2 Classification
4.3 Clustering
3 HIV WEKA Classificatio J 48 81.88 1. M. Durai raj, and V. Ranjani, Data Mining
/AIDS 3.6 n and Applications in Healthcare Sector: A Study,
Association international journal of scientific & technology
Rule Mining
research volume 2, issue 10, October 2013 ISSN
4 Blood WEKA Classificatio J 48 89.9 2277-
Bank n 8616
Sector
5 Brain K- Clustering MAFI 85
Cancer means A 2. D.Shobana, N.Uthra, The Data Mining Concepts
clusteri and Techniques: A Survey International Journal of
ng Trend in Research and Development, Volume 2(6),
6 Tubercu WEKA Nave Bayes KNN 78 ISSN 2394-9333
l osis Classifier
7 Diabete ANN Classificatio C 4.5 82.6
s n algorithm
3. Introduction to Data Mining and Knowledge
Mellitus
Discovery, Third Edition ISBN: 1-892095-02-5,
8 Kidney RST Classificatio Decision 75.97 Two
Dialysis n Maki Crows Corporation, 10500 Falls Road, Potomac, MD
ng 20854 (U.S.A.), 1999.
9 Dengue SPSS C 5.0 80 4. Han, J. & M. Kamber, and Data mining: concepts
Modeler and techniques, San Francisco: Morgan Kaufman
Table 1: Tools and Techniques in Healthcare
(2001).
The above table 1 show the various tools and techniques 5. D. S. Rajput, R. S. Thakur and G. S.
are used to find the accuracy level of various diseases. Thakur(2014)"Karnaugh Map Approach for
Mining Frequent Termset from Uncertain Textual
Data".
7. Conclusion