Data Mining Techniques for Analysis of Nutrition, Physical Activity and Obesity from Behavioral Risk Factor Surveillance System Data

This project aims to analyze the real-time business situation by applying data mining techniques to the Behavioral Risk Factor Surveillance System data. The project identifies the behavioral risk factors for surveillance system by analyzing the nutrition, obesity, and physical activity by using the Naive Bayes classifier. The project discusses the issues with the Naive Bayes classifier and suggests using effective rules and trees data mining algorithms for future analysis.

DATA MINING

INTRODUCTION
This project aims to analysis the real time business situation by applying
the data mining techniques. Here, we will analysis the physical activity, obesity,
nutrition from behavioral risk factor surveillance system by applying the data
mining techniques. The data mining techniques is used to provide the flexible and
effective system for sharing the evidence are in place and it is has the ability to
obtain and effectively use the best of available evidence including the surveillance
and data evaluation. It is prevent the obesity and promote the healthy behaviors. It
accesses the high quality of evidence and it has ability to use it appropriate public
health care practices including assessment, implementation and evaluation.

RELATED WORK
This paper is describes (Sunmoo YOON, 2018), to discover the
behavior correlates of chronic disease based on nutrition, obesity and
physical activity by using the data mining approach. The purpose of this
paper is to describe the data mining methods for busing classification modes
for a chronic disease using the US behavioral risk factor surveillance system
data and to illustrate the application of the methods using a case study of
depressive disorder. The data mining methods includes the six steps of data
mining to build a disease model using the classification techniques and it
use the innovative approach to analyzing the dimensionality data.

RELATED WORK
According to this paper (S. M and J. C, 2018), describes the
behavioral risk factor surveillance system ad it is a collects the information on
a wide variety of health related behaviors. This paper summarizes the
behavioral risk factor surveillance system procedures and questionnaire
design. It provides the representativeness of the evaluation and response rate.
It uses the health care data to calculate the variable including the weights
status. It analysis the critical features of behavioral risk factor surveillance
system and to survey and weighting the methodology to analysis the data sets
by using the data mining methodologies.

DATA SET
 This project uses the Nutrition, physical activity and obesity from
behavioral risk factor surveillance system data (Datasciencecentral.com,
2018).
 The dataset includes data on adult's diet, physical activity, and weight status
from Behavioral Risk Factor Surveillance System.
 This data is used for DNPAO's Data, Trends, and Maps database, which
provides national and state specific data on obesity, nutrition, physical
activity, and breastfeeding.

DATA SET

DATA MINING TECHNIQUES
The data mining is a process of the data and identifying the patterns and trends in
that information.
The big data uses the more extensive data mining techniques because the big data
has large size of the information and information trends. The data mining is used to
understand the how to relate, cluster, associate and map it with other data to produce
the result.
It also identify the source formats and data and then mapping the information to
provide the result can change and it discover the various elements and aspects of the
data by using the various data mining techniques .

DATA MINING TECHNIQUES
The naive bayes techniques are a classification techniques based on bayes theorem
with an assumption of independence among predictors. It assumes that the presence
of a particular features in a class.
 It is used to build and particularly useful for verge large data sets. It uses the highly
sophisticated classification methods. It provides a way of calculating the posterior
probability.
It fast and easy to predict the class of test data set and it perform the well in multi
class prediction.
The naive bayes classifier performs the better compare to other model like logistic
regression (EDUCBA, 2018).

ANALYZE AND DISCUSSION
 To analysis the nutrition, physical activity and obesity from behavioral risk
factor surveillance system data by using the weka data mining tool.
 The weka data mining tool is uses the classification method to provide the
effective data analysis for behavioral risk factor surveillance system.

DATA MINING ANALYSIS
 Open weka tool and upload the data set. It is shown below.

DATA MINING ANALYSIS
Naïve bayes Analysis for Age

DATA MINING ANALYSIS
Naïve bayes Analysis for Education

DATA MINING ANALYSIS
Naïve bayes Analysis for Gender

DATA MINING ANALYSIS
Naïve Bayes for Race/Ethnicity

GRAPH VISUALIZATION
 Year Start vs. Income

GRAPH VISUALIZATION
 Gender vs. Class

DISCUSSION
 In naïve bayes classifier, this classifier is used to provide the effective data analysis
compared to other classifier.
 It has easy and fast data analysis to provide the efficient system for sharing the evidence
are in place and it is has the ability to obtain and effectively use the best of available
evidence including the surveillance and data evaluation.
 Most of female genders are correctly maintaining the nutrition’s and physical activity
from behavioral risk factor. These data analyses provide the effective the data evidence
for behavioral risk factor surveillance system.

ISSUES
 The naive bayes classifiers techniques has some issues such as incomplete
training data, attributes independence and continuous variables.
 The training data is needs to compute the several conditional probabilities
which state the probability that an attribute assumes a particular value.
 The continuous variable is computing the probability by the traditional
method of frequency counts and it is not possible.
 The attributes independence is most important weakness and it requires the
little bit of extra effort.

CONCLUSION
This project aims to analysis the real time business situation by applying
the data mining techniques. Here, we will analysis the physical activity, obesity,
nutrition from behavioral risk factor surveillance system by applying the data
mining techniques. The data mining techniques is used to provide the flexible and
effective system for sharing the evidence are in place and it is has the ability to
obtain and effectively use the best of available evidence including the surveillance
and data evaluation. This project identified the behavioral risk factor for
surveillance system by analyzing the nutrition, obesity and physical activity by
using the naive bayes classifier. These are discussed in detail.

FUTURE WORK
In future, user does not use the naïve bayes classifier techniques to
analysis the nutrition, physical and obesity activity from behavioral risk
factor surveillance system because it has three main issues. It needs to
recover. So, we are using the effective rules and trees data mining algorithm
to analysis the behavioral risk factor surveillance system in future.

REFERENCES
 S. M, B. and J. C, T. (2018). CALIFORNIA BEHAVIORAL RISK FACTOR
SURVEILLANCE SYSTEM (BRFSS) SAS DATASET DOCUMENTATION AND
TECHNICAL REPORT. Public Health Survey Research Program. [online]
Available at: http://www.csus.edu/research/phsrp/docs/brfss
%202014%20codebook.pdf [Accessed 22 Apr. 2016].
 Statsoft.com. (2018). Naive Bayes Classifier. [online] Available at:
http://www.statsoft.com/textbook/naive-bayes-classifier [Accessed 31 May 2018].