Arthur Samuel (1959) :
"Field of study that gives computers the ability to learn without being explicitly programmed“
Tom Mitchell (1998) :
“A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E”.
There are several ways to implement machine learning algorithms such as,
Automating automation
Getting computers to program themselves
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Predicting and visualizing the heart diseases by machine learning algorithms with big data
1. PREDICTING AND VISUALIZING THE HEART
DISEASES BY MACHINE LEARNING
ALGORITHMS WITH BIG DATA
KANNAN.R,
Research Scholar, Department of Computer Science,
Rathinam College of Arts & Science,
Coimbatore, Tamil Nadu, India.
dschennai@outlook.com
PAPER PRESENTATION – ICGICCS18
HINDUSTHAN COLLEGE OF ENGINEERING & TECHNOLOGY
COIMBATORE, TAMIL NADU, INDIA
2. Predicting and Visualizing the Heart Diseases by Machine Learning
Algorithms with Big Data
Machine LearningIntroduction Heart Diseases
Materials & Methods Experimental Evaluation Results & Conclusion
1 2 3
4 65
3. I.Introduction
• Machine Learning
• Big Data
• Heart Diseases
• Materials and Methods
• Wireless heart rate monitoring system (HRMS)
• American heart association (AHA) dataset
• Logistic regression
• ROC Curve
• Software Tools
• Experimental Evaluation
• Results and Conclusion
4. II.Machine Learning
• Arthur Samuel (1959) :
• "Field of study that gives computers the ability to learn without being explicitly programmed“
• Tom Mitchell (1998) :
• “A computer program is said to learn from experience E with respect to some task T and some performance
measure P, if its performance on T, as measured by P, improves with experience E”.
• There are several ways to implement machine learning algorithms such as,
• Automating automation
• Getting computers to program themselves
7. IV. Big Data
describes as “high-volume, high-velocity, and/or high-variety information assets that require new forms
of processing to enable enhanced decision making, insight discovery and process optimization.
8. III. Heart Diseases
• Heart Diseases is now becoming the leading cause of mortality in India with a significant risk of both males and females.
9. V(A). Wireless heart rate monitoring system (HRMS)
• To extract the heart rate, blood pressure body temperature and present location of the patients data
• Work with Android and IOS mac smart phones.
• we have obtained the 15 heart patients with 21 set of variables such as
• Name of patient,
• Age,
• Sex,
• heart rate,
• blood pressure,
• temperature,
• the patient contact no,
• patient location,
• hospital name,
• hospital contact no,
• specialist name,
• contact no
10. V(B). American heart association (AHA) dataset
We have utilized the American heart Association (AHA) dataset to compare the patient threshold value of
heart rate, blood pressure, and temperature with the patients on time data to alarm the heart patient
abnormality.
11. V (C). LOGISTIC REGRESSION
• Statistical method for analyzing a dataset in which there are one or more independent variables that determine an
outcome.
• The outcome is measured with a dichotomous variable (in which there are only two possible outcomes).
12. V (D). ROC curve
• The Receiver Operating Characteristics (ROC)
• Measure for evaluating classifier performance.
• Based on two basic evaluation measures such as,
• Specificity - completely negative part of a dataset
• Sensitivity - completely positive part
• These measures are calculated by moving threshold values across the scores.
13. VI . EXPERIMENTAL EVALUATION
In this section, we have applied and evaluated the six process to achieve the results for the below
process
• Extract the real patient’s data from the heart monitor to Smart Phone via Bluetooth,
• Transfer the patient’s data from Smart Phone to Big Data via internet,
• Use the Logistic regression and Spark ML framework in R programming for classifying and
labeling the patients on time.
• Validate the real patient’s data with American Heart Association (AHA) Dataset by ROC
curve and boosting algorithms.
• After compression, the patient, heart specialists and hospitals get alarm when the
patient abnormal by the smart phone with location.
• In addition, the patients, heart specialists and hospitals can keep track and visualize the
patient’s details in any time.
15. VI(B) . EXPERIMENTAL EVALUATION
Age wise categorized Heart Diseases Data
We have categorized the prediction data by the age wise for the alert messages
16. VI(C) . EXPERIMENTAL EVALUATION
Visualizing and Predicting the Heart Prediction Data
We have developed the tool for monitoring and visualizing the prediction data with help of R programming
17. VI . RESULTS AND CONCLUSION
• Twenty one predictor variables from the fifteen patients dataset are used to predict the diagnosis of heart disease.
• The performance of the logistic regression are compared the accuracy obtained between AHA dataset and patients real
time dataset to predict the heart diseases.
• A comparison of the area under the ROC and the accuracy of this model predictions shows and performs with 0.97 % of
accuracy prediction.