Diabetes Prediction System Using Gaussian Algorithm

: Diabetes is among critical diseases and lots of people are suffering from this disease. Age, obesity, lack of exercise, hereditary diabetes, living style, bad diet, high blood pressure, etc. can cause Diabetes. To deal with this problem we have created a system which can predict Diabetes at early stages. We have made it in the form of a mobile app so everyone can use anywhere with just a touch on their mobile. The system is capable of predicting Diabetes by using users data such as age, BMI, insulin, blood pressure, glucose, pregnancy and with the help of our trained model which is using Gaussian algorithm we are predicting if the user has Diabetes or not with high accuracy. This system helps people to save their money and time from the tests that they have to undergo.


I. INTRODUCTION
Healthcare sectors have large volume databases.Such databases may contain structured, semi-structured or unstructured data.Big data analytics is the process which analyses huge data sets and reveals hidden information, hidden patterns to discover knowledge from the given data.Considering the current scenario, in developing countries like India, Diabetic Mellitus (DM) has become a very severe disease.Diabetic Mellitus (DM) is also known as Non-Communicable Disease (NCB).
A technique called, Predictive Analysis, incorporates a variety of machine learning algorithms, data mining techniques and statistical methods that uses current and past data to find knowledge and predict future events.By applying predictive analysis on healthcare data, significant decisions can be taken and predictions can be made.
Predictive analytics can be done using machine learning and regression technique.Predictive analytics aims at diagnosing the disease with best possible accuracy, enhancing patient care, optimizing resources along with improving clinical outcomes.Machine learning is considered to be one of the most important artificial intelligence features supports development of computer systems having the ability to acquire knowledge from past experiences with no need of programming for every case.
Machine learning is considered to be a dire need of today's situation in order to eliminate human efforts by supporting automation with minimum flaws.Existing method for diabetes detection uses lab tests such as fasting blood glucose and oral glucose tolerance.However, this method is time consuming.
There has been drastic increase in rate of people suffering from diabetes since a decade.Current human lifestyle is the main reason behind growth in diabetes.In order to avoid or reduce severity of such impact, there is a need to create a system using machine learning algorithm and data mining techniques which will provide accurate results and reduce human efforts.Our project will be basically focusing on diabetes prediction using Gaussian algorithm.II.LITERATURE SURVEY Diabetes prediction using Machine learning was discussed by Jingyu Xue, FanchoMin[1] as they used few machine learning algorithms to predict diabetes at younger stages.Here they used SVM and Light GBM and they found that the SVM algorithm gives the highest accuracy and hence they proved that machine learning can also be an integral part of medical science.
Mitushi Soni and Dr. Sunita Varma [2] proposed KNN, Random Forest and SVM algorithms and showed results using popular algorithms here the performance of these algorithms was shown.In this she compared the algorithms and showed their accuracy.S.R. Priyanka Shetty and Sujata Joshi [3] used some data mining techniques and showed that whether the user is diabetic or no diabetic the also added few features to show the blood sugar levels and spread some awareness.
Huaping Zhou, Raushan Myrzashova and Rui Zheng [4] used some enhanced deep neural networks to predict diabetes.They used DNN model as their prediction system they used different techniques to increase the accuracy of model like Hyper parameter Tuning N.Sneha and Tarun Gangil [5]also used SVM, Random Forest, KNN, Naive Bayes and showed performance and compared the accuracy.When working with continuous data, an assumption often taken is that the continuous values associated with each class are distributed according to a normal (or Gaussian) distribution.The likelihood of the features is assumed to be-

III. PROPOSED SYSTEM
Sometimes assume variance  is independent of Y (i.e., σi),  or independent of Xi (i.e., σk)  or both (i.e., σ) Gaussian Naive Bayes supports continuous valued features and models each as conforming to a Gaussian (normal) distribution.
An approach to create a simple model is to assume that the data is described by a Gaussian distribution with no covariance (independent dimensions) between dimensions.This model can be fit by simply finding the mean and standard deviation of the points within each label, which is all what is needed to define such a distribution.Above figure shows working of the app.After entering the app, user will be greeted with login screen, if it is an existing user, user can login by entering their credentials.If it is a new user, then user can register by entering their personal details, and then login to the app.After login, they could predict diabetes by tapping the predict button and entering the desired information.Then the result is displayed and then they can logout from the app.

Figure 1 :
Figure 1: Proposed SystemThe data set in this project comes from the open source standard test data set website Kaggle.The objective of the dataset is to diagnostically predict whether or not a patient has diabetes, based on certain diagnostic measurements included in the dataset.The dataset is trained and tested using gaussian algorithm and is stored in a server.When a user enters the Diabetes Prediction System after login id it cannot change it's username and password.After login the user can predict diabetes by entering the appropriate values.In the end the user will get to know if it has diabetes or not.