Description:DETAILED DESCRIPTION OF THE INVENTION
Collection of Patient’s Data
? The basic module of this project is the data collection of Patients and Guardian’s reviews. This step deals with collecting the right dataset, which is filtered based on a number of different aspects.
? The data in this project is mainly composed of the patient’s details and the symptoms of the disease . The CSV dataset is analyzed and the model with the highest accuracy is chosen and made to work with the data collected to provide predictions of a certain accuracy.
Collection of Hospital’s Data
? This step deals with collecting the right dataset, which is filtered based on a number of different aspects. The data of various Hospitals and various reviews given by patient is being listed here.
? The data in this module mainly consists of the user and guardians review to the Hospitals where in Hospital’s review is being taken into consideration and patient’s and guardians are being guided.
Doctor’s Data
? The fourth step as well as the basic module of this project is the data collection of Patients and Guardian’s reviews. This step deals with collecting the right dataset, which is filtered based on a number of different aspects.
? The data in this project is mainly composed of the patient’s details and the symptoms of the disease . The CSV dataset is analyzed and the model with the highest accuracy is chosen and made to work with the data collected to provide predictions of a certain accuracy.
Prediction of disease
? Here the user actually mentions the type of symptoms he is suffering and on that basis the diseases are being predicted.
? The data in this project is mainly composed of the patient’s details and the symptoms of the disease . The CSV dataset is analyzed and the model with the highest accuracy is chosen and made to work with the data collected to provide predictions of a certain accuracy.
Scoring the Data
? Applying a dataset to a predictive model is known as data scoring. This dataset is processed by Random Forest Algorithm. It is used not only for regression but also for classification.
? The previous module explains how the parameters are optimized in order to predict the rise and fall of the stocks in the stock market.
? A user authentication feature is implemented so that only the authorized users can access the results.
A. User’s Side Enter the symptoms: User enters the symptoms he is having, or the guardian enters the patient’s symptoms. Find the disease with trained data: The disease is being predicted by the structure and also based on the information given by the user. Select Hospital and Doctor: This module helps in selecting the appropriate doctor and hospital according to requirement specified by the user. Display the recommendation details: The doctor is being predicted by the structure based on the information given by the operator. Database: The data given by the user is being stored on the hard disks in the form of the table it is made more efficient.
B. Admin Side Collect the reviews: Admin collects the reviews from patients. Pre -process the review: Admin confirms that the information provided is genuine and correct. Find positive, negative and neutral review: Admin then classifies the user provided review as positive, negative and neutral review. Store the review: Admin then stores the reviews on the table.
Algorithms
KNN Algorithm
1. Start
2. Load the data
3. Initialize k to selected number of neighbors
4. For each sample in data
5. Calculate the distance between the query sample and the current sample from the data
6. Add the distance and the index of the example to an ordered collection
7. Arrange the ordered collection of the distances and the indices from smaller to the largest in ascending order by the distances
8. Choose the first k entries from the sorted collection get the labels of the selected k entries
9. If regression return the mean of K labels
10. If classification, return the mode of K labels
11. Stop
Decision Tree
1. It begins with the original set S as the root node
2. On each iteration of the algorithm, I iterates through the very unused attribute of the set S and calculate the Entropy(H) and Information gain(IG) of this attribute
3. It then selects the attribute which has the smallest entropy or largest information gain
4. The set S is then split by the selected attribute to produce a subset of the data.
5. The algorithm continues to recur on each subset, considering only attributes never sleeted before.
6. On each iteration of the algorithm, I iterates through the very unused attribute of the set S and calculate the Entropy(H) and Information gain(IG) of this attribute
7. It then selects the attribute which has the smallest entropy or largest information gain
8. Stop
Random Forest Algorithm
Step 1: First begin with the choosing of random samples from a given set of data.
Step 2: Next this step builds a decision tree for each model. Then we get the expected result from the decision tree.
Step 3: This step election will be performed for each predicted consequence and the non-predicted outcome.
Step 4: Finally select the highest polling consequence as non-predicted outcome.
EXPERIMENTS AND RESULTS

• Entering the details of the symptoms and submission button page
• The symptom entry page for guardians or patients after entering the symptoms.
• Hospital recommendation page
• The hospital recommendation page to view the list of hospitals
Results
Table 1: The diseases which are being predicted by the application
Symptom Analysis and Disease Prediction, pain in chest is considered.
Test case Test Unit Pass/Fail
Heart Disease General parameters like pain in hand Specific analysis like BP, ECG Pass
Diabetes Analysis in various parameters like pedigree analysis, range Diabetes mellitus(range) Pass
Breast Cancer Chest radiograph is done for analysing the lump size(ranges between 1-5 cm) Pass
General Symptoms The symptoms are analysed, and the prediction of the disease is done accordingly Pass
, Claims:General parameters like pain in hand Specific analysis like BP, ECG

Analysis in various parameters like pedigree analysis, range Diabetes mellitus(range)

Chest radiograph is done for analysing the lump size(ranges between 1-5 cm)

The symptoms are analysed, and the prediction of the disease is done accordingly