Pratham Sahay's Projects
USING THE LOGISTIC REGRESSION ACHIEVED 100% ACCURACY WITH THE TEST DATA AND NORMALIZED OR STANDARDIZED DATA. I HAVE ALSO VISUALIZED DATA TO UNDERSTAND DEPENDENCIES PRECISELY AND CLEAN THE DATA APPROPRIATELY.
Different Models are created for the Breast Cancer Diagnosis on the basis of the Dataset provided by WISCONSIN, applied different feature scaling and then came up with the model with the feature scaling which diagnosis highly precisely.
This is the directory for the diabetic retinopathy detection built under the artificial intelligence project.
I and my teammates have built a web app which can detect fake news to very good accuracy
Breast Cancer dataset is Visualized and Support Vector Machine Model is deployed and the accuracy matrix is improved by using the Grid Search CV to set the right parameter for a better score. Prediction Accuracy is 94%
The Repository contains solution to all the questions of HackerRank. Proper and optimized solution in Python with all test cases passed.
The Repository Contains the Code to convert xml file to CSV and upload it to the S3 Bucket
IRIS dataset is Visualized and Support Vector Machine Model is deployed and the accuracy matrix is improved by using the Grid Search CV to set the right parameter for a better score. Prediction Accuracy is 100%
Performed data analysis on classified data along with prediction using KNN and also applied Elbow Method in order to find the Accurate value for K and also applied standardization of Data to get much accurate result
About Performed data analysis on classified data along with prediction using KNN and also applied Elbow Method in order to find the Accurate value for K and also applied standardization of Data to get much accurate result
Kyphosis dataset analyzed and predictions carried out using two Classifier Algorithm called Decision Tree and Random Forest and achieved accuracy of 76% for Decision Tree and an Accuracy of 96% for Random Forest. The Dataset was small hence the predictions can be much more better if the dataset was more
LOAN DATA FROM LENDINGCLUB.COM WHICH CONNECTS THE BORROWERS WHO NEED MONEY TO THE INVESTORS. WE ACHIEVE AN ACCURACY OF 73.7% FOR DECISION TREE MODEL AND A ACCURACY OF 84.5% FOR RANDOM FOREST MODEL FOR PREDICTION IF LOAN PROVIDED WILL THE BORROWER REPAY IT OR NOT DATA IS VISUALIZED AND ASSESED WITH DIFFERENT SORTS OF DATA WE OBSERVE THAT RANDOM FOREST MODEL IS BETTER THEN THE DECISION TREE MODEL AND WE DID NOT APPLY FEATURE SCALING BECAUSE DECISION TREE AND RANDOM FOREST IS INDEPENDENT OF THE FEATURE SCALING HANCE THE RAW DATA CAN BE APPLIED DIRECTLY.
A GUI which helps you scrape the Wikipedia and helps blind people listen to the information and Technology news
The Features of Titanic Dataset is visualized and Predictions are made using the classification algorithm LOGISTIC REGRESSION all the operations were carried out on Google Colab.