ProjectS

Coronavirus DASHBOArd

Obtained worldwide coronavirus (Covid19) data from John Hopkins University GitHub repository. Data requires to be pre-processed and needs many transformations so that it is ready for visualization in Power BI.

Pre-processed the data using Python programming language and Jupyter Notebook as an IDE. Finally transferred the processed data into Power BI and started visualizing it using many different different types of graphs and charts. And also forecasted it to see the trends and the upcoming number of Covid19 positive cases.

Project data and code:- click here
IDE used:- Jupyter Notebook and MS Power BI

Diabetes prediction

In this particular project, I have used data from Kaggle of diabetes patients. In this dataset, I have 2000 readings and 9 columns. First I have divided it into dependent and independent variables. At initially I used the Logistic Regression model, and I got 78.2% accuracy for the training set. As I started improving my model by using the Logit function from the statsmodels module with all variables (i.e. with 8 variables) I got 77.4% accuracy for testing the dataset. Again tried to improve the model by using the backward elimination method. Eliminated variable "SkinThickness" with 0.5 as a threshold value, as it has the highest 'p' value. I got 77.8% accuracy for the testing dataset. Now plotted the ROC (Receiver Operating Characteristic) curve and tried to improve the roc_auc_score. At last, I was able to achieve 0.75 roc_auc_score and accuracy was 76%, which is good enough. I compromised with the accuracy so that I can get a minimum value of False Negative. False Negative value is the number of patients who have Diabetes but my model is predicting them as non-diabetic patients.

Project data and code:- click here
IDE used for this project:- Spyder

Credit risk

This project is all about predicting if someone is eligible for a loan or not, making this process automatically by training a Logistic Regression model by using some variables that a user gives in the loan form. My main task was to clean the data and pre-process it. So that I can use it in my classifier model and it should be able to give higher accuracy of predictions. In this dataset, I have 614 readings and 13 columns'. First I have divided the dataset into dependent and independent variables. After that encoded dependent variable (i.e. 'y'). After that cleaned up all independent variables and then trained the Logistic Regression model. It has 80% accuracy for the training set only and ROC (Receiver Operating Characteristic) AUC (Area Under the Curve) score was 0.70. The same model gives 82% accuracy when predicted for the testing dataset. Finally, for a more optimal model, I made another classifier and trained with the Logit function from the statsmodels module with a threshold value of 0.5. I got 83.7% accuracy and the ROC AUC score was 0.725.

Project data and code:- click here
IDE used for this project:- Spyder

PyShop

PyShop is an e-commerce website built on the Django framework. I have implemented models for maintaining the product list, its price, and quantity through the admin panel. You don't need to code for every product one can just add, delete, and modify any product using the admin panel. Also, there is the option of add to the cart after that you can purchase the products (these all are future implementations).

Project code:- click here

Location Tracing app

This project is based on android app developments, where I have used Android Studio, Firebase Cloud Messaging Service for managing notifications in the app, and Google Maps API. I have learned geo-fence creations (circular and polygon) and tried to filter out the points whether they're inside or outside of the geo-fence based on their positions compared to geo-fence. I Learned how to integrate Firebase with Android Apps and create your custom notifications in the FCM service.

Project code:- click here

Stay tuned for more upcoming projects.

Page updated

Google Sites

Report abuse