This dashboard provides an updated overview of the Novel Coronavirus (COVID-19 / SARS-CoV-2) epidemic for Bulgaria and its surrounding countries. This dashboard is built with R using the R Makrdown and flexdashboard framework and was adapted from the dashboard of Rami Krispin, courtesy of Antoine Soetewey.The input data for this dashboard is the dataset available from the `{coronavirus}` R package. The raw data is pulled from the Johns Hopkins University Center for Systems Science and Engineering (JHU CCSE) Coronavirus.

Predictive analysis to target employee attrition. Used variety of tree based and ensemble models in H2O, embedded as a notebook in Azure Machine Learning.

Results & Conclusions

Nested Forecasting Approach - Pick the Best

Global Ensemble Approach - Superlearner

English version

This dashboard provides an overview of the Novel Coronavirus (COVID-19 / SARS-CoV-2) epidemic for Bulgaria and its surrounding countries. This dashboard is built with R using the R Makrdown and flexdashboard framework and was adapted from the dashboard of Rami Krispin, courtesy of Antoine Soetewey.The input data for this dashboard is the dataset available from the `{coronavirus}` R package. The raw data is pulled from the Johns Hopkins University Center for Systems Science and Engineering (JHU CCSE) Coronavirus.

The Data Science team has identified 4 customer segments. The 4 were given descriptions based on the customer's top product purchases.

Using XGBOOST to predict the price of 2 new items in existing product gap.

Pricing algorithm to determine a new product price in a category gap. Evaluating five different models to determine the price - lm, glmnet, decision tree, random forest and xgboost.

The data in this analysis comes form the S&P 500 Index. Using a method of UMAP 2D projection with K-Means cluster assignment of Customer-Item matrix to classify companies based on how their stocks trade using their daily stock returns (percentage movement from one day to the next). This analysis can help to determine which companies are similar to each other in various segments of the market.

Sentiment analysis of 3500 COVID-19 related tweets.

UMAP 2D Projection with K-Means Cluster Assignment of Customer-Item Matrix.

This plot demonstrates how calibrating the model to a testing set maps with the test predictions and residuals inside. Calibration is how confidence intervals and accuracy metrics are determined. Calibration Data is simply forecasting predictions and residuals that are calculated from out-of-sample data.

Interactively visualizing Principal Component Analysis with R using the Motor Trend Car Road Tests dataset.