This our data analysis project for the Caterpillar competition.
*************************************** HOW TO ***************************************************** IMPORTANT: First click link below that includes a brief summary of our project.
https://www.youtube.com/watch?v=sIlNIVXpIns
1) To run just run the dashboard program and click link displayed in output to reach localhost server or run individual program files alone and follow instructions through scanner.
*************************************** Details *****************************************************
2) Dashboard_main file is a visual interface displaying the various analysis we conducted on data for Machine_Z while Dashboard2 holds samples of our graphs since running the graph programs on main dashboard would time out localhost sometimes.
a) We have a basic outlier detection on the data by comparing extreme z-scores to a linear regression normalized base set. This is visualized as a scatter plot with blown up outliers.
b) We also have a line graph displaying values over time clustering likely anomaly points within valid range values.
c) Expandability included to also compare multiple machine stats together given data sets.
3) Dashboard2 shows pictures of the actual graphs we did on small data sets from the overall.
4) Code for the actual analysis below. Graphs not shown on main dashboard since it took too long to process the graphs for whole data set on our computers.
Log in or sign up for Devpost to join the conversation.