Space Object Interactive Dashboard

Motivation

This visualization tool is a solution to the problem scientists have when trying to understand and use space data for research. The Sloan Digital Sky Survey is a giant database containing tens of thousands of data points about space objects with many numerical characteristics. Our tool will allow users to easily view specific points in space that are contained in the survey that fit numerical characteristics they want these points to have.

This is a very useful tool for anyone who needs easy access to large amounts of data to continue research and know where in space certain objects with specific qualities are. While having access to space data publicly online is extremely convenient, making sense of it just by reading through rows may not be especially helpful for someone performing research. It is important that our tool allows the user to not only clarify questions they have about the data but also be able to visualize what those answers are.

Specifically, the tool will accomplish a few specific tasks. It will allow the user to study celestial objects in a specific coordinate plane of the celestial sphere. The tool allows this by letting the user display these objects of a desired class in that plane, then more specifically allowing for viewing of redshift for a group of objects and where those objects are in terms of the Thuan-Gunn astronomic magnitude system (using band lengths for the u, g, r, i, and z bands).

Background

Data

The tool will visualize the position of celestial objects in a specific cordinate plane of the celestial sphere, allowing users to filter results based on redshift, telescope band (u, g, r i, z), and type of celestial object. The user will see the characteristics of the visualized points by hovering/clicking on them. With visual access, anyone studying these points in space will be able to more accurately and confidently make assumptions about the characteristics of certain areas of space and perform more in-depth research after using the tool.

The dataset was posted on Kaggle by Lennart Grosser, a machine learning engineer from Berlin, Germany. The dataset contains 10,000 rows of observations, each having 17 numerical columns and one class column and is a subset of the SDSS dataset released, that were taken by the Sloan Digital Sky Survey (SDSS). The SDSS provides us with the information needed to learn and discover more about our universe. This exact data table was creating by querying the CasJobs database, joining the photometric and spectral data tables published by the SDSS. The dataset we worked with can be found at this link.

Since this data was collected from observations of our universe, there are not any apparent ethical issues that arise. When it comes to biases in the data, the subset of the SDSS dataset provided on Kaggle may not have been fully representative of the true distribution of observations across the universe. Since we do not know if the subset of data was chosen at random, or handpicked, we do not now how widely distributed the values are for the true scale, but should still be able to provide us with sufficient information.

Visualization

Band Distribution (All 450 Data Points)