Multiple industries leverage SafeGraph data to provide value and improve quality in their analytics pipelines (see examples). As a Data as a Service (DaaS) company, they provide high quality and high volume data representing foot traffic at varied points of interest (POI). Their data are rich in their spatial and temporal detail.
Businesses that want to leverage this type of data need to see the data value. They also want to understand your specific skills at wrangling, feature creation, predictive modeling, and app development using Spark. In addition to your data science development skills, they want to understand your ability to generate value from data.
We need a final predictive model accessible by non-technical employees within the designed use case. Carefully documented guides should support this predictive model, data ingestions, data engineering, application development, and process implementation and installation.
We would like you to leverage the SafeGraph data to support the placement of retail brick-and-mortar locations. This is a proof of concept process, so we are open to your team picking the retail space in line with the following requirements.
- As much data as possible is leveraged from the SafeGraph source.
- The retail placement doesn't focus on a specific brand.
- The output is either a probability or a count.
- Decisions for placement can be made at a relatively refined spatial footprint (not predicting at the state level, for example).
- A proposal of how to use the OpenWeather API with our tool.
- Your submission will receive added weight if you exemplify incorporating additional data sets beyond those available from the SafeGraph.
Your proposal should be no more than two pages and should include the following;
- Team member names and majors
- Business needs you will address and why this need is of value
- Explanation of the current SafeGraph data format and some descriptive visualizations
- Additional data that you will leverage beyond that found in the SafeGraph API
- Description of what constitutes a successfully implemented product