Sponsor Challenges
Description
Spatial data analysis to support the humanitarian mission
Input Data
We take a region in an under-mapped country (like Tanzania), and provide:
- Population density dataset (HRSL, recently published by Facebook)
- Existing OpenStreetMap roads
- Deep Neural Network road predictions for the same area (100Mb - 1Gb of raster data
Challenge Questions
- Warm Up: How many people live in the unmapped area (further than 100 meters from the closest road)?
- Main Challenge: With some basic assumptions about the speed limits for the known roads, ML-predicted roads and unmapped areas, estimate the distance and approximate travel time the between the two points.
Prizes
- 1st Place: Oculus Quest VR Gaming Headset (Winners)
Plus, an opportunity of working towards the next “State of the Map” Conference
Description
As with all mechanical equipment, things break and when things break money is lost in the form of repairs and lost oil production. When costs go up cash goes down, but how can we predict when equipment will fail and use this information to drive down our costs?
Input Data
A data set will be provided that has documented failure events that occurred on surface equipment and down-hole equipment. For each failure event, data has been collected from over 107 sensors that collect a variety of physical information both on the surface and below the ground.
DatasetChallenge Questions
Predict surface and down-hole failures using the data set provided.
Prizes
Description
Best project that helps the community and/or environment!
Input Data
None
Challenge Questions
Project should show a contribution to the community or the evironment or an insight into bettering them.
Prizes
- 1st Place: Amazon Echo (Winners)
Description
One thing that brings everyone together is food, and in Texas that means burritos and tacos. A list of 19,439 restaurants and similar businesses with menu items containing burritos and tacos from across the US has been compiled for users to mine the data and derive insights.
Input Data
The dataset includes the category, cuisine, restaurant information, and more for a menu item. Each row corresponds to a single menu item from the restaurant, and the entirety of each restaurant's menu is not listed. Only burrito or taco items are listed.
DatasetChallenge Questions
The final product of your efforts should include a visualization of your output, with supporting documentation detailing the modeling and analysis performed.
Prizes
- 1st Place: Amazon Echo (Winners)
Description
By developing a better understanding of the consumer and the marketplace than their peers, fuel retailers can deliver more appealing products and services to their customers across multiple categories. The key to improving the revenue potential of each customer lies in understanding as much as possible about buyers’ needs, preferences and purchasing behaviors
Input Data
Please find external data from Kaggle / Google for this challenge.
Challenge Questions
- When people visit a fuel station, how many purchase Non-Fuel Retail (NFR) items from the store?
- Can we predict which items will be purchased more in the future, taking into account weather, season, events, and prior purchases?
Prizes
- 1st Place: Shell Goodie Bags & 50$ Shell gift-card (Winners)
Description
Black Friday is coming, and Walmart has a lot of great deals for its customers. As a smart shopper, you will use computer to make a plan to visit the in-store deals as quick as possible.
Input Data
- Store Map
- Location of deals
- Traffic heatmap
Challenge Questions
- Make a model to generate the optimal travel plan
- Stretch goal: Create a simulation environment to visualize the model
- Stretch goal: Help store manager by providing a model to optimize the deals locations
