Methods

We’re using data science to save lives. Want to learn more (and maybe nerd out a little bit)? Read on…

Data Science Analysis

  • We’ll use a convolutional neural network to analyze the appearance of the animals. Breed is a very complex issue in animal shelters, but certain animals are very obviously one particular breed or a mix. We can identify this using our model. We can also identify other aspects of the animals’ appearance, like coloration, long or short hair, etc.
  • What about the adoption rate of older/younger dogs? Larger/smaller dogs? The model will analyze all these factors.
  • We will use Natural Language Processing to obtain insights from the descriptions given on adoption websites for each animal. For example, does using certain words increase the likelihood that an animal will be adopted?
  • A number of factors will go into the process of determining which animals to move to which locations, including the distance from one shelter to another, the euthanasia rate, and other factors. Shelters will receive an automated notification every time we have a recommended transport.
  • To measure impact, we will track the animals transported to determine how long it takes for them to be adopted to see if there’s an improvement. Over time, we can also examine shelters’ euthanasia rates to note an improvement.

If the animal is transported to a shelter…

….the animal and its friends at the shelter will receive new, handmade blankets courtesy of our friends at Comfort for Critters! Whoever adopts the pet gets to keep the blanket.