Last year my wife and I had a child, Garen Jay Marrs. Finding a name for him was an interesting task due to my wife having difficulties in picking names. Our first pet together, a cat, did not have a name for around 2 months.
I spent quite a bit of time researching different apps available, but they did not seem to do what I wanted. Generally the apps only offered going through a list of names and liking or disliking them. As a data scientist, I was hoping to find an app that would suggest names as you provided feedback on which names you like and dislike. None of the apps that I reviewed had this capability so I decided to create my own.
This post is the starting point of a short blog series that describes how I created my ideal baby naming application. It will cover both high and low level details of how it works and my reasoning behind the functionality. At the end of this blog series, the application will be freely available.
What Is In A Name?¶
When choosing a name, most of us try to sound out the name and just create a list of what we like. However, when you look at it from an analytical eye, you see patterns in the names. Some features that my wife and I found important when choosing a name for our son included:
- The letter it starts with
- 2 - 3 letters that the name starts with
- 2 - 3 letters that the name ends with
- What it sounds like when pronounced
- Percentage of the population that have the name as male or female
The dataset that I used is from the Social Security Administration. They provide various statistics (percent male and female) over a period of time for names. This was a great set of names consisting of around 10,000 names. Unfortunately, it does not include a good mix of various ethnicities; Arabic, Chinese and Indian to name a few. For our case this was not an issue.
You can learn more about the data offered here: