# Baby Naming Algorithm (Python)

To help name my coworkers babies, I wrote this script. It works as follows:

1. Popularity: This is done via a beta distribution. An alpha parameter of 1 prioritizes the most common names, and higher numbers produce less common names.
2. Alliteration: Prioritizes first names to match the first letter of the last name.
3. Syllable count: Lowers priority of names in which the first name has the same number of syllables as the last name. It uses the syllables python package.

To run it, you will need the top baby name list for a given year (as a txt file), which you can get from this zip file from the Social Security website.

This is the code:

``````import pandas as pd
import syllables
from scipy.stats import beta

# inputs
gender = 'M'
lastName = 'Matthews'
alphaParam = 1 # commonness(1 = very common, 1.5 = medium common 100 = very uncommon)

# parameters
betaParam = 2 # how much commonness matters (probably dont change)
wgt_common = 1000
wgt_allit = 0

# pull data and filter gender
data = data.loc[data.gender==gender,]

# Compute a commonality score
data['pctle'] = (data['count']/sum(data['count']))[::-1].cumsum()[::-1]
data['beta'] = beta.pdf(abs(data['pctle']-1),alphaParam,betaParam)
data['common'] = data['beta']/sum(data['beta'])

# Compute alliteration score
data['allit'] = data['name'].str[:1] == lastName[:1].upper()

# Compute syllable score
data['sylcount'] = data['name'].apply(syllables.estimate)