us-state-abbr-anagrams

Using PHP and R to find eight-letter words that are anagrams of four adjacent US state postal abbreviations

The Challenge

I heard a puzzle challenge on the radio to find a common eight-letter word spelled from the US state postal abbreviations of states that can be driven between. For more details, see the This week's challenge section of the January 31st, 2021 Sunday Puzzle page.

It sounded like an interesting problem that could easily be sovled with computer science, so I decided to give it a try.

Preparing Data

First, I pulled up a map of the United States that showed state boundaries. I went state by state, creating a list of adjacent states and making a list of their abbreviations.

The resulting file can be found here.

MySQL Setup

After that, I created a simple MySQL database for my PHP script to reference. The pairings data was loaded into the connections table. I also created a simple states table that contained the results of looking at the distinct values in the origin column of the connections table - both as a way to confirm I had entered pairings for all 48 states, and to create a list for my PHP script to work through. A combinations8 table was created to store the results of my PHP script.

The details of that database can be found here.

PHP Script to Find All Eight-Letter Combinations

I wrote a script in PHP which generated every possible four state combination based on the connections I had documented earlier. Nearly 3000 combinations were generated and written to the combinations8 table.

My script can be found here.

The CSV containing the combinations generated can be found here.

R Script to Find Anagrams

With all possible combinations in hand, I then needed to check them for anagrams that were common English word. I did not find a web API I was satisfied would be up for the task, but did find the wfindr R package that seemed perfect for the job. I deduplicated the data in my combinations8 table and exported the combinations to a CSV file for use with R. That file can be found here.

My R script looped through each of the eight-letter state abbreviation combinations and fed each through the find_word function to find any eight-letter anagrams and saved any words it found to a dataframe which was exported as a CSV once the script finished running.

My R script can be found here.

The resulting CSV of eight-letter state abbreviation strings and the anagrams found within each can be found here.

Results

My scripts found 50 eight-letter anagrams out of the nearly 3000 possible four state combinations, of which only 9 words I would consider "common":

Word	Four Adjacent States
Animator	IA > MO > TN > AR
Condemns	MN > SD > NE > CO ; NM > CO > NE > SD
Diamonds	ND > SD > IA > MO
Dioramas	SD > IA > MO > AR
Flagrant	FL > GA > TN > AR
Moleskin	IL > MO > KS > NE
Nominate	NE > IA > MO > TN
Ornament	NE > MO > TN > AR
Ransomed	AR > MO > NE > SD

Note: The above state combinations found the same words when checked in a different order, which is why the matched_words.csv is so much longer.

The complete list of anagrams can be found here.

ericcawthon / us-state-abbr-anagrams Goto Github PK