A journalist friend uses a combination of rhyming words and common phrases as an inspiration source when writing headlines, however, they do this by using a handful of different websites.
Let's do something about that...
Of the three datasources I looked at, I wound up using two:
- Datamuse: A great source for word associations
- The free dictionary: A collection of common phrases
(The other if you want to check it out is Stands4.com which has a large selection of different APIs for word asscociation)
While Datamuse has a REST API, the Free Dictionary didn't; in order to use this a combination of a scraper to get the page content and (as the page structure was consistent) a set of string queries to isoloate the content I required.
While, for the sake of this POC challenge this will suffice, a better API or a cleaner solution to extracting this information is something to come back to in the future.
- Get a search term
- Use the searched word to get a list of rhyming words and related phrases
- Listen for click events on the rhyming words and swap those into the phrases.
While experimenting with the APIs I went down a rabbit hole of word association; this can be seen on the /associations page.
It takes the searched word(s) and uses them to find a list of related words including rhyme, synonyms, antonyms, found before/after, and many others.
The additioanl 'chaining' of related words takes this a step further for example, adjectives for ocean but only when related to the topic of temperature. The Datamuse api docs has more information on what is available.
Ultimately this was not as useful to the initial direction so this remain unfinished for now
This project is built with Next.js and on the the start template from Learn Next.js.