Text analysis in the voluntary sector

Chart from Google n-grams showing results for data and cheese since 1800. Data overtakes cheese in 1970ish

Last week at the Data 4 Good event in Birmingham I gave a talk on the opportunities and challenges that text analysis offers to the voluntary sector. It was a short lightning talk so I was only able to give a quick overview of the possibilities, and it's a big area so there's lots more to explore.

You can find the slides on Google drive.

In the last slide I suggest some tools you can use to take your data further with text analysis. They are:

  • DataBasic – SameDiff - a simple tool for comparing two sets of text
  • DataBasic – word counter - create word clouds and look in more detail at the frequency of words in a piece of text.
  • Reuters OpenCalais - a proprietary tool for getting started with entity recognition. I'd be interested to know if there are any similar open-source projects - one I've seen is spacy but it requires some programming knowledge to use.
  • OpenRefine - a useful data wrangling tool with lots of useful features for working with text.