Sections
Portfolio checklist
Here’s a summary of all the tasks and activities you should have added to your portfolio as you worked through this module:
-
Treating text as data: Repeat the searches you used to create your QueryPic chart in the ngram viewer. Try changing the corpus used and the capitalisation of your terms. In a minimum of 100 words describe the differences you observe.
-
Counting words: Save the answers to the questions above in your portfolio.
-
Comparing texts: Grab a screenshot of your streamgraph and save it and the url to your portfolio.
-
Overviewing with overview: Look at the two examples above carefully and think about how TF-IDF values are calculated. Can you find any interesting patterns in the speeches? Write a minimum of 100 words on what you notice. Then look at the Hansard word clouds – a lot of the high-value words seem unfamiliar. Why do you think that is? You might notice ‘Corinella’ in the first cloud – trying searching for it in Historic Hansard. Can you work out why it has a high TF-IDF value in the period 1901-1910? Explain your theory in a minimum of 100 words.
-
Looking critically: Choose two of the visualisations above and in a minimum of 400 words try to answer the questions above. Be thoughtful and critical – question what you see.
-
Change over time: So what did happen in May 2011? There’s enough information in the chart for you to present a hypothesis. Write a few sentences in your portfolio.
-
Change over time: And now a challenge – without any more assistance, can you create the same sort of stacked bar chart in Plot.ly? Test your new dataviz skills! Once you’ve recreated the chart in Plot.ly, save it, and then click of the Share button to get a shareable link. Add a screenshot of you chart and the link to your portfolio.