Group items tagged

Nice resource from the UK Chartered Institute of Public Relations on use of statistics in PR and some pitfalls to avoid - Stats matters in the professions!!

On Friday, the Australian Taxation Office, as it does every April, released the latest batch of annual taxation statistics. And as ever, the data was used in rather contorted ways to suggest the budget needed to reduce the level of taxation paid by the wealthiest and to make it seem like the richest were the ones doing it tough. The other old chestnut that got a run in the Australian is that more school teachers actually use negative gearing than company executives. Again, is it really a shock that "while 72,000 investors were listed as company executives, 99,000 people claiming rental losses on their tax returns were either teachers, nurses or midwives"? Given there are about 300,000 more people working as teachers, nurses or midwives than there are company executives, does anyone really think that because 27,000 more of them might use negative gearing is proof of anything? The crucial thing is not the total number, but the proportion of teachers and nurses (and any other profession) who use negative gearing.

some examples of how the 3 types of machine learning - supervised, unsupervised, and reinforcement - have already been used for journalistic purposes, and using those to explain what those are along the way. Examples include: supervised learning to investigate doctors and sex abuse; unsurprivsed learning to identify motifs in Wes Anderson films; reinforcement learning to create a rock-paper-scissors that can beat you...