Getting Back into Text Analytics

In my current paper project (which will hopefully be the last paper of my dissertation) I am diving into text analytics. I am planning to analyze the strategic priorities and digitalization activities from annual reports of different companies over the course of 10 years. However, since it has been over 10 years since I studied information retrieval (back then at the University of Hildesheim) and I am not up to date with the latest tools and techniques, which takes a bit of work to catch up.

It is helpful that we live in times where information is ubiquitous and besides researching tools and articles you can also take a complete course on text mining and analytics on Coursera (or any other course platform), which is probably helpful to repeat the basics. I will see how that goes and if I am disciplined enough to stick with the course program and complete it.

And then another event occurred last week that demonstrated how useful text analytics skills might be in the future: Donald Trump was elected president of the United States. This came as a surprise to many, since the official polls had predicted a clear win for Hillary Clinton. Many articles are now discussing the reasons for this – why typical surveys and polls no longer appear to work. However, at the same time the artificial intelligence Mog-IA correctly predicted Donald Trump as winner, just like every winner of the U.S. presidential elections since 2004. It is not exactly said, how Mog-IA works, but it analyzes mentions in social media and combines that with sentiment analysis. By machine-learning the system is constantly improving. The fact that social media analysis is well suited to predict the outcomes of electoral campaigns has been researched in a couple of studies: