Analysing a TwitterBot using TextBlob, NLTK and Python

@MagicRealismBot is a Twitter bot that randomly generates succinct story synopses that it can tweet. In this talk we will show how we can use TextBlob (and NLTK) to analyse the stories generated by @MagicRealismBot and extract the underlying patterns used to generate the tweets. We will show how TextBlob can be used to analyses the tweets (for example, by sentence extraction and part-of-speech tagging) allowing us to 'reverse engineer' the underlying patterns of the Twitter bot. We will also highlight the variety of other analyses that TextBlob can perform and touch on the state of natural language processing in Python, and elsewhere.

Presented by

Lachlan Blackhall

Lachlan is the Co-founder and CTO of Reposit Power (, an Australian technology company that develops hardware and software to manage, control, optimise and aggregate energy storage in the electricity grid. Lachlan is an experienced Python developer having developed and implemented Python-based software for organisations in the energy, telecommunications, consumer goods, and government sectors. Lachlan holds a PhD in applied mathematics and network theory and has a broad professional interest in control theory, artificial intelligence, and cognitive computing.