Bitcoin has gained widespread interest for anonymous payments in recent times, driving the prices for a bitcoin through the roof.
Those prices are generally rising correlated with waves of ransomware, as bitcoin is the most common way of paying those ransoms.
After all, it's an anonymous currency, perfect for anonymous payments, isn't it?
In this talk, we will challenge the assumption of anonymity of the bitcoin network and see what conclusions can be drawn from the publicly available Bitcoin blockchain.
We will discuss options to track bitcoins, evaluate the possibilities to attribute addresses and transactions to users and their interactions with one another.
We will also take a look at how to characterize payments in the blockchain and what conclusions we can draw from that about the usage of Bitcoin as a cryptocurrency.
No previous knowledge about Bitcoin is necessary, a brief review of the properties of the bitcoin blockchain will be included in the talk to set the necessary starting point for everyone interested in the topic.
Scripts used for analysis can be found here: https://github.com/cherti/bitcoin-analysis
Machine Learning has gained traction in recent times for a vast majority of purposes.
This ranges from data analysis over self-driving cars to playing Board-Games like Go, the applicants spread from companies like Google, Amazon and Facebook to insurances and banks who want to use Machine Learning to decide if a person is suited for a loan or to set insurance rates.
In this talk I want to give an introduction to the topic and introduce concepts behind Machine Learning and enable the audience to separate buzzwords from actual content.
No previous knowledge about Machine Learning will be required.
(The focus of this talk will be more on basics and concepts and less on forefront of research, intended to provide a general entry-level overview over the field of Machine Learning.)
This workshop intents do enable the audience to use Python to analyse Data out there. We will analyse an exemplary dataset using the Python Scientific Library as well as Python's statistics toolset Pandas.
Participants should be fluent in Python (version 3, we will not consider Python 2 in this workshop).
Knowledge in the aforementioned libraries is not necessary and will be demonstrated in the workshop.