Fighting Terrorism with Django, Big Data and NLP

Miercuri, 24 February 2016

Oraș: Cluj-Napoca

Tip: Prezentări


For this meet-up, the first one this year, we bring you special guests from ComplyAdvantage (London, UK)

ComplyAdvantage provides anti-money laundering exposure data and compliance technology solutions to help companies control risk and prevent their products being used by criminals, terrorists and money launderers.


  1. "Fighting terrorism with Django"

    Fighting terrorism in the modern world is not only about hearts and minds, guns and violence, but also about cutting off the source of funding that enables such criminals to conduct their activities. This is no mean feat - the amount of money involved is significant, and hence there is a significant incentive for those involved to find ways and means around any systems that are designed to block them. ComplyAdvantage is one of the few companies to offers its customers a real-time transaction monitoring platform. We’ve built the product an existing product on Laravel which provides the powerful, modern framework necessary to build rules and risk based transaction monitoring, and to proactively respond as the terrorists change their strategy. But fighting money laundering means we need to constantly evolve, and we will soon be starting a rewrite, based in Django supported by the powerful mathematical and scientific libraries offered by Python. Luke and Oliver have worked extensively with financial firms from remittance agencies to banks and insurance companies. They have an expert understanding of the challenges involved, and what is necessary to develop a product that can outwit the best minds in corruption.

  2. "Python, Big Data and Natural Language Processing: The tools James Bond ought to be using"

    ComplyAdvantage is building the best database in the world for politically exposed people and adverse media. To do this we need to gather huge amounts of data and use modern technologies to solve some difficult problems so we can detect all the criminals and terrorists trying to do business across the world. In this talk we will discuss the tools we use to build profiles of criminals and terrorists from news sources. We will explore the power of ElasticSearch as an alternative to graph databases and other useful features for text- based systems. Finally, we'll talk about how we put everything together, how we can merge data from different sources and what remains to be done. This talk will give you an overview of the cutting-edge building blocks we need to build the most exciting product in its space today. Q eat your heart out, we make the tools which let James Bond get out of the office at 5 o'clock sharp and drink more martinis.


  • Luke Mawbey studied computational mathematics and electrical engineering at the University of Melbourne before working on 4G/LTE research and development for Analog Devices and NEC, both in Australia and Japan. After a 4-year stint working for the United Nations in Bangladesh, Senegal, Malawi, Kenya, and Lesotho on data and information management, he managed his own start-up to provide quantitative based news analysis which allowed humanitarian organisations to make real-time evidence-based decisions. Luke joined ComplyAdvantage in 2015 primarily on the basis that Romania was one of the few countries he'd neither worked in nor visited, and fighting financial crime with big data sounded like a superhero job and a fun thing to do. Luke manages end-to-end tech projects for ComplyAdvantage and is also the local Hadoop specialist in his spare time.
  • Oliver Butterfield graduated from Durham University in 2004 and joined fledgling server hosting company Gyron Internet where he was instrumental in establishing their first self-managed datacentre in 2009. Following time spent travelling in Africa, he made the move into consultancy. Oliver met Charlie Delingpole (founder of ComplyAdvantage) in early 2014, and Charlie's plans for the as-yet- unnamed anti-money laundering platform coupled with his successful track record in start-ups made the opportunity too hard to refuse. Soon after Oliver became Employee Number 1. Today Oliver lives with his wife and two daughters, and heads up ComplyAdvantage's platform team. He coordinates delivery of the transaction monitoring and screening platforms built on Laravel via a restful API as well as web applications that utilise VueJS and AngularJS.
  • Chris Leow graduated from Cambridge with a Mathematics degree, and cut his teeth as a software developer architecting robust and scalable solutions to business problems for media companies in London. Inspired to take on new challenges and expand his horizons, he co-founded a start-up in mobile geo-located media and social media. Chris joined ComplyAdvantage in 2014 and heads the data team, responsible for building the best database of sanctions, politically exposed people and adverse media mentions in the world.
  • Dr. Thiago Galery has a PhD in linguistics at University College London on the relationships between linguistic structure and contextual information. During that time, he worked with sentiment analysis for IBM. After graduating, Thiago worked at a number of start-ups in the UK, on topics including semantic recommendation, text classification, news tracking, and unsupervised semantic modelling. He was a tutor for Google Summer of Code in 2015 and currently works on the multi lingual NLP pipeline at Comply Advantage and is enthusiastic about using NLP to thwart the financing of criminal activity.
  • Kensuke Muraki is a Natural Language Processing (NLP) and Machine Learning (ML) expert at ComplyAdvantage. He started his career in Japan as a software engineer, working for developing a job recommendation system. He came to London to pursue his master's degree in Machine Learning at University College London, focusing on Statistical Natural Language Processing in his dissertation. After working in several start-ups as a machine learning engineer, he joined ComplyAdvantage in April 2015. He currently focuses on automated data extraction from unstructured natural language texts using NLP and ML. He is interested in technologies such as named entity recognition, text classification, and deep neural networks as well as applying NLP technologies into non-English languages.


18:45 Arrival time. Come in, get a ticket and your name tag. Have some snacks, pour yourself a cup of coffee/tea and meet us!
19:00 Introductions
19:05 Django/Platform talk
19:35 Let's take a break and network next to some pizza and refreshments!
19:50 Python/Data talk
20:20 Closing remarks

This event is organized with: