Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Real-World Azure Machine Learning at EF Education First by Stefano Tempesta

Real-World Azure Machine Learning at EF Education First by Stefano Tempesta

In this session, Stefano gives an overview about the Power of Azure Machine Learning (ML) and how they use ML at “EF Education First” to drive their business forward.

Speaker: Stefano Tempesta, EF Education First
Stefano is the Vice President of Engineering at EF Education First (http://www.efswiss.ch/de/), the world’s leading private international education company. He is a regular speaker at conferences, including Microsoft Ignite, TechEd, NDC, API World and the European SharePoint Conference. Stefano’s interests span across Cloud, Mobile and IoT applications. You can reach him via his personal web site http://www.tempesta.space/.

Azure Zurich User Group

October 26, 2017

More Decks by Azure Zurich User Group

Other Decks in Technology


  1. TOPICS Online Payment Fraud Detection Intranet Sentiment Analysis Cache Hit

    Ratio  Improved Page Load Time Emergency Response
  2. VISA handles 2000 transactions / sec = >172M tpd PayPal

    handles = >135M tpd In 2015, financial fraud totaled a cost of € 900 million
  3. Traditional Payment Fraud Detection Analyze transactions and human-review suspicious ones

    Use a combination of data, horizon-scanning and “gut-feel” Every attempted purchase that raises an alert is either declined or reviewed ▪ False positive ▪ Fail to predict unware threats ▪ Need to update risk score regularly
  4. ML Payment Fraud Detection Large, historical datasets across many clients

    and industries ▪ Benefit also small companies Self-learning models not determined by a fraud analyst ▪ Update risk score quickly
  5. Sentiment Analysis “Sentiment Analysis is the process of detecting whether

    a piece of writing is positive, negative or neutral” ▪ ML-driven Text Mining ▪ Opinion Polarity
  6. Text Analytics Applications: ▪ Product Reviews ▪ Case / Document

    Classification ▪ Social Media Analytics ▪ Intellectual Property ▪ Plagiarism Check
  7. Text Analytics API Analyze unstructured text for tasks • Language

    detection • Key phrase extraction • Sentiment analysis Returns a numeric score between 0 and 1 ▪ Negative 0 .. 1 Positive sentiment Advanced natural language processing https://azure.microsoft.com/en-us/services/cognitive-services/text-analytics/
  8. 9