BEGIN:VCALENDAR VERSION:2.0 PRODID:-//hacksw/handcal//NONSGML v1.0//EN METHOD:PUBLISH BEGIN:VEVENT DTSTAMP:20240329T121057Z DESCRIPTION:Click for Latest Location Information: http://edw2020.dataversi ty.net/sessionPop.cfm?confid=128&proposalid=11700\n
FINRA was an early ad opter of cloud computing and machine learning using big data sets. FINRA pr ocesses up to 135 billion market events a day. How much is 135 billion, you may wonder? If one were to save $10,000 a day it would take 37,000 years t o reach $135 billion! Human analysis and traditional software development p rocesses cannot meet this challenge, but machine learning can.
\n< p>We’ve made great strides in this area and would like to share our e xperiences with you! This hands-on workshop will review case studies from r eal-world problems solved by FINRA technologists, and will demonstrate how to apply the best practices we’ve established to common, real-wo rld data sets. You do not need to have experience with cloud or ML to find this session useful. \nAttendees to this three-hour workshop w ill learn how to:
\n\n Design a data infrastructure to support an effective machi ne learning platform: what’s a data lake, why is it important, and ho w do I find things in it?\n Choose an algorithm: an explanation of basic algorithms, w hat makes an algorithm suitable for a data set, and how you apply them to d ata\n Data labeling: Harness your company’s intellectual p roperty by labeling your data, and how labeling can improve ML model outcom es\n Ensure data quality: quality is not about looking for need les in haystacks, it’s knowing what besides needles are hidden in the re and how to find them\n Operational Excellence: \n \n Advanced data quality checks for large complex big data workloads.\n How Machine Learning can help with data anomaly predictions with big data\n \n \n\n\n DTSTART:20200322T143000 SUMMARY:W7: Taming the Data Tsunami – Harnessing Actionable Intelligence wi th Machine Learning DTEND:20200322T174459 LOCATION: See Description END:VEVENT END:VCALENDAR