Big Data and Machine Learning - London - Meetup #8

Date: 
Tuesday, January 16, 2018 - 18:30
Source: 
Big Data and Machine Learning - London
Attendees: 
200
City: 
London

PLEASE NOTE: Limit of 130 attendees!

Meetup #8
Welcome to Meetup #8, and what we hope will be another interesting evening of presentations and lightning talks. The agenda is listed below, followed by further details about the main presentations and their presenters.

Should you wish to contact me, email me at [masked].

Kindest regards
Mark


Agenda

18:30 Doors open and networking

18:55 Welcome
Mark Whalley
(5 mins)

19:00 Teach an e-commerce search engine a thing or two about how users search: mine query logs with language models and word2vec
Guillaume Allain
(30 mins)

19:30 “That was fun! ...What next?”
Tate Bartlett
(30 mins)

20:00 The Lab Series
Mark Whalley
(30 mins)

20:30 Networking

21:30 Close

---

Teach an e-commerce search engine a thing or two about how users search: mine query logs with language models and word2vec
Training statistical spellchecker traditionally required an annotated dataset of misspelled sentences and their corrections. Instead, web companies have relied on massive query logs and used iterative procedures. In this talk, we will showcase a simpler approach, complementing Elastic phrase suggester with an approximated error model based on the distributional hypothesis, trained using the word2vec algorithm.

Guillaume Allain
Guillaume has a PhD in applied mathematics and statistics and has learnt software engineering on the job while building recommender and NLP systems in the industry. He is passionate about applying a scientific approach and an agile approach to deploy ML systems in the wild world. He recently joined Argos where he is taking part in the digital transformation of this traditional retailer into a tech driven company.
---

“That was fun! ...What next?”
A recruitment professional’s guide to scaling London’s Machine Learning landscape in 2018.

Tate Bartlett
Tate leads the AI & Machine Learning desk for TEKsystems in London, a new section of our business that he has taken sole responsibility for developing from scratch. He works with senior managers across both the start-up space and with establish conglomerates. Recent projects include founding hires at the likes of Factmata, Scape Technologies & Creative AI (all from Entrepreneur First) as well as senior Machine Learning hires at IHS Markit, Experian & Bloomberg LP.

TEKsystems is the IT focused arm of the Allegis Group, the world’s largest privately owned staffing firm. We became TEKsystems (formerly Aston Carter) after an acquisition in 2011 on the back of having a presence in London since founding in 1997.

---

The Lab Series
This presentation continues to discuss a project that has been developing through 2017 – tracking commercial aircraft in near real-time using a Raspberry Pi, Kafka and Vertica.

In the previous presentations, we covered some of the many components of this project including, using Automatic Dependent Surveillance Broadcast (ADS-B) to track the position, speed and other metrics of commercial aircraft, Raspberry Pi computers to perform data capture and decoding, a custom-built Extract Transform and Load (ETL) function, Apache Kafka, integration between Kafka and Vertica, simple data visualisation using geospatial coordinates and a Google Maps API, time series gap filling and interpolation, outlier detection and sessionization to name but a few!

This presentation will review the progress of the project to-date, and look at some of Vertica’s tools for simplifying data exploration and having a conversation with your data.

Mark Whalley
From the early 1980s, Mark worked with Michael Stonebraker's Ingres RDBMS and then a number of column-store big data analytic technologies. In 2016, he joined HPE Big Data Platform as a Systems Engineer specialising in Vertica and Vertica SQL in Hadoop, and from September 2017 followed Vertica as it moved over to Micro Focus.

Campus London

4-5 Bonhill St, Shoreditch London EC2A 4BX