ML/AI becomes ubiquitous in our daily life - Five main trends at Big Data Tech Warsaw 2021 - Part 2/5.

Each day this week I'll share one of the five Big Data trends that will be covered in detail by numerous presentations at the forthcoming edition of Big Data Tech Warsaw (February 25-26th, 2021).

Trend 2.

ML/AI becomes ubiquitous in our daily life

While the adoption of Machine Learning, Data Science, and AI algorithms and techniques is nothing new, it always required a lot of work, skills, and time. One of the examples that I love to mention is Discover Weekly implemented by Spotify. I attach slides created by my ex-colleagues at Spotify that describe how Discover Weekly came to be, highlight technical challenges, data-driven development, and the ML models used to power their recommendations engine. It was not done overnight. The amount of work to integrate all necessary (open-source) technologies, built scalable architecture, implement smart algorithms, and monitor it was very huge. At least 5 years ago.

Today companies significantly increase their productivity when working on new ML models because they build their dedicated ML platforms and MLOps toolkits. Often, they also switch to the public cloud to take advantage of ready to use libraries and hardware to make their job even easier. As a result, they are able to experiment, train, and deploy more models to production faster and cheaper.

We clearly more ML models in our daily life these days.

ML/AI-related presentations at Big Data Tech Warsaw

During the Big Data Tech Warsaw conference, we will have many presentations that (a) describe use-cases, algorithms, and techniques that solve real-world business problems with ML/AI and (b) share their lessons learned that come from working with ML, Data Science, and advanced analytics. Here I highlight a few examples:

• Mikio Braun (ex-Zalando) will share his lessons on building large-scale production recommender systems. In particular, he will explain how to bridge the gap from the raw mathematical models and algorithms to robust and scalable software systems. In other words, he will explore how theory and practice come together.
• Boxun Zhang (ex-Spotify, currently at Unity) will give a similar presentation, but he will mainly focus on real-time aspects of large-scale, real-time ML systems. His experience comes from implementing a real-time ads bidding system that processes tens of thousands of ad auction requests per second. Boxun will also share several generalizable lessons that make ML systems performant from an ML perspective and scalable from an engineering perspective. It's worth noting that Boxun spoke at our conference a few years ago and he gave one of the best-rated presentations (see slides below). We are happy to have him back with new great Data Science content!

• On the other hand, my colleagues at GetInData will share their year-long journey in developing Kcell (a large Kazach telecom’s) big data analytics platform and building data-driven solutions on top of it that help to reduce costs, improve the quality of the services and understand users need better.
• The very popular use-case for ML is prediction, forecasting, and anomaly detection. We will learn about a near real-time ML model built by Ericsson for predicting telecom systems degradation and outage based on historical fault & performance data. It helps operations teams to move from reactive to proactive monitoring what reduced man-hours spent by support engineers on fixing issues by 30-40%. It also improved the user experience in pre-paid calls and made customer retention higher. But this will not be the only presentation about predictions and forecasting. Peltarion (a Swedish company that specializes in AI) will describe their state-of-the-art weather forecasting AI service. Sotrender (a Polish company that analyses data from social media) will explain their ML model used for predicting the effectiveness of marketing campaigns on the Facebook platform.
• We will also have presentations that use data, science & technology to generate insights for search and recommendation systems in an e-commerce platform (Etsy), built content personalization systems in e-commerce (eBay), run A/B experiments for growth (Atlassian), analyze geophysical data from ground-penetrating radars using deep-learning techniques (SGPR.TECH), and more.

What's next?

On Wednesday, I will share the third trend. Please stay tuned!

In the meantime, I encourage you to check our agenda and register before January 15th to take advantage of New Year Promotion (link).

As you might expect, this year the conference will be organized in form of an online interactive. Please check my recent blog post that explains how COVID-19 changes Big Data Tech Warsaw 2021 but makes it greater at the same time.

If you like this post, please give us a like, or share it or leave a comment. Thanks!