We are happy to announce our new line of Yotpo Engineering meetups!
Our first meetup will be held by our Yotpo Data Group.
In this meetup, we will present cutting-edge technologies in the fields of data science & big data development, and describe how we used them to develop Yotpo Insights - our newest engine for extracting and analyzing opinions from customer reviews.
We are also honored to host a talk by Yoel Zeldes from Taboola’s Algorithms Group, who will present some of Taboola’s most recent efforts in improving their content recommendation engine.
Please note: the lectures will be given in Hebrew.
18:00 - 18:30: Gathering
18:30 - 19:00: Talk #1: Graph Processing at Scale using Spark & GraphFrames
19:00 - 19:15: Talk #2: Metorikku - Make ETL Simple Again
19:15 - 19:30: Break
19:30 - 20:00: Talk #3: Computer Vision Models for Recommender Systems
20:00 and on: Mingling
18:30-19:00 - Graph Processing at Scale using Spark & GraphFrames
At Yotpo, we developed an Insights engine which automatically extracts customer topics and opinions from unstructured reviews. A key step in developing our Insights engine was to group related topics (i.e., words or phrases to which an opinion refers) into topic clusters at scale, as the number of initial topics can be extremely large. In this talk, I will introduce GraphFrames - a distributed graph processing library for Apache Spark, describe how and why we decided to use GraphFrames, and our key takeaways from that experience.
Big Data Developer at Yotpo
19:00-19:15 - Metorikku - Make ETL Simple Again
Metorikku is an open-source library that simplifies the writing and execution of ETLs on top of Apache Spark. Using the Metorikku library, users can create complex ETL jobs by only relying on simple configuration files and SQL commands. Moreover, the library includes an infrastructure for creating unit tests. In this talk, we will show how we leveraged Metorikku at Yotpo and made spark accessible to our entire R&D.
Big Data Team Leader at Yotpo
19:30-20:00 - Computer Vision Models for Recommender Systems
Taboola’s content discovery platform leverages computational models to match content to users who are likely to engage with it. Their content recommendation algorithm relies on a short description text and an accompanying image and integrates deep learning techniques from the field of NLP and computer vision. In this short talk, Yoel will describe how Taboola have incorporated vision models into their recommendation system, outline the challenges faced in using transfer learning to predict click-through rate and review the technical aspects involved in incorporating these models into production.
Algorithms Engineer at Taboola