DataTalks #5: Advanced Topics in Clustering

https://a248.e.akamai.net/secure.meetupstatic.com/photos/event/7/7/d/b/600_449850683.jpeg

DataTalks @ Gett (https://gett.com/il/)

DataTalks (http://datahack-il.com/) #5: Advanced Topics in Clustering.

Our fifth meetup will be hosted by Gett (https://gett.com/il/), and will focus on clustering.

Language: Hebrew

Location: HaBarzel 19d, Tel Aviv

Schedule:

• 18:00 - 18:15 - Gathering, snacks & mingling

• 18:15 - 18:20 - Opening words

• 18:20 - 18:30 - Introduction:
Boris Korenfeld, VP R&D, Gett - Data Science at Gett

• 18:30 - 19:40 - First talk:
Ilai Falach, StoreSmarts - K-means++: Harder, Better, Faster, Stronger

• 19:40 - 19:50 - A short break

• 19:50 - 20:40 - Second talk:
Nadav Bar, Google - A Practical Intro To Density Based Clustering

==== Talk #1 ===

Speaker: Ilai Falach, StoreSmarts
Title: K-means++: Harder, Better, Faster, Stronger
Abstract: In this talk I will give an overview of center-based clustering methods, starting from the well known k-center and k-means methods. These will give the motivation for the k-means++ method, which extends k-means by making the random initialization of data points more intelligent. We will show guarantees on convergence and approximation of the algorithm, and go through the actual proofs.

About the speaker: Ilai is the CTO and Co-Founder of StoreSmarts, and is finishing an BSc. in Computer Science from Tel Aviv University.

==== Talk #2 ===

Speaker: Nadav Bar, Google
Title: A Practical Intro To Density Based Clustering
Abstract: Although they have received less attention compared to Centroid-based clustering methods, such as k-means, density based clustering methods offer some very appealing features for their users, including the ability to discover the number of clusters automatically, as well as the detection of clusters of different shapes and sizes. In this talk, I will present several density-based clustering methods, starting from the classic DBSCAN method, and moving forward to newer and more advanced methods. As part of the talk, we will walk through each algorithm’s inner workings, and we will also see live code examples for each of the clustering methods.

About the speaker: Nadav currently works as a Software Engineer at Google, and holds an MSc. in Computer Science from Tel Aviv University. In his thesis, he focused on developing a novel approach for density-based clustering, under the supervision of Prof. Daniel Cohen-Or.

-------------------
DataHack (http://datahack-il.com/) is a data-driven community and annual hackathon for data-enthusiast programmers, researchers and designers.

You can also find us on Facebook (https://www.facebook.com/datahackil/) and twitter (http://twitter.com/DataHackIL), and join our monthly newsletter (http://eepurl.com/bH6BoX).