AlgoIL #8 - Everything is a Graph

Don't we all just love graphs theory??
What a pleasure it is to learn new ways in which modeling our data as a graph can help us in our research 💜

Agenda:
18:00 Gathering and snacks
18:30 Opening words
18:40 [English] Hannah Danan Moise - Putting a stop to illicit maritime activity using Graph and Network Analysis
19:25 Short break
19:35 [Hebrew] Ella Bor - Generating knowledge graphs with data
20:20 Time for more snacks and chats

Thanks Windward for hosting this event!
_____________________________________________________________________________________

Hannah Danan Moise - Putting a stop to illicit maritime activity using Graph and Network Analysis

Shipping is complicated. A single ship, on a single voyage has dozens of different stakeholders, including ship operators, cargo owners, marine insurers, governments, and banks.
With 90% of global trade underpinned by shipping, vessels can also be seen as central to the movement of illicit goods. Gaining a deep, accurate understanding of the various players involved is key to identifying sea-based activities that pose a threat, be it to nations, energy traders or anyone else. This makes Graph Analysis an essential tool for the maritime domain.
In this talk, we will discuss how we approached our research, encompassing our understanding of Graph Theory and our reasons for looking at maritime dataset as graphs. We developed various research tools: an investigation application that is now used by our own analysts, as well as a research infrastructure that we built to enable rapid experimentation. Finally, we will present the insights and models that can be extracted from data by looking at it through the lens of Graph Theory, using basic (and more advanced) graph algorithms.

About the speaker:
Hannah Danan Moise is a Data Scientist at Windward. Her recent work focuses on research and implementation of Graph Theory to support decisions in the maritime domain. With an academic background in economics and mathematics, she made “Aliyah” from France in 2016 and started working at Windward as a data analyst. She fell in love with data, and graduated from the Israel Tech Challenge Data Science Fellows program. Like Windward, Hannah combines domain expertise with data sciences and business savvy.

_____________________________________________________________________________________

Ella Bor - Generating knowledge graphs with data

At Twiggle, we learned and understood users' queries and matched them with relevant products. To that end, we built a knowledge graph representing our own ontology of the e-commerce world. Probably not the easiest task, but surely an adventurous and challenging one.
We built a data-driven ontology by applying Node2Vec and clustering methods on query-to-product clicks along with minimal information about these products.
Join me for this talk where we will explore our usage of graphs, learn about Node2Vec, and reminisce on knowledge and experience that lives longer than any product or company.

About the speaker:
With an MSc in Physics from the Weizmann institute, I entered the fascinating world of algorithms. In Twiggle I took my first steps (or huge jumps, actually) in NLP and ML algorithms and learned from the best of the best, all while having the greatest time. When I'm not representing things as a graph, I enjoy traveling abroad - I even visited Alaska and Hawaii!

_____________________________________________________________________________________