Neo4j Budapest Meetup

Conference

Thursday, February 13, 2020, 6:30 pm – 9:30 pm

Details

We are back to 2020 with three very interesting talks about Neo4j and network science.

DANIEL DARABOS, Senior Software Engineer at Lynx Analytics

Intro to Graph Neural Networks

Neural networks are the best known solution for a wide variety of problems on tabular data, images, sounds, natural language text, and even games. Applying them to each different form of input comes with its own challenges. This talk is about applying neural networks to graphs.
I will start with a detailed explanation of the most commonly used architecture, the graph convolutional neural network. I will briefly mention a few variations and share some real world applications. I will end with a practical walkthrough so you can do this on your own graph.

🔗 LinkedIn: https://www.linkedin.com/in/darabos/

JACK WAUDBY, PhD Student at Newcastle University

Reciprocal Consistency in Distributed Graph Databases

This talk will discuss data consistency in a distributed, partitioned graph database. Specifically, it will focus on reciprocal consistency, which if not maintained can lead to irreversible data corruption that can spread at alarming rates. A lightweight mechanism we have developed for preserving reciprocal consistency will be presented and discussed.

🔗 LinkedIn: https://www.linkedin.com/in/jack-waudby-b780b1153/

MATTEO NERI, PHD Candidate at CEU, Network and Data Scientist at FNA

Communication Networks in Organizations: How to Benefit from Your Own Data

What would happen if all the interactions among the individual making part of the same company were stored and accessible to be analyzed? In this age where everything is digital companies with a remote culture are arising and with them the possibility to study their evolution and their composition. Is it possible to infer the structure of the company just looking at the pattern of interaction between people? How much can we say about the efficiency of the communication inter and intra team? Is it possible to predict the formation of a team even before the team exists?
In this talk, I will present a use case focused on a UK based company.

🔗 LinkedIn: https://www.linkedin.com/in/matteo-neri-70a3b898/

Schedule

6:30pm - The space is open
6:45pm - Welcome speech
7:00pm - First talk
7:30pm - Second talk
8:00pm - Third talk
around 8:30pm - We're moving to a nearby bar, where we can eat and drink. This time there is no food at the meetup location.

Thursday, February 13, 2020 - 5:30PM - Thursday, February 13, 2020 - 8:30PM Add to Calendar 2020-02-13 17:30:00 2020-02-13 20:30:00 Neo4j Budapest Meetup Details We are back to 2020 with three very interesting talks about Neo4j and network science. DANIEL DARABOS, Senior Software Engineer at Lynx Analytics Intro to Graph Neural Networks Neural networks are the best known solution for a wide variety of problems on tabular data, images, sounds, natural language text, and even games. Applying them to each different form of input comes with its own challenges. This talk is about applying neural networks to graphs. I will start with a detailed explanation of the most commonly used architecture, the graph convolutional neural network. I will briefly mention a few variations and share some real world applications. I will end with a practical walkthrough so you can do this on your own graph. 🔗 LinkedIn: https://www.linkedin.com/in/darabos/ JACK WAUDBY, PhD Student at Newcastle University Reciprocal Consistency in Distributed Graph Databases This talk will discuss data consistency in a distributed, partitioned graph database. Specifically, it will focus on reciprocal consistency, which if not maintained can lead to irreversible data corruption that can spread at alarming rates. A lightweight mechanism we have developed for preserving reciprocal consistency will be presented and discussed. 🔗 LinkedIn: https://www.linkedin.com/in/jack-waudby-b780b1153/ MATTEO NERI, PHD Candidate at CEU, Network and Data Scientist at FNA Communication Networks in Organizations: How to Benefit from Your Own Data What would happen if all the interactions among the individual making part of the same company were stored and accessible to be analyzed? In this age where everything is digital companies with a remote culture are arising and with them the possibility to study their evolution and their composition. Is it possible to infer the structure of the company just looking at the pattern of interaction between people? How much can we say about the efficiency of the communication inter and intra team? Is it possible to predict the formation of a team even before the team exists? In this talk, I will present a use case focused on a UK based company. 🔗 LinkedIn: https://www.linkedin.com/in/matteo-neri-70a3b898/ Schedule 6:30pm - The space is open 6:45pm - Welcome speech 7:00pm - First talk 7:30pm - Second talk 8:00pm - Third talk around 8:30pm - We're moving to a nearby bar, where we can eat and drink. This time there is no food at the meetup location. UTC public

Thursday

February

2020

6:30PM

RSVP not required

Budapest Site

Nador u. 15

Room : 106

Host

Department of Network and Data Science

Audience

Public

Please, be informed that video/photo recording might take place at the event and the edited version of the video material might be published to communicate or promote CEU PU’s activities. Please, find our Privacy Notice here.