Issue #128

July 4, 2021

Dear friends,
Welcome to a very special 128th issue of the Sunday Briefing.

We continue on our hiatus from blogging, but while we prepare the next post, you can catch up on Network Assortativity and the Configurational Model where we explore Degree-Degree Correlations in the airline network. Over at the V4Sci substack you’ll find Geographical Maps with Cartopy: The US Airline Network where we introduce some of the amazing functionality of Cartopy.

After several months of working behind the scenes, we can finally announce what has been keeping us busy (and away from the usual blogging schedule which we hope to resume soon). We’ve just completed recording and editing two new long form on-demand videos.

1. Natural Language Processing, has almost five and a half hours of new NLP content using NLTK and Keras that will bring you up to speed on state of the art NLP techniques and approaches. And while we’re on the topic of NLP, we would like to remind you that the next NLP webinar, NLP with Deep Learning for Everyone is coming up on November 19. Sign up now so you don’t miss out!

2. Times Series Analysis for Everyone, is six hours of videos covering data pre-processing, visualization, ARIMA, ARCH and Deep Learning models for Time series analysis.

We’re also proud to announce our first webinars of 2022: Graphs and Network Algorithms for Everyone on Jan 12, followed by Why and What If – Causal Analysis for Everyone on Jan 26.

On our regularly scheduled content, we have a deep dive into graph neural networks with Computation graphs and graph computation and Graph Representation Learning: From Simple to Higher-order Structures. We also explore a Visual Git Reference and Cognition Without Computation.

From the halls of academia we have an exploration of how conspiracists exploited COVID-19 science, the effects of remote work on collaboration among information workers and how hierarchical Transformers Are More Efficient Language Models.

Finally, this weeks ‘Data Science Book ’ highlight is “Graph Representation Learning” by W. Hamilton. As always you can find all the previous book recommendations on our website. In the video of the week we have an An Introduction to Graph Neural Networks: Models and Applications.

Data shows that the best way for a newsletter to grow is by word of mouth, so if you think one of your friends or colleagues would enjoy this newsletter, just go ahead and forward this email to them. This will help us spread the word!

Semper discentes,

The D4S Team

This weeks Data Science Book is " Graph Representation Learning " by W. Hamilton. This short (141 pages) book gives a grounded, well-written, and to the point introduction to representation learning for graphs that helps you grasp the fundamental concepts as well understand when to and when not to. The extensive bibliography provides entry points for further study for the motivated reader. The algorithm descriptions are clear and intuitive in a way that will give you a leg up to implement pre-existing algorithm and even develop your own variations.

Graph Representation Learning

1. Computation graphs and graph computation [breandan.net]
2. Surprising Limits Discovered in Quest for Optimal Solutions [quantamagazine.org]
3. Cognition Without Computation [spectrum.ieee.org]
4. Graph Representation Learning: From Simple to Higher-order Structures [opendatascience.com]
5. How to flatten a list in Python [treyhunner.com]
6. A Visual Git Reference [marklodato.github.io]
7. Facebook to Shut Down Use of Facial Recognition Technology [bloomberg.com]

• How conspiracists exploited COVID-19 science (K. H. Jamieson)
• Cryptic transmission of SARS-CoV-2 and the first COVID-19 wave (J. T. Davis, M. Chinazzi, N. Perra, K. Mu, A. Pastore y Piontti, M. Ajelli, N. E. Dean, C. Gioannini, M. Litvinova, S. Merler, L. Rossi, K. Sun, X. Xiong, I. M. Longini Jr, M. E. Halloran, C. Viboud, A. Vespignani)
• The effects of remote work on collaboration among information workers (L. Yang, D. Holtz, S. Jaffe, S. Suri, S. Sinha, J. Weston, C. Joyce, N. Shah, K. Sherman, B. Hecht, J. Teevan)
• Two provably consistent divide-and-conquer clustering algorithms for large networks (S. S. Mukherjee, P. Sarkar, P. J. Bickel)
• Characterizing partisan political narrative frameworks about COVID-19 on Twitter (E. Jing, Y.-Y. Ahn)
• Hierarchical Transformers Are More Efficient Language Models (P. Nawrot, S. Tworkowski, M. Tyrolski, Ł. Kaiser, Y. Wu, C. Szegedy, H. Michalewski)
• Identification and prioritization of urban traffic bottlenecks (N. Serok, S. Havlin, E. B. Lieberthal)
• Community structure detection algorithm based on link prediction (G. Dai, Q. Wang, B. Xu, L. Sun)

An Introduction to Graph Neural Networks: Models and Applications

July 4, 2021

​