Day 27 of #100daysofnetworks

Dense Network Community Exploration pt. 2

I want to write this post tonight so that I’ll have it out of my mind and I can work on other things this weekend. For the last two posts, I’ve been working on a technique to explore different network communities based on density and node size. In other words, I built a tool that allows me to explore network communities based on where I think I’ll find interesting content. I made a few observations over the last three posts. I’m using arXiv data to create author collaborative networks. Here’s my impression, so far.

• Sparser communities are comprised of authors who have collaborated on papers written by a few authors, and large sparse communities thus have more variety in papers than large dense communities.

• Dense communities are comprised of authors who have collaborated on papers written by many authors, and large sparse communities might just contain one or two papers. There are arXiv papers with dozens of authors.

• In between, you get a blend.

Personally, I felt that sparser communities had more interesting papers, and more of them, but that is my opinion. So, for today’s exercise, I explored network communities with a density between 0.1 and 0.3. I’m going to give a tour of those today.

You can read the previous two posts here and here.

As those two posts already describe the how, today will be showcasing results and discovered articles. There will be many articles linked to at the end, for your enjoyment.

Communities and their Papers

I use Natural Language Processing, Social Network Analysis (SNA), and Network Science together, not separately.

As such, you’re going to see two things here:

• The community that wrote the papers (SNA and Network Science)

• The papers they wrote (Language, NLP)

That is how I will show these. Let’s go!

Social Network Analysis: this is a network community. It was identified using the Louvain Method for Community Detection.

Network Science: several of the key nodes are colored differently, by their Page Rank values. You can see that Xun Liu and Stefan are central figures in this ecosystem as outer parts of the network are unreachable except through them. Other important nodes are also colored light blue.

Natural Language Processing: these are the papers that came from this community. This is text data, and could be analyzed using NLP.

['A Deep Neural Model Of Emotion Appraisal',
 'A Humanoid Social Agent Embodying Physical Assistance Enhances Motor Training Experience',
 'A Review of Critical Features and General Issues of Freely Available mHealth Apps For Dietary Assessment',
 'A User-Centred Framework for Explainable Artificial Intelligence in Human-Robot Interaction',
 'Assessing the Contribution of Semantic Congruency to Multisensory Integration and Conflict Resolution',
 'Can AI detect pain and express pain empathy? A review from emotion recognition and a human-centered AI perspective',
 'Focusing and directional beaming effects of airborne sound through a planar lens with zigzag slits',
 'Hierarchical principles of embodied reinforcement learning: A review',
 'Highly efficient anomalous refraction of airborne sound through ultrathin metasurfaces',
 'Improving interactive reinforcement learning: What makes a good teacher?',
 'Incorporating Rivalry in Reinforcement Learning for a Competitive Game',
 'Inference of Affordances and Active Motor Control in Simulated Agents',
 'Intelligent behavior depends on the ecological niche: Scaling up AI to human-like intelligence in socio-cultural environments',
 'Intelligent problem-solving as integrated hierarchical reinforcement learning',
 'Lifelong Learning from Event-based Data',
 'Observation of Acoustic Non-Hermitian Bloch Braids and Associated Topological Phase Transitions',
 'Orbital Deflection of Comets by Directed Energy',
 'Rethinking Continual Learning for Autonomous Agents and Robots',
 'Sequential Attention GAN for Interactive Image Editing']

I’m only going to point out the three distinctions once. Try to learn how they relate. These are the papers this community wrote, and if these papers are of interest to you, then you might want to get to know these people.

While doing this kind of exploration, I was keeping a side text file with articles I wanted to look closer at. At the end of this, I will paste in the articles so you can read them. I’ll show a few more communities, first.

This is an excellent network for demonstration purposes. Can you identify the edge (line) that probably has the highest Edge Betweenness Centrality value? It is the edge that would split the network in two if cut. I think it’s obvious. You will learn to identify these things visually. Serge and Thomas are bridge nodes. They connect two communities together, via a single relationship. One relationship unlocks a world of potential to everyone in this network.

I find this community visually very interesting, the way it is winding and has areas of differing density. If you were to use the Louvain Method one more time on this subgraph, you would be able to identify communities within the community, but you can also see them with your eyes. What kind of papers do these people write?

['A Recipe for Geophysical Exploration of Enceladus',
 'Black hole and neutron star mergers in Galactic Nuclei: the role of triples',
 'Differing Enceladean ocean circulation and ice shell geometries driven by tidal heating in the ice versus the core',
 'Dynamics or Geysers and tracer transport over the south pole of Enceladus',
 'Ocean Worlds Exploration and the Search for Life',
 'Stability of exomoons around the Kepler transiting circumbinary planets',
 'When do star clusters become multiple star systems? II. Toward a half-life formalism with four bodies',
 'When does a star cluster become a multiple star system? I. Lifetimes of equal-mass small-N systems']

Nice. These people would be fun to talk to.

This one is actually fascinating. I don’t know if I’ve ever noticed one like this before. This is almost a star graph, but containing small dense communities connected to one individual in the middle. That one person is very important in this community.

If you want to explore more, you can get the code here. There are thousands of communities to explore in this one dataset.

What Did I Find?

Today, I used the Artificial Life collaboration dataset to build this network. There are about 20,000 arXiv articles in the dataset. Rather than being overwhelmed with 20,000 articles, browsing by communities allows a few things:

• First, I’m not overwhelmed, trying to read 20,000 articles at the same time.

• Second, I’m not so overwhelmed that I just gave up, falling back to raw luck and random internet searches to find things.

• Third, it felt very approachable going slow, looking at the community, admiring the network, and slowly reading through the list that the groups managed to write.

And because I didn’t feel overwhelmed and could explore by ecosystem, I could find useful information. I’ll link to my articles of interest below, for your enjoyment.

That’s All Folks!

That’s all for today! Thanks for reading! If you would like to learn more about networks and network analysis, please buy a copy of my book!

This may get cutoff in email. Please visit on a browser to see the full list.

Articles Found

Category: cs.CL

Title: Find the Conversation Killers: a Predictive Study of Thread-ending Posts
Date Published: 2017-12-22 19:58:01+00:00
URL: http://arxiv.org/pdf/1712.08636v1

Title: A Deep Learning Approach for Multimodal Deception Detection
Date Published: 2018-03-01 12:38:13+00:00
URL: http://arxiv.org/pdf/1803.00344v1

Title: Causal Explanation Analysis on Social Media
Date Published: 2018-09-04 19:06:34+00:00
URL: http://arxiv.org/pdf/1809.01202v2

Title: Fine-Grained Emotion Classification of Chinese Microblogs Based on Graph Convolution Networks
Date Published: 2019-12-05 12:56:28+00:00
URL: http://arxiv.org/pdf/1912.02545v1

Title: Natural Backdoor Attack on Text Data
Date Published: 2020-06-29 16:40:14+00:00
URL: http://arxiv.org/pdf/2006.16176v4

Title: GraphTMT: Unsupervised Graph-based Topic Modeling from Video Transcripts
Date Published: 2021-05-04 12:48:17+00:00
URL: http://arxiv.org/pdf/2105.01466v4

Title: Thinking Fast and Slow in Large Language Models
Date Published: 2022-12-10 05:07:30+00:00
URL: http://arxiv.org/pdf/2212.05206v2

Title: Machine Psychology: Investigating Emergent Capabilities and Behavior in Large Language Models Using Psychological Methods
Date Published: 2023-03-24 13:24:41+00:00
URL: http://arxiv.org/pdf/2303.13988v4

Title: Large Language Models for User Interest Journeys
Date Published: 2023-05-24 18:40:43+00:00
URL: http://arxiv.org/pdf/2305.15498v1

Title: Human-Like Intuitive Behavior and Reasoning Biases Emerged in Language Models -- and Disappeared in GPT-4
Date Published: 2023-06-13 08:43:13+00:00
URL: http://arxiv.org/pdf/2306.07622v2

Category: cs.SI

Title: Cultures in Community Question Answering
Date Published: 2015-08-20 16:55:42+00:00
URL: http://arxiv.org/pdf/1508.05044v1

Title: On the Behaviour of Deviant Communities in Online Social Networks
Date Published: 2016-10-26 15:13:35+00:00
URL: http://arxiv.org/pdf/1610.08372v1

Title: The Remarkable Benefit of User-Level Aggregation for Lexical-based Population-Level Predictions
Date Published: 2018-08-29 01:33:21+00:00
URL: http://arxiv.org/pdf/1808.09600v1

Title: Fast Algorithms for Intimate-Core Group Search in Weighted Graphs
Date Published: 2019-08-30 15:28:54+00:00
URL: http://arxiv.org/pdf/1908.11788v1

Title: Mining Bursting Communities in Temporal Graphs
Date Published: 2019-11-07 07:25:13+00:00
URL: http://arxiv.org/pdf/1911.02780v1

Title: RP-DNN: A Tweet level propagation context based deep neural networks for early rumor detection in Social Media
Date Published: 2020-02-28 12:44:34+00:00
URL: http://arxiv.org/pdf/2002.12683v2

Title: A Large-scale Friend Suggestion Architecture
Date Published: 2022-12-24 16:42:11+00:00
URL: http://arxiv.org/pdf/2212.12773v1

Title: The Systemic Impact of Deplatforming on Social Media
Date Published: 2023-03-20 14:30:52+00:00
URL: http://arxiv.org/pdf/2303.11147v1

Title: Lady and the Tramp Nextdoor: Online Manifestations of Economic Inequalities in the Nextdoor Social Network
Date Published: 2023-04-11 14:05:05+00:00
URL: http://arxiv.org/pdf/2304.05232v2

Category: cs.LG

Title: Real-Time Bidding by Reinforcement Learning in Display Advertising
Date Published: 2017-01-10 09:30:29+00:00
URL: http://arxiv.org/pdf/1701.02490v2

Title: The Hanabi Challenge: A New Frontier for AI Research
Date Published: 2019-02-01 18:59:07+00:00
URL: http://arxiv.org/pdf/1902.00506v2

Title: SINGA-Easy: An Easy-to-Use Framework for MultiModal Analysis
Date Published: 2021-08-03 08:39:54+00:00
URL: http://arxiv.org/pdf/2108.02572v1

Title: AutoHEnsGNN: Winning Solution to AutoGraph Challenge for KDD Cup 2020
Date Published: 2021-11-25 07:23:44+00:00
URL: http://arxiv.org/pdf/2111.12952v1

Title: Large Graph Models: A Perspective
Date Published: 2023-08-28 12:17:51+00:00
URL: http://arxiv.org/pdf/2308.14522v1

Category: cs.CV

Title: Physical world assistive signals for deep neural network classifiers -- neither defense nor attack
Date Published: 2021-05-03 04:02:48+00:00
URL: http://arxiv.org/pdf/2105.00622v1

Title: SCB-dataset: A Dataset for Detecting Student Classroom Behavior
Date Published: 2023-04-05 15:02:30+00:00
URL: http://arxiv.org/pdf/2304.02488v1

Title: PRAT: PRofiling Adversarial aTtacks
Date Published: 2023-09-20 07:42:51+00:00
URL: http://arxiv.org/pdf/2309.11111v1

Title: SCB-Dataset3: A Benchmark for Detecting Student Classroom Behavior
Date Published: 2023-10-04 01:43:46+00:00
URL: http://arxiv.org/pdf/2310.02522v1

Category: cs.AI

Title: Ease-of-Teaching and Language Structure from Emergent Communication
Date Published: 2019-06-06 03:59:37+00:00
URL: http://arxiv.org/pdf/1906.02403v2

Title: Secure Artificial Intelligence of Things for Implicit Group Recommendations
Date Published: 2021-04-23 16:38:26+00:00
URL: http://arxiv.org/pdf/2104.11699v1

Title: Computational Metacognition
Date Published: 2022-01-30 17:34:53+00:00
URL: http://arxiv.org/pdf/2201.12885v1

Title: Towards Cognitive Bots: Architectural Research Challenges
Date Published: 2023-05-26 23:51:49+00:00
URL: http://arxiv.org/pdf/2305.17308v1

Category: cs.DB

Title: Capturing Topology in Graph Pattern Matching
Date Published: 2011-12-31 05:34:57+00:00
URL: http://arxiv.org/pdf/1201.0229v1

Title: Graph Pattern Matching for Dynamic Team Formation
Date Published: 2018-01-03 14:24:08+00:00
URL: http://arxiv.org/pdf/1801.01012v1

Title: Efficient Top-k Ego-Betweenness Search
Date Published: 2021-07-21 12:46:52+00:00
URL: http://arxiv.org/pdf/2107.10052v1

Category: hep-ph

Title: Wandering in Color-Space -- why the life of pentaquark is so long ? --
Date Published: 2004-08-04 13:29:07+00:00
URL: http://arxiv.org/pdf/hep-ph/0408056v4

Title: XENON1T Anomaly and its Implication for Decaying Warm Dark Matter
Date Published: 2020-06-22 15:46:01+00:00
URL: http://arxiv.org/pdf/2006.12348v1

Category: cs.CR

Title: Malware Detection using Artificial Bee Colony Algorithm
Date Published: 2020-12-01 21:32:09+00:00
URL: http://arxiv.org/pdf/2012.00845v1

Title: A Survey of Neural Trojan Attacks and Defenses in Deep Learning
Date Published: 2022-02-15 04:26:44+00:00
URL: http://arxiv.org/pdf/2202.07183v1

Category: physics.soc-ph

Title: The Twitter of Babel: Mapping World Languages through Microblogging Platforms
Date Published: 2012-12-20 20:43:12+00:00
URL: http://arxiv.org/pdf/1212.5238v1

Title: Entropy and the Predictability of Online Life
Date Published: 2013-12-01 01:34:09+00:00
URL: http://arxiv.org/pdf/1312.0169v2

Category: physics.pop-ph

Title: Funding the Search for Extraterrestrial Intelligence with a Lottery Bond
Date Published: 2013-11-11 15:40:35+00:00
URL: http://arxiv.org/pdf/1311.2467v2

Title: Why do we find ourselves around a yellow star instead of a red star?
Date Published: 2017-05-22 15:48:36+00:00
URL: http://arxiv.org/pdf/1705.07813v1

Category: cs.CY

Title: Mining the Social Media Data for a Bottom-Up Evaluation of Walkability
Date Published: 2017-12-12 14:31:13+00:00
URL: http://arxiv.org/pdf/1712.04309v1

Title: Uncovering Bias in Personal Informatics
Date Published: 2023-03-27 20:49:42+00:00
URL: http://arxiv.org/pdf/2303.15592v2

Category: astro-ph.EP

Title: Does the evolution of complex life depend on the stellar spectral energy distribution?
Date Published: 2019-05-17 15:58:07+00:00
URL: http://arxiv.org/pdf/1905.07343v1

Title: Did life originate from low-temperature areas of the Universe?
Date Published: 2020-10-21 11:26:17+00:00
URL: http://arxiv.org/pdf/2010.10905v2

Category: cs.IR

Title: A Survey on Personality-Aware Recommendation Systems
Date Published: 2021-01-28 18:03:23+00:00
URL: http://arxiv.org/pdf/2101.12153v2

Title: SceneRec: Scene-Based Graph Neural Networks for Recommender Systems
Date Published: 2021-02-12 09:06:12+00:00
URL: http://arxiv.org/pdf/2102.06401v1

Category: cs.SE

Title: Can Requirements Engineering Support Explainable Artificial Intelligence? Towards a User-Centric Approach for Explainability Requirements
Date Published: 2022-06-03 11:17:41+00:00
URL: http://arxiv.org/pdf/2206.01507v1

Category: cs.FL

Title: What can we learn from universal Turing machines?
Date Published: 2021-10-16 08:43:29+00:00
URL: http://arxiv.org/pdf/2110.08511v1

Category: cs.SD

Title: MES-P: an Emotional Tonal Speech Dataset in Mandarin Chinese with Distal and Proximal Labels
Date Published: 2018-08-30 03:02:46+00:00
URL: http://arxiv.org/pdf/1808.10095v2

Category: q-bio.QM

Title: Local Causal Structure Learning and its Discovery Between Type 2 Diabetes and Bone Mineral Density
Date Published: 2020-06-27 08:27:00+00:00
URL: http://arxiv.org/pdf/2006.16791v1

Category: math.OC

Title: G-networks and the optimization of supply chains
Date Published: 2019-03-17 22:57:22+00:00
URL: http://arxiv.org/pdf/1903.10691v1

Category: stat.ME

Title: Outlyingness: why do outliers lie out?
Date Published: 2017-08-12 10:48:58+00:00
URL: http://arxiv.org/pdf/1708.03761v1

Category: astro-ph.GA

Title: Four hot DOGs in the microwave
Date Published: 2015-10-14 16:02:10+00:00
URL: http://arxiv.org/pdf/1510.04179v1

Category: cs.HC

Title: BEAMERS: Brain-Engaged, Active Music-based Emotion Regulation System
Date Published: 2022-11-26 16:37:13+00:00
URL: http://arxiv.org/pdf/2211.14609v1