constructing maps of intellectual influence from publication data

Overview

Influencemap Project @ ANU

Influence in the academic communities has been an area of interest for researchers. This can be seen in the popularity of applications like google scholar and the various metrics created for ranking papers, authors, conferences, etc.

We aim to provide a visualisation tool which allows users to easily search and visualise the flow of academic influence. Our visualisation maps influence in the form of an influence flower. We calculate influence as a function of the number of citations between two entities (look below for information on our definition of influence).

The node in the centre of the flower denotes the ego entity, the entitiy in which we are looking at influence with respect to. The leaf nodes are the most influential entities with respect to the ego. (We define the ego as a collection of papers. If it is an author, it is the collection of papers that the author has authored)

Each of the edges of the graph signifies the flow of influence to and from the ego node, the strength of this relation is reflected in the thickness of the edge. The red edges denote the influence the ego has towards the outer entities (an outer entity citing a paper by the ego). The blue edges denote the influence the outer entities have towards the ego (the ego cites a paper by one of the outer entities).

The colour of the outer nodes signifies the ratio of influence in and out. A blue node indicates that the associated entity has influenced the ego more than the ego has influenced itself. Likewise, a red node indicates the ego has influenced the node's entity more than it has influenced the ego.

We define two entities to be coauthors if the entities have contributed to the same paper. Coauthors of the ego are signified by nodes with greyed out names.

Data

We use the microsoft academic graph (MAG) dataset for our visualisation. The dataset is a large curation of publication indexed by Bing. From MAG, we use the following fields of the paper entries in the dataset,

  • Citation links
  • Authors
  • Conferences
  • Journals
  • Author Affiliations

Influence

To quantify academic influence, we define influence as a function of paper citations. Each citation which the ego is apart of contributes to the overall influence map of an ego. To prevent papers with a large number of entities contributing from creating an overwhelming amount of influence, we normalise the influence contribution by the number of entities in the cited paper.

For example, consider the following four paper database where we only consider entities which are authors.

Name Paper no. authors cites papers
John Smith Algorithms 2 [Linear Algebra]
John Smith Machine Learning 3 [Linear Algebra, Computation]
Maria Garcia Linear Algebra 2 None
Maria Garcia Computation 4 [Algorithms]

In this case John's influence on Maria is 0.5 (John's paper Algorithm's has a weight of 0.5 and was cited once by Maria).

On the other hand Maria's influence on John is 1.25 (Linear Algebra has a weight of 0.5 and it was cited twice by John, Computation has a weight of 0.25 and was cited once by John).

We aggregate the pairwise influence of entities associated with the papers of the ego to generate the nodes of a flower. Each flowers' outer nodes can be a collection of several types of entities. In our influence flower application, we present 4 different flower types:

  1. Author outer nodes
  2. Venue (conferences or journals) outer nodes
  3. Author Affiliation outer nodes
  4. Paper topic outer nodes

Filtering self-citations

We define a self-citation between papers and a cited paper as a relation dependent on the ego. A paper citation is a self-citation if both papers have the ego as an author (a venue, an institution, or a topic).

Filtering co-contributors

The Influence Flower is able to capture less obvious influence outside of one’s co-author networks with the filtering. We define two entities to be co-contributors if the entities have contributed to the same paper. For the venue type entity, co-contribution indicates if the ego has published a paper to the venue. For the topic type entity, it means that the ego has written a paper of the topic. Co-contributors of the ego are indicated by nodes with greyed out names.

Other candidate definitions of influence

We have described influence as the sum of citations from one person (or venue or affiliation) to another, weighted by the number of authors in the cited paper. Similar methods were considered early on in the project which included combinations of different weighting schemes. We looked at the eight combinations of three mutually exclusive weightings:

  1. Weighting by the number of authors on the citing paper;
  2. Weighting by the number of authors on the cited paper; and
  3. Weighting by the number of papers referenced by the citing paper.

Due to the lack of a ground truth value of influence to compare these definitions to, we evaluated the eight combinations of these weightings empirically by discussing with researchers which of the definitions produced flowers that most accurately reflected their opinions of who they have influenced and been influenced by.

Other definitions of influence which have not yet been explored with this data include existing measures for node centrality in graphs. By using citation data from MAG to define a directed graph where nodes represent authors, venues or affiliations, and edges are derived from citations between nodes, we could explore using metrics such as closeness, betweenness and eigenvector centrality. These metrics are more appropriate for defining the influence of an entity relative to the whole network.

Owner
CS Metrics
CS Metrics
NCVX (NonConVeX): A User-Friendly and Scalable Package for Nonconvex Optimization in Machine Learning.

The source code is temporariy removed, as we are solving potential copyright and license issues with GRANSO (http://www.timmitchell.com/software/GRANS

SUN Group @ UMN 28 Aug 03, 2022
Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation

Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation

Bae, Gwangbin 95 Jan 04, 2023
Tensorflow 2 Object Detection API kurulumu, GPU desteği, custom model hazırlama

Tensorflow 2 Object Detection API Bu tutorial, TensorFlow 2.x'in kararlı sürümü olan TensorFlow 2.3'ye yöneliktir. Bu, görüntülerde / videoda nesne a

46 Nov 20, 2022
A static analysis library for computing graph representations of Python programs suitable for use with graph neural networks.

python_graphs This package is for computing graph representations of Python programs for machine learning applications. It includes the following modu

Google Research 258 Dec 29, 2022
SciFive: a text-text transformer model for biomedical literature

SciFive SciFive provided a Text-Text framework for biomedical language and natural language in NLP. Under the T5's framework and desrbibed in the pape

Long Phan 54 Dec 24, 2022
Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21)

Learning Structural Edits via Incremental Tree Transformations Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21) 1.

NeuLab 40 Dec 23, 2022
Implementation for paper "Towards the Generalization of Contrastive Self-Supervised Learning"

Contrastive Self-Supervised Learning on CIFAR-10 Paper "Towards the Generalization of Contrastive Self-Supervised Learning", Weiran Huang, Mingyang Yi

Weiran Huang 13 Nov 30, 2022
InvTorch: memory-efficient models with invertible functions

InvTorch: Memory-Efficient Invertible Functions This module extends the functionality of torch.utils.checkpoint.checkpoint to work with invertible fun

Modar M. Alfadly 12 May 12, 2022
The devkit of the nuPlan dataset.

The devkit of the nuPlan dataset.

Motional 264 Jan 03, 2023
Creating predictive checklists from data using integer programming.

Learning Optimal Predictive Checklists A Python package to learn simple predictive checklists from data subject to customizable constraints. For more

Healthy ML 5 Apr 19, 2022
Self-Guided Contrastive Learning for BERT Sentence Representations

Self-Guided Contrastive Learning for BERT Sentence Representations This repository is dedicated for releasing the implementation of the models utilize

Taeuk Kim 16 Dec 04, 2022
A different spin on dataclasses.

dataklasses Dataklasses is a library that allows you to quickly define data classes using Python type hints. Here's an example of how you use it: from

David Beazley 752 Nov 18, 2022
Learnable Boundary Guided Adversarial Training (ICCV2021)

Learnable Boundary Guided Adversarial Training This repository contains the implementation code for the ICCV2021 paper: Learnable Boundary Guided Adve

DV Lab 27 Sep 25, 2022
Streamlit app demonstrating an image browser for the Udacity self-driving-car dataset with realtime object detection using YOLO.

Streamlit Demo: The Udacity Self-driving Car Image Browser This project demonstrates the Udacity self-driving-car dataset and YOLO object detection in

Streamlit 992 Jan 04, 2023
AITUS - An atomatic notr maker for CYTUS

AITUS an automatic note maker for CYTUS. 利用AI根据指定乐曲生成CYTUS游戏谱面。 效果展示:https://www

GradiusTwinbee 6 Feb 24, 2022
Accurate Phylogenetic Inference with Symmetry-Preserving Neural Networks

Accurate Phylogenetic Inference with a Symmetry-preserving Neural Network Model Claudia Solis-Lemus Shengwen Yang Leonardo Zepeda-Núñez This repositor

Leonardo Zepeda-Núñez 2 Feb 11, 2022
Project Tugas Besar pertama Pengenalan Komputasi Institut Teknologi Bandung

Vending_Machine_(Mesin_Penjual_Minuman) Project Tugas Besar pertama Pengenalan Komputasi Institut Teknologi Bandung Raw Sketch untuk Essay Ringkasan P

QueenLy 1 Nov 08, 2021
Red Team tool for exfiltrating files from a target's Google Drive that you have access to, via Google's API.

GD-Thief Red Team tool for exfiltrating files from a target's Google Drive that you(the attacker) has access to, via the Google Drive API. This includ

Antonio Piazza 39 Dec 27, 2022
Pytorch implementation of the paper Improving Text-to-Image Synthesis Using Contrastive Learning

T2I_CL This is the official Pytorch implementation of the paper Improving Text-to-Image Synthesis Using Contrastive Learning Requirements Linux Python

42 Dec 31, 2022
Official implementation for Multi-Modal Interaction Graph Convolutional Network for Temporal Language Localization in Videos

Multi-modal Interaction Graph Convolutioal Network for Temporal Language Localization in Videos Official implementation for Multi-Modal Interaction Gr

Zongmeng Zhang 15 Oct 18, 2022