Task-Guided and Path-Augmented Heterogeneous Network Embedding for Author Identification

In this paper, we study the problem of author identification under double-blind review setting, which is to identify potential authors given information of an anonymized paper. Different from existing approaches that rely heavily on feature engineering, we propose to use network embedding approach to address the problem, which can automatically represent nodes into lower dimensional feature vectors. However, there are two major limitations in recent studies on network embedding: (1) they are usually general-purpose embedding methods, which are independent of the specific tasks; and (2) most of these approaches can only deal with homogeneous networks, where the heterogeneity of the network is ignored. Hence, challenges faced here are two folds: (1) how to embed the network under the guidance of the author identification task, and (2) how to select the best type of information due to the heterogeneity of the network. To address the challenges, we propose a task-guided and path-augmented heterogeneous network embedding model. In our model, nodes are first embedded as vectors in latent feature space. Embeddings are then shared and jointly trained according to task-specific and network-general objectives. We extend the existing unsupervised network embedding to incorporate meta paths in heterogeneous networks, and select paths according to the specific task. The guidance from author identification task for network embedding is provided both explicitly in joint training and implicitly during meta path selection. Our experiments demonstrate that by using path-augmented network embedding with task guidance, our model can obtain significantly better accuracy at identifying the true authors comparing to existing methods.

• Publication year

  2016

• Venue

  Web Search and Data Mining

• Publication date

  2016-12-08

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1145/3018661.3018735arXiv 1612.02814
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• Using path-augmented network embedding with task guidance improves accuracy for identifying the true authors compared with existing methods.

  Confidence 0.94

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• Task guidance enters the embedding process both explicitly through joint training and implicitly through meta path selection.

  Confidence 0.91

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• Node embeddings are shared and jointly trained under task-specific and network-general objectives.

  Confidence 0.95

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• The task-guided and path-augmented heterogeneous network embedding model extends unsupervised network embedding with meta paths for heterogeneous networks and uses task-driven path selection for author identification.

  Confidence 0.96

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review

• author identification task

  task

  The problem of identifying likely authors of an anonymized paper from available information.

  Aliases: author identification

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• heterogeneous network

  network

  A network containing multiple types of nodes or relations that can be exploited by embedding methods.

  Aliases: heterogeneous graph

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• joint training

  training method

  A training scheme in which embeddings are optimized together using multiple objectives.

  Aliases: jointly trained

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• meta path

  representation, path

  A typed path schema used to capture structured connectivity patterns in a heterogeneous network.

  Aliases: metapath, meta-path

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• network embedding

  method

  A representation learning approach that maps network nodes into low-dimensional feature vectors.

  Aliases: node embedding

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• task-guided and path-augmented heterogeneous network embedding model

  method, model

  The proposed embedding model that combines task guidance with path information for heterogeneous networks.

  Aliases: path-augmented heterogeneous network embedding model, task-guided heterogeneous network embedding model

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review
• task-specific and network-general objectives

  objective, loss function

  The pair of optimization goals used to balance performance on the author identification task with general network structure learning.

  Aliases: task-specific objective, network-general objective

  박진우 (dztg5apj7m) extractionAnonymous (12632b8b5f) review

• Meta Path Based Top-K Similarity Join In Heterogeneous Information Networks

  2016cited by this paper
• Entity Embedding-Based Anomaly Detection for Heterogeneous Categorical Events

  2016cited by this paper
• PTE: Predictive Text Embedding through Large-scale Heterogeneous Text Networks

  2015cited by this paper
• Mining Quality Phrases from Massive Text Corpora

  2015cited by this paper
• Semantic Path based Personalized Recommendation on Weighted Heterogeneous Information Networks

  2015influential reference
• LINE: Large-scale Information Network Embedding

  2015cited by this paper
• Heterogeneous Network Embedding via Deep Architectures

  2015cited by this paper
• Combination of feature engineering and ranking models for paper-author identification in KDD Cup 2013

  2015influential reference
• Personalized entity recommendation: a heterogeneous information network approach

  2014cited by this paper
• DeepWalk: online learning of social representations

  2014influential reference
• PathSelClus

  2013cited by this paper
• Efficient Estimation of Word Representations in Vector Space

  2013influential reference
• Contextual rule-based feature engineering for author-paper identification

  2013cited by this paper
• Combination of feature engineering and ranking models for paper-author identification in KDD Cup 2013

  2013cited by this paper
• Translating Embeddings for Modeling Multi-relational Data

  2013cited by this paper
• Distributed Representations of Words and Phrases and their Compositionality

  2013influential reference
• KDD Cup 2013 - author-paper identification challenge: second place team

  2013cited by this paper
• Feature engineering and tree modeling for author-paper identification challenge

  2013cited by this paper
• The scorecard solution to the author-paper identification challenge

  2013cited by this paper
• When will it happen?: relationship prediction in heterogeneous information networks

  2012influential reference
• Mining Heterogeneous Information Networks: Principles and Methodologies

  2012influential reference
• Integrating meta-path selection with user-guided object clustering in heterogeneous information networks

  2012cited by this paper
• PathSim

  2011influential reference
• Co-author Relationship Prediction in Heterogeneous Bibliographic Networks

  2011cited by this paper
• Reduction A Global Geometric Framework for Nonlinear Dimensionality

  2011cited by this paper
• Why Does Unsupervised Pre-training Help Deep Learning?

  2010cited by this paper
• Large-Scale Machine Learning with Stochastic Gradient Descent

  2010cited by this paper
• ArnetMiner: extraction and mining of academic social networks

  2008cited by this paper
• Graph Embedding and Extensions: A General Framework for Dimensionality Reduction

  2007cited by this paper
• Learning Deep Architectures for AI

  2007cited by this paper
• The myth of the double-blind review?: author identification using only citations

  2003cited by this paper
• Statistical mechanics of complex networks

  2001cited by this paper
• Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering

  2001cited by this paper
• Nonlinear dimensionality reduction by locally linear embedding.

  2000cited by this paper
• A global geometric framework for nonlinear dimensionality reduction.

  2000cited by this paper
• Effect of open peer review on quality of reviews and on reviewers'recommendations: a randomised trial

  1999cited by this paper
• An Efficient Method for Generating Discrete Random Variables with General Distributions

  1977cited by this paper

• Navigating complexity: a comprehensive review of heterogeneous information networks and embedding techniques

  2025cites this paper
• The community discovery method in heterogeneous networks: a survey

  2025cites this paper
• HGExplainer: Toward Interpretable Heterogeneous Graph Neural Networks via Meta-path Perturbation

  2025cites this paper
• Temporal Enhancement of Top-N Recommendation on Heterogeneous Graphs

  2025cites this paper
• Attack Smarter: Attention-Driven Fine-Grained Webpage Fingerprinting Attacks

  2025influential citation
• SiHGNN: Leveraging Properties of Semantic Graphs for Efficient HGNN Acceleration

  2024cites this paper
• REHREC: Review Effected Heterogeneous Information Network Recommendation System

  2024cites this paper
• A Heterogeneous Graph Neural Network With Attribute Enhancement and Structure-Aware Attention

  2024cites this paper
• Graph embedding on mass spectrometry- and sequencing-based biomedical data

  2024cites this paper
• Towards Fine-Grained Webpage Fingerprinting at Scale

  2024cites this paper
• MRH-GCN: An Efficient GCN Accelerator for Multi-Relation Heterogeneous Graph

  2024cites this paper
• Review of heterogeneous graph embedding methods based on deep learning techniques and comparing their efficiency in node classification

  2024cites this paper
• Accelerating Mini-batch HGNN Training by Reducing CUDA Kernels

  2024cites this paper
• ES-HMS: Efficient Spatiotemporal Few-Shot Learning in Hyperbolic Space

  2024cites this paper
• Few-Shot Causal Representation Learning for Out-of-Distribution Generalization on Heterogeneous Graphs

  2024cites this paper
• A Unified Generative Adversarial Learning Framework for Improvement of Skip-Gram Network Representation Learning Methods

  2023cites this paper
• ASIAM-HGNN: Automatic Selection and Interpretable Aggregation of Meta-Path Instances for Heterogeneous Graph Neural Network

  2023cites this paper
• HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation

  2023cites this paper
• Cross-heterogeneity Graph Few-shot Learning

  2023cites this paper
• Understanding writing style in social media with a supervised contrastively pre-trained transformer

  2023cites this paper
• Robust representation learning for heterogeneous attributed networks

  2023cites this paper
• Zoo guide to network embedding

  2023cites this paper
• TwHIN-BERT: A Socially-Enriched Pre-trained Language Model for Multilingual Tweet Representations at Twitter

  2022cites this paper
• Heterogeneous Information Networks: the Past, the Present, and the Future

  2022cites this paper
• Interpretable Relation Learning on Heterogeneous Graphs

  2022cites this paper
• Simple and Efficient Heterogeneous Graph Neural Network

  2022cites this paper
• PART: Pre-trained Authorship Representation Transformer

  2022cites this paper
• TwHIN: Embedding the Twitter Heterogeneous Information Network for Personalized Recommendation

  2022cites this paper
• MultiBiSage: A Web-Scale Recommendation System Using Multiple Bipartite Graphs at Pinterest

  2022cites this paper
• Learning heterogeneous graph embedding for Chinese legal document similarity

  2022cites this paper
• Feature-enhanced embedding learning for heterogeneous collaborative filtering

  2022cites this paper
• Spatiotemporal-Enhanced Network for Click-Through Rate Prediction in Location-based Services

  2022cites this paper
• Cluster-Aware Heterogeneous Information Network Embedding

  2022cites this paper
• Predicting coauthorship using bibliographic network embedding

  2022cites this paper
• Predicting combinations of drugs by exploiting graph embedding of heterogeneous networks

  2022cites this paper
• SILK-BPR: Identify Implicit Friends with Self-guided Walking for Social Recommendation

  2022cites this paper
• Knowledge Concept Recommendation Model for MOOCs with Local Sub-graph Embedding

  2022cites this paper
• Attention based Collaborator Recommendation in Heterogeneous Academic Networks

  2022cites this paper
• Movie Recommender System Based On Heterogeneous Graph Neural Networks

  2022cites this paper
• A group behavior prediction model based on sparse representation and complex message interactions

  2022cites this paper
• Hotness prediction of scientific topics based on a bibliographic knowledge graph

  2022cites this paper
• Inductive Contextual Relation Learning for Personalization

  2021influential citation
• A Framework for Joint Unsupervised Learning of Cluster-Aware Embedding for Heterogeneous Networks

  2021cites this paper
• Local Structural Aware Heterogeneous Information Network Embedding Based on Relational Self-Attention Graph Neural Network

  2021cites this paper
• Bug Triage Model Considering Cooperative and Sequential Relationship

  2021cites this paper
• Walking With Attention: Self-Guided Walking for Heterogeneous Graph Embedding

  2021cites this paper
• Learning multi-resolution representations of research patterns in bibliographic networks

  2021cites this paper
• HINTS: Citation Time Series Prediction for New Publications via Dynamic Heterogeneous Information Network Embedding

  2021cites this paper
• MixSp: A Framework for Embedding Heterogeneous Information Networks With Arbitrary Number of Node and Edge Types

  2021cites this paper
• Predicting Paper Acceptance via Interpretable Decision Sets

  2021cites this paper
• Network representation learning systematic review: ancestors and current development state

  2021cites this paper
• Higher-Order Attribute-Enhancing Heterogeneous Graph Neural Networks

  2021cites this paper
• Survey on graph embeddings and their applications to machine learning problems on graphs

  2021cites this paper
• Network representation learning: A macro and micro view

  2021cites this paper
• Graph neural networks

  2021cites this paper
• Meta-Path-Free Representation Learning on Heterogeneous Networks

  2021cites this paper
• Heterogeneous Graph Representation for Recommendation

  2021cites this paper
• Heterogeneous Hypergraph Variational Autoencoder for Link Prediction

  2021cites this paper
• Introduction

  2021cites this paper
• The State-of-the-Art of Heterogeneous Graph Representation

  2021cites this paper
• Graph Neural Networks in Computer Vision

  2021cites this paper
• Deep Learning on Graphs: An Introduction

  2021cites this paper
• Topic-aware Heterogeneous Graph Neural Network for Link Prediction

  2021cites this paper
• Attention-Aware Encoder–Decoder Neural Networks for Heterogeneous Graphs of Things

  2021cites this paper
• Automated Unsupervised Graph Representation Learning

  2021cites this paper
• Attribute-Assisted Heterogeneous Graph Representation

  2021cites this paper
• Beyond GNNs: More Deep Models on Graphs

  2021cites this paper
• Graph Embedding

  2021cites this paper
• Advanced Topics in Graph Neural Networks

  2021cites this paper
• Preface

  2021cites this paper
• Deep Learning on Graphs

  2021cites this paper
• Bibliography

  2021cites this paper
• Graph Neural Networks in Natural Language Processing

  2021cites this paper
• Foundations of Deep Learning

  2021cites this paper
• RPT: Toward Transferable Model on Heterogeneous Researcher Data via Pre-Training

  2021cites this paper
• Index

  2021cites this paper
• HisRect: Features from Historical Visits and Recent Tweet for Co-Location Judgement

  2021cites this paper
• Scalable Graph Neural Networks

  2021cites this paper
• Learning Embeddings Based on Global Structural Similarity in Heterogeneous Networks

  2021cites this paper
• ISHNE: Influence Self-attention for Heterogeneous Network Embedding

  2021cites this paper
• mSHINE: A Multiple-Meta-Paths Simultaneous Learning Framework for Heterogeneous Information Network Embedding

  2021cites this paper
• CMG2Vec: A composite meta-graph based heterogeneous information network embedding approach

  2021cites this paper
• Advanced Applications in Graph Neural Networks

  2021cites this paper
• Modeling Complementarity in Behavior Data with Multi-Type Itemset Embedding

  2021cites this paper
• FONDUE: A Framework for Node Disambiguation and Deduplication Using Network Embeddings

  2021cites this paper
• Foundations

  2021cites this paper
• Robust Graph Neural Networks

  2021cites this paper
• An Efficient Neighborhood-based Interaction Model for Recommendation on Heterogeneous Graph

  2020cites this paper
• Deep Inductive Graph Representation Learning

  2020cites this paper
• A novel node-level structure embedding and alignment representation of structural networks for brain disease analysis

  2020cites this paper
• Co-purchaser Recommendation for Online Group Buying

  2020cites this paper
• Embedding Based Personalized New Paper Recommendation

  2020cites this paper
• Path-Based Reasoning over Heterogeneous Networks for Recommendation via Bidirectional Modeling

  2020cites this paper
• HisRect: Features from Historical Visits and Recent Tweet for Co-Location Judgement

  2020cites this paper
• Heterogeneous Network Representation Learning

  2020cites this paper
• FONDUE: A Framework for Node Disambiguation Using Network Embeddings

  2020cites this paper
• Attentional Graph Convolutional Networks for Knowledge Concept Recommendation in MOOCs in a Heterogeneous View

  2020cites this paper
• Unsupervised Differentiable Multi-aspect Network Embedding

  2020cites this paper
• SeHNE: Semi-supervised Heterogeneous Network Embedding for Drug Combination

  2020cites this paper
• CODE: Contrastive Pre-training with Adversarial Fine-Tuning for Zero-Shot Expert Linking

  2020cites this paper