Ranking-based classification of heterogeneous information networks

  title={Ranking-based classification of heterogeneous information networks},
  author={Ming Ji and Jiawei Han and Marina Danilevsky},
  booktitle={Knowledge Discovery and Data Mining},
It has been recently recognized that heterogeneous information networks composed of multiple types of nodes and links are prevalent in the real world. [] Key Method Specifically, we build a graph-based ranking model to iteratively compute the ranking distribution of the objects within each class. At each iteration, according to the current ranking results, the graph structure used in the ranking algorithm is adjusted so that the sub-network corresponding to the specific class is emphasized, while the rest of…

Figures and Tables from this paper

Clustering and Ranking in Heterogeneous Information Networks via Gamma-Poisson Model

A probabilistic generative model is proposed that simultaneously achieves clustering and ranking on a heterogeneous network that can follow arbitrary schema, where the edges from different types are sampled from a Poisson distribution with the parameters determined by the ranking scores of the nodes in each cluster.

Ranking-based Clustering on General Heterogeneous Information Networks by Network Projection

This paper studies the ranking-based clustering problem in a general heterogeneous information network and proposes a novel solution HeProjI, which can achieve better clustering and ranking performances compared to other well-established algorithms.

A Noval Weighted Meta Graph Method for Classification in Heterogeneous Information Networks

A novel framework for the weighted Meta graph-based Classification of Heterogeneous Information Networks (MCHIN), which can effectively outperform the baselines algorithms and significantly enlarges the training sets by introducing the concept of Extension Meta Graphs in HINs.

Meta path-based collective classification in heterogeneous information networks

This paper introduces the concept of meta-path based dependencies among objects, where a meta path is a path consisting a certain sequence of linke types, and shows that the quality of collective classification results strongly depends upon the meta paths used.

Multi-label Collective Classification Using Adaptive Neighborhoods

A neighborhood ranking method for multi-label classification, which can be further used in the Multi-label Collective Classification framework, and experimental results show that the use of ranking in neighborhood selection for collective classification improves the performance of the classifier.

Meta-Path-Based Search and Mining in Heterogeneous Information Networks

This work proposes to explore the meta structure of the information network, i.e., the network schema, to systematically capture numerous semantic relationships across multiple types of objects, defined as a path over the graph of network schema.

Ensemble Learning for Multi-Type Classification in Heterogeneous Networks

The system Mr-SBC is adopted, which is originally able to analyze heterogeneous networks of arbitrary structure, within an ensemble learning approach, to improve the classification accuracy of Mr- SBC by exploiting the possible presence of correlation and autocorrelation phenomena.

A General Recommendation Model for Heterogeneous Networks

This paper proposes a graph-based model, called HeteRS, which can solve general recommendation problems on heterogeneous networks and significantly outperforms the baseline methods in the authors' experiments for all the recommendation tasks, and the learned influence weights help understanding user behaviors.

Tensor Based Relations Ranking for Multi-relational Collective Classification

An algorithm, TensorRRCC, is proposed, able to determine the ranking of relations and the labels of objects simultaneously, which is compared with current collective classification algorithms on two real-world data sets and the experimental results show the superiority of the method.



Ranking-based clustering of heterogeneous information networks with star network schema

This paper studies clustering of multi-typed heterogeneous networks with a star network schema and proposes a novel algorithm, NetClus, that utilizes links across multityped objects to generate high-quality net-clusters and generates informative clusters.

Graph Regularized Transductive Classification on Heterogeneous Information Networks

This paper considers the transductive classification problem on heterogeneous networked data which share a common topic and proposes a novel graph-based regularization framework, GNetMine, to model the link structure in information networks with arbitrary network schema and arbitrary number of object/link types.

Improving web search results using affinity graph

A novel ranking scheme named Affinity Ranking (AR) is proposed to re-rank search results by optimizing two metrics: diversity -- which indicates the variance of topics in a group of documents; and information richness -- which measures the coverage of a single document to its topic.

Object-level ranking: bringing order to Web objects

The experimental results show that PopRank can achieve significantly better ranking results than naively applying PageRank on the object graph, and the proposed efficient approaches to automatically decide these factors are proposed.

Classification in Networked Data: a Toolkit and a Univariate Case Study

The results demonstrate that very simple network-classification models perform quite well---well enough that they should be used regularly as baseline classifiers for studies of learning with networked data.

Graph-based Consensus Maximization among Multiple Supervised and Unsupervised Models

This paper proposes to consolidate a classification solution by maximizing the consensus among both supervised predictions and unsupervised constraints, and casts this ensemble task as an optimization problem on a bipartite graph, where the objective function favors the smoothness of the prediction over the graph, as well as penalizing deviations from the initial labeling provided by supervised models.

Combining link and content for community detection: a discriminative approach

A discriminative model for combining the link and content analysis for community detection from networked data, such as paper citation networks and Word Wide Web is proposed and introduced and hidden variables are introduced to explicitly model the popularity of nodes.

Spectral clustering for multi-type relational data

A general model, the collective factorization on related matrices, is proposed for multi-type relational data clustering and a novel algorithm is derived, the spectral relational clustering, to cluster multi- type interrelated data objects simultaneously.

Grouper: A Dynamic Clustering Interface to Web Search Results

Transfer Learning for Collective Link Prediction in Multiple Heterogenous Domains

This paper proposes a nonparametric Bayesian framework for solving the CLP problem, which allows knowledge to be adaptively transferred across heterogeneous tasks while taking into account the similarities between tasks.