INTRODUCTION

Various application domains such as scientific data mining, precision medicine, talent intelligence, communication networks naturally generate complex heterogeneous data. These heterogeneous data are typically multi-modal, graph structure and dynamic. In this big data era, effectively mining from such data is becoming more and more important. The 1st International Workshop on Complex Heterogeneous Data Mining (CHDM) aims to provide a forum for researchers from academia and industry to exchange ideas, techniques and application scenarios in complex heterogeneous data mining as well as discuss open challenges and identify new research directions in the area. It aims to integrate techniques from various areas into solving problems in complex heterogeneous data mining by focusing on research topics such as machine learning techniques for complex heterogeneous data and AI techniques for complex heterogeneous data. The proposed workshop well matches the interests of ICDM 2023. It is closely related to a number of research tracks in ICDM 2023 on the topics of mining from heterogeneous data sources, deep learning and statistical methods for data mining, foundations, algorithms, models and theory of data mining. The workshop is also expected to spark discussions on novel technique paradigms and application scenarios for complex heterogeneous data mining. Besides regular research papers, we also welcome vision papers, demonstration papers and papers with industry showcase from various applications.

TOPICS OF INTEREST

The workshop will be of interest to researchers in developing techniques for complex heterogeneous data mining in various application domains. The intended audiences include researchers from both academia and industry who are interested in exploiting the value of complex heterogeneous data.

Topics of interest include but not limited to:

  • Multi-modal data mining techniques
  • Multi-modal models for real-world applications, such as classification, clustering, retrieval, generation
  • Big data driven multi-modal learning for foundation models
  • Multi-modal Graph Learning
  • Graph mining techniques
  • Dynamic and streaming graph data analytics
  • Graph analytics in various application domains such as social networks multimedia, semantic web, biological data, business processes, transport data, etc.
  • Vision papers to survey the area of heterogenous data analytics as well as describe the future research directions

CALL FOR PAPER

We cordially request the submission of research papers on a regular basis, (max 8 pages plus 2 extra pages), including all content and references. The submissions must be in PDF format, and must adhere to the new Standard IEEE Conference Proceedings Template.

Submitted papers will undergo an evaluation based on their originality, technical quality, potential impact, insightfulness, depth, clarity, and reproducibility. In light of the practical nature of this workshop, we strongly encourage authors to submit their demonstrations, which will also be subject to evaluation during the review process.

All papers must be submitted through the Cyberchair system.

Continuing the unique tradition of ICDM, all workshop papers that are accepted will be published in the dedicated ICDM proceedings by the IEEE Computer Society Press.

We will be awarding this thrilling best paper award, announced in our closing remarks:

● Best Paper Award: [The exact amount may be subject to changes since we are still defining them.]

Important Dates

● Workshop Paper submission: September 15, 2023

● Notification of workshop papers acceptance to authors: September 24, 2023

● Camera-ready deadline and copyright form: October 1, 2023

● Workshop Day: December 4, 2023

Please Note: All times are at 11:59PM Beijing Time.

Organizers

Yang Yang

Nanjing University of Science and Technology

Long Yuan

Nanjing University of Science and Technology

Wenjie Zhang

University of New South Wales Australia

Weili Guo

Nanjing University of Science and Technology

SPONSORS

×
Processing SPARQL Queries Over Distributed RDF Graphs -
A Partial Evaluation and Assembly Approach

Prof. Lei Zou

Peking University

Abstract

We propose techniques for processing SPARQL queries over a large RDF graph in a distributed environment. We adopt a "partial evaluation and assembly" framework. Answering a SPARQL query Q is equivalent to finding subgraph matches of the query graph Q over RDF graph G. Based on properties of subgraph matching over a distributed graph, we introduce local partial match as partial answers in each fragment of RDF graph G. For assembly, we propose two methods: centralized and distributed assembly.

Short Biography

Lei Zou received his BS degree and Ph.D. degree in Computer Science at Huazhong University of Science and Technology (HUST) in 2003 and 2009, respectively. He received a CCF (China Computer Federation) Doctoral Dissertation Nomination Award in 2009, won Second Class Prize of CCF Natural Science Award in 2014 and Second Class Prize of Natural Science of the Ministry of Education, China in 2017. Since September 2009, he joined Institute of Computer Science and Technology (ICST) of Peking University (PKU) as a faculty member. He has been a professor in PKU since August 2017. Before joining PKU, he visited Hong Kong University of Science and Technology in 2007 and University of Waterloo in 2008 as a visiting scholar. His recent research interests include graph databases, knowledge graph, particularly in graph-based RDF data management. He has published more than 50 papers, including more than 30 papers published in reputed journals and major international conferences, such as SIGMOD, VLDB, ICDE, TODS, TKDE, VLDB Journal. Lei Zou's research is supported by NSFC-Young Excellent Talent Project and National Key Research and Development Program of China.

Close
×
Architecting a Comprehensive Enterprise Graph Platform

Dr. Yinglong Xia

Huawei Research America

Abstract

Graph technology has been playing increasingly important roles in various machine learning, data analytics, and resource management domains, thus more and more companies have been adopting/utilizing graph platforms, either on cloud or on premise, to support their business. In this talk, we will investigate various factors that contribute to the success of a graph platform for enterprise use, ranging from graph data organization, runtime scheduling, analytics optimization, to some thoughts on recent graph deep learning frameworks. We will discuss through some concrete examples on how to effectively put together the above building blocks, so as to form a comprehensive graph platform delivering efficient end-to-end performance to meet the requirements in many industrial scenarios. At last, we will brief some recent activities in graph technology community, such as the discussions on its standardization, along with the summary of the challenges and opportunities.

Short Biography

Yinglong Xia is a chief architect at Huawei Research America, working on AI platforms and Graph Engine Service (GES, https://www.huaweicloud.com/en-us/product/ges.html). Prior to that, he was a technical leader and research staff member at IBM Watson Research Center, exploring graph database and reasoning framework, creating the IBM System G (http://systemg.mybluemix.net/) platform. He has solid experience in both industrial research and product development, already published 60+ technical papers and filed 30+ patents. He serves as a technical advisory committee (TAC) member in Linux Foundation, a board member of LDBC, and an associate editor of IEEE trans. Knowledge and Data Engineering (TKDE), and IEEE trans. Big Data (TBD); he is a general co-chair of IEEE HiPC'19, vice co-chair of IEEE BigData'19, TPC member of KDD'19, VLDB'19, and ICDE'19, etc.

Close