ICDM Tutorial on Social Multimedia as Sensors (by Jiebo Luo and Tao Mei)

Call for Papers

Different social networking platforms have become a ubiquitous means for information sharing and communication, especially with the ever increasing mobile device availability. As a result, increasingly rich and large-scale social multimedia data (in the form of text, image, audio, and video) are being generated and posted to different social networking and media sharing platforms such as Twitter, Facebook, Orkut, Instagram, Pinterest,  Flickr, Vine, and Youtube. Often the multimedia content posted in these social platforms is accompanied with user comments, tags, likes, hashtags, upvotes, and so on. Such large-scale multimedia data with rich contextual information has wide ranging practical applications such as user profiling, behavior analysis, advanced personalization and recommendation systems, marketing, etc. and offers unique research challenges that require efforts from multiple disciplines such as data mining, machine learning, image and video processing, computer vision, and information retrieval. Through this workshop, we intend to offer a common platform to multidisciplinary researchers from academia as well as industry to:

  • present recent advances in social multimedia data mining and multimedia content analysis
  • present next generation technologies for managing rich social multimedia data, with special emphasis on organizing, indexing, retrieving and mining social multimedia data
  • identify novel applications and key industry drivers

Manuscripts are solicited to address a wide range of topics in social multimedia data mining, including but not limited to the following:

·        Machine learning and data mining methods for social multimedia content

·         Personalization and recommendation algorithms based on social data

·         Social context-based media content analysis

·         Prediction and forecasting models based on social multimedia

·         User profiling across multiple social media channels

·         Event driven media creation

·         Organization, indexing and navigation of social multimedia content

·         Behaviour analysis across multiple social media networks

·         Multi-modality fusion for heterogeneous social media content

·         Large scale image, video and audio classification using social contextual cues

·         Image, video and audio recommendation in social networks

·         Social media-based advertisement

·         Social network enablement via media

Prospective authors should submit high quality, original manuscripts that have not appeared, nor are under consideration elsewhere. All workshop submissions should be formatted following the same guidelines of ICDM'14 conference papers (maximum of 10 pages, in the IEEE 2-column format). Detailed formatting instructions will be available at http://icdm2014.sfu.ca/submission.html. All papers will be reviewed by the Program Committee based on technical quality, relevance to workshop theme, originality, significance, and clarity. The authors can opt for a double blind review, in which case the authors should therefore avoid using identifying information in the text of the paper. All papers should be submitted through here..

To encourage attendance and attract quality submissions, we are considering a special issue of extended versions of selected papers in a suitable journal, a Best Paper cash award, and invited keynotes from academia and industry to stimulate discussions at the workshop.

Important Dates

Submission deadline: August 1, 2014 Extended to August 20, 2014
Decision notification: September 26, 2014
Camera-ready paper due date: October 20, 2014 
Workshop date: December 14, 2014 

Please submit your papers here.


Time & Location: 9:00am to 12:30pm December 14th, 2014, Room: Madrid 6  


Session 1: Opening  and Invited Talk I

9:00am – 9:05am

Welcome and Introduction

9:05am -10:00am

Invited Talk I: Advancing the Frontier of Social Media Mining 

Prof. Huan Liu, Arizona State University

10:00am - 10:10am

Coffee break


Session 2:  Paper Presentation

10:10am - 10:30am

Recommender Systems Using Harmonic Analysis

Gilbert Badaro, Hazem Hajj, Ali Haddad, Wassim El-Hajj, and Khaled Shaban

10:30am - 10:50am

The eyes of the beholder: Gender prediction using images posted in OSNs

Quanzeng You, Sumit Bhatia, and Jiebo Luo

10:50am - 11:10am

Topical Influential User Analysis with Relationship Strength Estimation in Twitter
Xinyue Liu, Fenglong Ma, Hua Shen, and Wenxin Liang

11:10am - 11:30am

Semantic Feature for Food image Recognition with Geo-constraint
Xinhang Song, Shuqiang Jiang, Ruihan Xu, and Luis Herranz


Session 3: Invited Talk II and Closing

11:30am - 12:25pm

Invited Talk II: Supervised Deep Learning with Auxiliary Networks

Dr. Wei Fan, Noah’s Ark Laboratory

12:25pm - 12:30pm

 Discussion & Closing


1. Advancing the Frontier of Social Media Mining


Prof. Huan Liu

Data Mining and Machine Learning Lab

Arizona State University, Tempe, Arizona



Social media mining differs from traditional data mining in many ways, offering unique opportunities to advance data mining. As detailed in our recent textbook “Social Media Mining: An Introduction”, we face challenges such as “the evaluation dilemma”, “sampling bias”, and “the noise removal fallacy”. We will present these challenging problems as well as a recent research issue we encounter - a big-data paradox unique to social media where many social networking sites are present but only minimum information is available. We will exemplify the intricacies of social media data, and show how to exploit unique characteristics of social media data in developing novel algorithms and tools for social media mining. A free download of the textbook can be accessed via the speaker’s homepage.


Dr. Huan Liu is a professor of Computer Science and Engineering at Arizona State University. He obtained his Ph.D. in Computer Science at University of Southern California and B.Eng. in EECS at Shanghai JiaoTong University. He was recognized for excellence in teaching and research in Computer Science and Engineering at Arizona State University. His research interests are in data mining, machine learning, social computing, and artificial intelligence, investigating problems that arise in real-world applications with high-dimensional data of disparate forms. His well-cited publications include books, book chapters, encyclopedia entries as well as conference and journal papers. He serves on journal editorial/advisory boards and numerous conference program committees. He is a Fellow of IEEE and a member of several professional societies.

2. Supervised Deep Learning with Auxiliary Networks

Dr. Wei Fan

Associate Director of Noah's Ark Laboratory

Huawei, Hong Kong


Deep learning well demonstrates its potential in learning latent feature representation and have shown great success in computer vision, voice recognition and machine translation. Recent years have witnessed an increasing enthusiasm for regularizing deep neural networks by incorporating various side information, such as user-provided labels or pairwise constraints. However, the effectiveness and parameter sensitivity

of such algorithms have been major obstacles for putting them into practice, i.e., it has been shown that exiting frameworks that incorporate supervision actually have worse performance than unsupervised pre-training.

The major contribution of our work is the exposition of a novel supervised deep learning framework, which distinguishes from two unique traits. First, it regularizes the network construction by utilizing similarity or dissimilarity constraints between data pairs, rather than sample-specific annotations. Such kind of side information is more flexible. Secondly, unlike prior works, our proposed algorithm decouples the supervision information and intrinsic data structure. We design two heterogeneous networks, each of which encodes either supervision or unsupervised data structure respectively.  Our experiments use a number of benchmark datasets. The main claim of this work is that the proposed framework has consistently higher accuracy, either fully supervised or semi-supervised, than unsupervised deep learning model.


Dr. Wei Fan is the associate director of Huawei Noah's Ark Lab.  He received his PhD in Computer Science from Columbia University in 2001. His main research interests and experiences are in various areas of data mining and database systems, such as, stream computing, high performance computing, extremely skewed distribution, cost-sensitive learning, risk analysis, ensemble methods, easy-to-use nonparametric methods, graph mining, predictive feature discovery, feature selection, sample selection bias, deep learning, transfer learning, time series analysis, bioinformatics, social network analysis,  big data processing platform, novel applications and commercial data mining systems. His co-authored paper received ICDM'2006 Best Application Paper Award, he led the team that used his Random Decision Tree method to win 2008 ICDM Data Mining Cup Championship. He received 2010 IBM Outstanding Technical Achievement Award for his contribution to IBM Infosphere Streams. He is the associate editor of ACM Transaction on Knowledge Discovery and Data Mining (TKDD). Since he joined Huawei in August 2012, he has led his colleagues to develop Huawei StreamSMART – a streaming platform for online and real-time processing, query and mining of very fast streaming data,. The StreamSMART team has received the presidential award from Huawei Labs. His current main research mission and interests are to use big data technologies to transform services related to carrier operations.


Program Committee (tentative)

Prakhar Biyani, Pennsylvania State University
Liangliang Cao, IBM Research
Cornelia Caragea, University of North Texas

Pablo Cesar, CWI

Shu-Ching Chen,  Florida International University
Yang Cong, Chinese Academy of Sciences & University of Rochester 

David Crandall, Indiana University

Sujatha Das, Institute for Infocomm Research

Munmun De Choudhury, Georgia Tech

Anlei Dong, Yahoo! Research

Yiannis Kompatsiaris, Information Technologies Institute (CERTH-ITI)

Fei Liu, Carnegie Mellon University

Bo Long, LinkedIn

Debapriyo Majumdar, Indian Statistical Institute

Tao Mei, Microsoft Research

Amrita Saha, IBM India Research Lab

Tong Sun, Xerox Research Centre Webster

Hanghang Tong, City University of New York

Dingding Wang, University of Miami

Lynn Wilcox, FX Palo Alto Laboratory

Changsheng Xu, Chinese Academy of Science

Yanfang Ye, West Virginia University

Yue Zhou, Twitter

Shenghuo Zhu, NEC Research


Quanzeng You, University of Rochester


  • Q: What is the page limit?
    A: The workshop papers have the same page limit as the main conference. 
  • Q: I have a recent paper published elsewhere. Can I submit a short version to the workshop?
    A:  We encourage submissions of new work. Authors are also welcome to submit papers that have been recently published or accepted at another venue, as long as this information is disclosed at the time of submission.
  • Q: Will papers accepted to the workshop be published?
  • A: Workshops papers will be published in the CPS ICDMW 2014 Proceedings.