SMHP Challenge News: Challenge dataset was released at Here!

OVERVIEW

On social media, authors often want to see their information shared more widely, potentially reaching thousands of readers in a short amount of time. For examples, Washington Post in 2017 started to look for ways to predict which stories will become popular and help editors to more efficiently allocate resources to support a better reading experience. Headline prediction is thus one of the most desired and powerful tools to crack the secret of popular social media content.

Social media today, with plenty of recorded user preferences and social behaviors, offers a rich online source for learning and predicting future headlines from the history data. Therefore, our challenge task is designed to get a deeper understanding of the temporal evolution of social media headlines. Given sequential and continuous post sharing data before a specific date from social media, the problem is to predict how popular of new posts will be for the near future and which would be the headlines?

As a continuous activity of the Social Media Prediction (SMHP) Challenge in ACM Multimedia 2017, the joint research team from the Academia Sinica (AS) and JD.com is running a new edition of the prediction challenge for social multimedia understanding and predictions, namely Social Media Headline Prediction (SMHP) Challenge in ACM Multimedia 2018, with over 340K posts and 80K users data in total. Different from existing social media datasets in the literature, ours is a time-ordering dataset to particularly capture the temporal dynamics of social media data over time to thoroughly represent the social media “world”.

PARTICIPATION

The Challenge is a team-based contest. Each team can have one or more members, and an individual can not be a member of multiple teams. At the end of the Challenge, all teams will be ranked based on both objective evaluation and human evaluation. The top performing teams will receive award certificates and/or cash prizes. At the same time, all accepted submissions are qualified for the conference’s grand challenge award competition.

CITATION

If you intend to publish results that use the information and resources provided by this challenge, please include the following references:

      @inproceedings{Wu2017DTCN,
  title={Sequential Prediction of Social Media Popularity with Deep Temporal Context Networks},
  author={Wu, Bo and Cheng, Wen-Huang and Zhang, Yongdong and Qiushi, Huang and Jintao, Li and Mei, Tao},
  booktitle={International Joint Conference on Artificial Intelligence (IJCAI)},
  year={2017},
  location = {Melbourne, Australia}}
  @inproceedings{Wu2016TemporalPrediction,
  author = {Wu, Bo and Mei, Tao and Cheng, Wen-Huang and Zhang, Yongdong},
  title = {Unfolding Temporal Dynamics: Predicting Social Media Popularity Using Multi-scale Temporal Decomposition},
  booktitle = {Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI)}
  year = {2016},
  location = {Phoenix, Arizona}}
  @inproceedings{Wu2016TimeMatters,
  author = {Wu, Bo and Cheng, Wen-Huang and Zhang, Yongdong and Mei, Tao},
  title = {Time Matters: Multi-scale Temporalization of Social Media Popularity},
  booktitle = {Proceedings of the 2016 ACM on Multimedia Conference (ACM MM)},
  year = {2016},
  location = {Amsterdam, The Netherlands}}
      

TIMELINE

Unless otherwise stated, all deadlines are at 23:59 Anywhere on Earth (AoE), UTC-12.
  • April 15, 2018

    Dataset available for download (training set)

  • June 1, 2018

    Test set available for download

  • June 18, 2018

    Results submission

  • June 20 - 25, 2018

    Objective evaluation and human evaluation

  • June 25, 2018

    Evaluation results announce

  • July 8, 2018

    Paper (all tasks) submission deadline (please follow the instructions on the main conference website)

  • September 5, 2018

    Dataset released (training and testing dataset)

CONTACTS

  • wenhuangcheng
  • wubo
  • meitao

SPONSOR