Problem Description

The Social Media Popularity Prediction task focuses on predicting the impact of sharing different posts for a publisher on social media. Given a photo (a.k.a. post) from a publisher, the goal is to automatically predict the popularity of the photo, e.g., view count for Flickr, Pin count for Pinterest, etc.

The contestants are asked to develop their dynamic prediction models based on the SMPD dataset provided by the Challenge (as training data), plus possibly additional public/private data, to address the given tasks. For the evaluation purpose, a contesting system is asked to produce prediction results of popularity. The accuracy will be evaluated by pre-defined quantitative evaluation. The contestants need to introduce their systems and datasets in the conference.

Evaluation Criteria

By quantitative evaluation, we measure the systems submitted to this challenge on a testing set. We adopt multiple metrics including Spearman’s Rho (SR) and Mean Absolute Error (MAE). The ranking for the competition is based on quantitative evaluation.

Submission

Each submission is required to be formatted in a JSON File as follows.

  {
      "version": "VERSION 1.2",
      "result": [
          {
              "post_id": "post6374637",
              "popularity_score": 2.1345
          },
          ...
          {
              "post_id": "post3637373",
              "popularity_score": 3.1415
          }
      ],
      "external_data": {
          "used": "true",
          "details": "VGG-19 pre-trained on ImageNet training set"
      }
  }