Task

Social Media Popularity Prediction. The overconsumption of online information has its limitations, so online word-of-mouth helps us to efficiently discover emerging topics, interesting news, or new products from the information ocean. Therefore, predicting online popularity became a crucial task for online media, brand owners, social influencers, and individuals.

The task focuses on predicting the impact of sharing different posts for a publisher on social media. Given a multimodal post from a publisher, the goal is to automatically predict the future popularity after the post is publicly shared.

Evaluation

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.

The evaluation will be based on the online evaluation server and platform. We released the training dataset and withhold the testing dataset for evaluation.

Track1 SMP-Image: We will measure the received solutions by the prediction correlation metric Spearman’s Rho (SR) and the error metric Mean Absolute Error (MAE).

Track2 SMP-Video: We will measure the received solutions by the Mean Absolute Percentage Error (MAPE).

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"
      }
  }