Loading
Feedback

πŸ›  Contribute : Found a typo? Or any other change in the description that you would like to see ? Please consider sending us a pull request in the public repo of the challenge here.

πŸ•΅οΈ Introduction

What does it take to go viral on social media ? We give you the features of news articles. Predict the number of 'share' it gets on social platforms.

Understand with code! Here is getting started code for you.πŸ˜„

πŸ’Ύ Dataset

The dataset contains 61 columns.
The goal is to predict shares. Thus remaining 58 columns are used for prediction. Some of the columns are as follows - + n_tokens_title - Number of words in the title + num_imgs - Number of images present in the blog + average_token_length - Average length of the words in the content

More info is contained in dataset_info.txt

Files

Following files can be found in resources section

  • train.csv - (26715 samples) This csv file contains the attributes describing the blog written along with share count .
  • test.csv - (13082 samples)File that will be used for actual evaluation for the leaderboard score.

πŸš€ Submission

  • Prepare a CSV containing header as share and predicted value as share count respectively denoting the number of shares it will recieve.
  • Name of the above file should be submission.csv.
  • Sample submission format available at sample_submission.csv in the resorces section.

Make your first submission here πŸš€ !!

πŸ–Š Evaluation Criteria

During evaluation MAE or Mean Absolute Error will be used for accuracy,

For secondary score we use RMSE or Root Mean Squared Error

πŸ”— Links

πŸ“± Contact

Pulkit Gera

πŸ“š Refrences

  • K. Fernandes, P. Vinagre and P. Cortez. A Proactive Intelligent Decision Support System for Predicting the Popularity of Online News. Proceedings of the 17th EPIA 2015 - Portuguese Conference on Artificial Intelligence, September, Coimbra, Portugal.

Participants

Getting Started