Description

* The articles were published by Mashable (www.mashable.com) and their content as the rights to reproduce it belongs to them. Hence, this dataset does not share the original content but some statistics associated with it. The original content be publicly accessed and retrieved using the provided urls.
* Acquisition date: January 8, 2015
* The estimated relative performance values were estimated by the authors using a Random Forest classifier and a rolling windows as assessment method.  See their article for more details on how the relative performance values were set.

Attribute Information:

Number of Attributes: 61 (58 predictive attributes, 2 non-predictive, 1 goal field)

Attribute Information:
    0. url:                           URL of the article (non-predictive)
    1. timedelta:                     Days between the article publication and the dataset acquisition (non-predictive)
    2. n_tokens_title:                Number of words in the title
    3. n_tokens_content:              Number of words in the content
    4. n_unique_tokens:               Rate of unique words in the content
    5. n_non_stop_words:              Rate of non-stop words in the content
    6. n_non_stop_unique_tokens:      Rate of unique non-stop words in the content
    7. num_hrefs:                     Number of links
    8. num_self_hrefs:                Number of links to other articles published by Mashable
    9. num_imgs:                      Number of images
   10. num_videos:                    Number of videos
   11. average_token_length:          Average length of the words in the content
   12. num_keywords:                  Number of keywords in the metadata
   13. data_channel_is_lifestyle:     Is data channel 'Lifestyle'?
   14. data_channel_is_entertainment: Is data channel 'Entertainment'?
   15. data_channel_is_bus:           Is data channel 'Business'?
   16. data_channel_is_socmed:        Is data channel 'Social Media'?
   17. data_channel_is_tech:          Is data channel 'Tech'?
   18. data_channel_is_world:         Is data channel 'World'?
   19. kw_min_min:                    Worst keyword (min. shares)
   20. kw_max_min:                    Worst keyword (max. shares)
   21. kw_avg_min:                    Worst keyword (avg. shares)
   22. kw_min_max:                    Best keyword (min. shares)
   23. kw_max_max:                    Best keyword (max. shares)
   24. kw_avg_max:                    Best keyword (avg. shares)
   25. kw_min_avg:                    Avg. keyword (min. shares)
   26. kw_max_avg:                    Avg. keyword (max. shares)
   27. kw_avg_avg:                    Avg. keyword (avg. shares)
   28. self_reference_min_shares:     Min. shares of referenced articles in Mashable
   29. self_reference_max_shares:     Max. shares of referenced articles in Mashable
   30. self_reference_avg_sharess:    Avg. shares of referenced articles in Mashable
   31. weekday_is_monday:             Was the article published on a Monday?
   32. weekday_is_tuesday:            Was the article published on a Tuesday?
   33. weekday_is_wednesday:          Was the article published on a Wednesday?
   34. weekday_is_thursday:           Was the article published on a Thursday?
   35. weekday_is_friday:             Was the article published on a Friday?
   36. weekday_is_saturday:           Was the article published on a Saturday?
   37. weekday_is_sunday:             Was the article published on a Sunday?
   38. is_weekend:                    Was the article published on the weekend?
   39. LDA_00:                        Closeness to LDA topic 0
   40. LDA_01:                        Closeness to LDA topic 1
   41. LDA_02:                        Closeness to LDA topic 2
   42. LDA_03:                        Closeness to LDA topic 3
   43. LDA_04:                        Closeness to LDA topic 4
   44. global_subjectivity:           Text subjectivity
   45. global_sentiment_polarity:     Text sentiment polarity
   46. global_rate_positive_words:    Rate of positive words in the content
   47. global_rate_negative_words:    Rate of negative words in the content
   48. rate_positive_words:           Rate of positive words among non-neutral tokens
   49. rate_negative_words:           Rate of negative words among non-neutral tokens
   50. avg_positive_polarity:         Avg. polarity of positive words
   51. min_positive_polarity:         Min. polarity of positive words
   52. max_positive_polarity:         Max. polarity of positive words
   53. avg_negative_polarity:         Avg. polarity of negative  words
   54. min_negative_polarity:         Min. polarity of negative  words
   55. max_negative_polarity:         Max. polarity of negative  words
   56. title_subjectivity:            Title subjectivity
   57. title_sentiment_polarity:      Title polarity
   58. abs_title_subjectivity:        Absolute subjectivity level
   59. abs_title_sentiment_polarity:  Absolute polarity level
   60. shares:                        Number of shares (target)

Relevant Papers:

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.

Citation Request:

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.