Page MenuHomePhabricator

[Spike] NLP for article quality models.
Open, LowPublic

Description

We should be able to get some more signal by looking for obvious grammatical mistakes and other issues. We should also consider using our dictionary-based features.

Ideas:

  • Readability metrics like (flesch kincaid)
  • Grammatical mistakes scanning
  • Sentence length

This task is done when we have a list of available libraries with a cursory assessment of their quality and utility. If applicable, new tasks are filed for taking advantage of them.

Event Timeline

Halfak created this task.Apr 12 2016, 11:49 PM
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptApr 12 2016, 11:49 PM
Halfak assigned this task to schana.Jul 26 2016, 2:40 PM
Halfak updated the task description. (Show Details)
Halfak triaged this task as Low priority.Aug 4 2016, 2:29 PM
Halfak renamed this task from NLP for article quality models. to [Spike] NLP for article quality models. .Aug 16 2016, 4:42 PM
Halfak reassigned this task from schana to Ladsgroup.
Halfak added a subscriber: schana.
Restricted Application added a project: User-Ladsgroup. · View Herald TranscriptAug 16 2016, 4:42 PM
Ladsgroup removed Ladsgroup as the assignee of this task.Jan 25 2019, 12:32 PM
Ladsgroup removed a project: User-Ladsgroup.
Ladsgroup added a subscriber: Ladsgroup.
Restricted Application added a project: artificial-intelligence. · View Herald TranscriptJan 25 2019, 12:32 PM
Harej moved this task from Ideas to Epic on the Scoring-platform-team board.Apr 3 2019, 4:45 AM