[new model] Identify spam comments #3994
Open
+581
−0
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
marco-c
reviewed
Jan 24, 2024
jpangas
commented
Jan 29, 2024
Member
suhaibmujahid
left a comment
suhaibmujahid
left a comment
There was a problem hiding this comment.
Thank you, @jpangas! Please see my comments.
Comment on lines
131
to
154
| class UnknownLinkAtBeginning(CommentFeature): | ||
| name = "Unknown Link found at Beginning of the Comment" | ||
|
|
||
| def __init__(self, domains_to_ignore=set()): | ||
| self.known_domains = domains_to_ignore | ||
|
|
||
| def __call__(self, comment, **kwargs): | ||
| urls = extract_urls_and_domains(comment["text"], self.known_domains)["urls"] | ||
|
|
||
| words = comment["text"].split() | ||
| return words[0] in urls if words else False | ||
|
|
||
|
|
||
| class UnknownLinkAtEnd(CommentFeature): | ||
| name = "Unknown Link found at End of the Comment" | ||
|
|
||
| def __init__(self, domains_to_ignore=set()): | ||
| self.known_domains = domains_to_ignore | ||
|
|
||
| def __call__(self, comment, **kwargs): | ||
| urls = extract_urls_and_domains(comment["text"], self.known_domains)["urls"] | ||
|
|
||
| words = comment["text"].split() | ||
| return words[-1] in urls if words else False |
Member
There was a problem hiding this comment.
Instead, we could return the index for the start of the first link and the index for the end of the last link.
@jpangas wdyt?
Collaborator
Author
There was a problem hiding this comment.
This sounds good. Let me try it out and I will share the results.
Collaborator
Author
|
This is currently blocked by #4097 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes #3377
This PR introduces the
spamcommentmodel which is trained to identify spam comments. Later, it will be used to tag potential spam comments.The model is a work in progress and still needs some features to be added, refined and experimented.
Train on Taskcluster: spamcomment