How to use code JAVA NLP information to delete NLP depth algorithm recommendation

news96 · November 2, 2020, 2:12pm

I am an old user and I am very happy to use newsblur. Now I have encountered a lot of source duplicate information. How to use newsblur to support Java to implement LPN and duplicate information filter🙏

samuelclay · November 5, 2020, 4:30pm

That’s a tall order to ask of that dialog box. JavaScript is intended to be used if you want to change some small things about the interface and have the chops to do it. You would code it up in the Console of your browser’s debugger and then copy and paste it into this box to save it for every session. Sort of a monkey-patching for the web.

What you’re asking is to use a different language, which is compiled and needs to be run on a host machine, and it have it perform actions that honestly can only be done on the backend.

I’d love to be proven wrong and somebody could show a JavaScript algorithm that compares two stories (by title, author, tags, content, permalink) and can “de-dupe” them, even if they have differing titles, permalinks, etc, but are 99% the same. I already do this on the backend, but sometimes duplicates come through.

Here’s the code I wrote to do the de-duplication:

github.com

samuelclay/NewsBlur/blob/5813e80a50b75e99e84bcdd816679ac4432d6197/apps/rss_feeds/models.py#L1989-L2089


def _exists_story(self, story, story_content, existing_stories, new_story_hashes):
    story_in_system = None
    story_has_changed = False
    story_link = self.get_permalink(story)
    existing_stories_hashes = existing_stories.keys()
    story_pub_date = story.get('published')
    # story_published_now = story.get('published_now', False)
    # start_date = story_pub_date - datetime.timedelta(hours=8)
    # end_date = story_pub_date + datetime.timedelta(hours=8)


    for existing_story in existing_stories.values():
        content_ratio = 0
        # existing_story_pub_date = existing_story.story_date
        # print 'Story pub date: %s %s' % (story_published_now, story_pub_date)


        if isinstance(existing_story.id, unicode):
            # Correcting a MongoDB bug
            existing_story.story_guid = existing_story.id
        
        if story.get('story_hash') == existing_story.story_hash:

This file has been truncated. show original

news96 · January 30, 2021, 2:57pm

Thank you, I suggest you provide a news library and some social media connections.Users can directly read the content of the following users with their own accounts.