Bot or Not? Automated Accounts

With all the controversy on Facebook and Twitter lately of accounts being used for inauthentic behavior and potential manipulation - I thought it might be interesting to take a look at how my bots stand up. The services created by Indiana University Bloomington, primarily Bot or Not, are being leveraged by social platforms to help identify accounts that are not real human beings. The service has mixed results for my accounts. It makes me wonder what sort of quality control is in place to identify the bots versus accounts that are just weird or unusual. The Bots: @Malinda_Lloyd My primary Twitter account, malinda_lloyd, is almost entirely generated from cross platform publishing. When I post to Instagram, I also publish to my twitter feed. I retweet a lot of cute pictures and videos of animals too. As I suspected they might, Bot or Not thinks this feed is managed by a human. @anime_girlie_ My secondary account for this site and all the Japanese inspired cuteness I find in Los Angeles, was slightly less correct. The anime_girlie_ account is 100% automated and while less bot-like than some of my feeds, it's still automated. I use the story bench tweeter app to publish tweets twice a day and auto publish from Instagram. I wrote my preset tweets like I speak (with hashtags) so they appear less bot-like than other feeds. @fabric_lover My fabric_lover twitter feed has been broken, off and on, for the past few months. I cannot get the story bench tweeter app to start working again (as you can see from this thread). This account is 99.9% automated; I preschedule the posts and forget about the feed until a reminder goes off to refresh the content. This feed determination is correct. @MaryShelleyFR The feed results I found most humorous is the determination for Shelley Bot! The feed uses an auto tweeter to post content. And Bot or Not has it misidentified as maybe not a bot. This feed is absolutely a bot! The content of the feed is nothing more than snippets of Mary Shelley's Frankenstein. I do not control what is tweeted except: The parameters I input at creation The book it tweets from The analyzer pegged this feed as less bot-like than @fabric_lover. The language may be more natural because it is from a well known book. I am not totally satisfied that these platforms are a good way to certainly identify automation. What business doesn't use prescheduled posts and canned responses in today's environment? None. They all do it but how should classify those feeds? What do you think of the results? Are they clear cut identifiers or do you think more human review is needed before a feed is considered a "bot?"

ANIMEGIRLIE

Life of a nerd, tech junkie & anime fan

Tag: technology

Notice for AdBlock users

Share this:

Notice for AdBlock users