How to interpret the result?

[Gautamshahi]

{ "cap": { "english": 0.0011785984309163565, "universal": 0.0016912294273666159 }, "categories": { "content": 0.058082395351262375, "friend": 0.044435259626385865, "network": 0.07064549990637549, "sentiment": 0.07214003430676995, "temporal": 0.07924665710801207, "user": 0.027817972609638725 }, "display_scores": { "content": 0.3, "english": 0.1, "friend": 0.2, "network": 0.4, "sentiment": 0.4, "temporal": 0.4, "universal": 0.1, "user": 0.1 }, "scores": { "english": 0.0215615093045025, "universal": 0.0254864249403189 }, "user": { "id_str": "1548959833", "screen_name": "clayadavis", "...": "..." } }

From the result, how to interpret the results? Which field indicates the bot? Can you please provide any description?

[mihirp161]

display_score -> english. That is a overall score, such that the higher the score, more chance for that twitter user to be a bot.

[Gautamshahi]

Thanks, What is the meaning of cap? In https://github.com/IUNetSci/botometer-python/issues/35, Clayton says cap, I read the publication, but it's not mentioned.

Which is more suitable display_score or cap?

On Sat, May 2, 2020 at 12:21 AM Mihir Patel notifications@github.com wrote:

display_score -> english. That is a overall score, so if it is higher than it's a bot.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/IUNetSci/botometer-python/issues/42#issuecomment-622590886, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA6IKFTI5ET3SBWOHQMQDULRPNDOPANCNFSM4MR5CKTQ .

-- Gautam Kishore Shahi,

[mihirp161]

Sorry for late reply. Anyways, I would say display_score if your content is exclusively English language, else universal is fine. clayadavis said it too, and that's what (meaning display_score) I use when doing bot analysis. The documentation is poor for this package, not doubt. Thing's became even crazier when mashup changed to rapidapi. This move broke a lot of R packages that depends on this on background. So I am here, opening up issue lol.

Cap (Complete Automation Probability) scores include probabilities and they are multiplied by 100 on display_scores and scores respectively.

Anyways, go with display score Gautam, that's my recommendation. Cheers!

[yang3kc]

Thanks opendatasurgeon for the help!

We have some FAQs for these questions: https://botometer.iuni.iu.edu/#!/faq#bot-threshold https://botometer.iuni.iu.edu/#!/faq#what-is-cap

Hope they help.

[Gautamshahi]

Thanks, @Mihir for the suggestion.

On Sun, May 3, 2020 at 12:05 AM Kaicheng(Kevin) Yang < notifications@github.com> wrote:

Thanks opendatasurgeon for the help!

We have some FAQs for these questions: https://botometer.iuni.iu.edu/#!/faq#bot-threshold https://botometer.iuni.iu.edu/#!/faq#what-is-cap

Hope they help.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/IUNetSci/botometer-python/issues/42#issuecomment-623021542, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA6IKFUOFJCJW5PU76GFOKDRPSKL5ANCNFSM4MR5CKTQ .

-- Gautam Kishore Shahi,