Earlier quoted context omitted.
Agreed. For better or worse, social networks are the fb marketplace of eyes.
how could this be for better?
How to automate Instagram engagements with computer vision (and get banned)
11–20 of 36 posts
Re: How to automate Instagram engagements with computer vision (and get banned)
#12>A full screenshot is huge. Over 7 million pixels on a typical screen. Hmmm i don't think that's true? 1080p is 2M, 2K is 4M, 4K is 8M. Are we really in an era where full 4K is "typical"? For reading Instagram? I've got 4 screens i use on a daily basis that are all 2K or less. I mean my TV was only 1080p until a few months ago lol. Maybe I'm just slow. Also: you're automating the browser, just make the window smaller…
Re: How to automate Instagram engagements with computer vision (and get banned)
#13e.g. contradictory feeds.
1. motivational post followed immediately be demotivational post iteratively. this will allow people not be stuck in one basin of attraction.
2. wealth flaunting posts followed by poverty posts to elicit a strong contrast and inequality.
3. science posts followed by pseudo-science posts
4. feeds that are generated like DJ sets. instead of playing with music, the feed would play with emotions.
Other such combinations.
Re: How to automate Instagram engagements with computer vision (and get banned)
#14Re: How to automate Instagram engagements with computer vision (and get banned)
#15I appreciate (and encourage) using real life problems as a pretext for researching new topics and developing new skills, but 1. I'm really curious as to what's the desired outcome here? A spambot to flood people's notifications? 2. I'll admit that I have absolutely no idea how Instagram's API layer (and protection) works but wouldn't capturing the HTTP calls a more appropriate and easier approach to take?
2. Leveraging their private APIs will get you banned even quicker than OPs method.
Re: How to automate Instagram engagements with computer vision (and get banned)
#16I tried all sorts of automation in the past and always get banned at the end. Not worth it. Use AI or whatever to generate your social media content but always post it manually for best engagement and not getting banned
Re: How to automate Instagram engagements with computer vision (and get banned)
#17i have no interest in juicing engagement, purely from a "20 years from now it'll be cool to know this existed at all" aspect.
Re: How to automate Instagram engagements with computer vision (and get banned)
#18Re: How to automate Instagram engagements with computer vision (and get banned)
#19I tried all sorts of automation in the past and always get banned at the end. Not worth it. Use AI or whatever to generate your social media content but always post it manually for best engagement and not getting banned
on business pages they literally give you access to the posting API.... And you can automate against that. Not that they will give you much action on that without the sweet sweet accelerant of ad money.
I realized months ago that social networks are distribution systems. Want acceleration? Pay.
There is little incentive to make anything viral/organically boosted.
Re: How to automate Instagram engagements with computer vision (and get banned)
#20I appreciate (and encourage) using real life problems as a pretext for researching new topics and developing new skills, but 1. I'm really curious as to what's the desired outcome here? A spambot to flood people's notifications? 2. I'll admit that I have absolutely no idea how Instagram's API layer (and protection) works but wouldn't capturing the HTTP calls a more appropriate and easier approach to take?
1. Social media accounts with lots of engagement and high follower counts are worth big money these days. However, it takes a ton of effort to build that up organically. It's not difficult to see possible motives for something like this. 2. Leveraging their private APIs will get you banned even quicker than OPs method.
Anything not from their vanilla app, the littlest dot on their charts will trigger severe alarms and actions.
Big correlation systems. The safest path used to be to automate the app itself, through mobile automation, but they even got too sensitive to that.
I don't doubt the whole app has a behavioural analysis component, full screen size, much like a big "I am not a robot" checkbox.
Also, it's very likely their private APIs are CSRF-protected or similar.