Spot A Bot

Get insights into the bot activity for Twitter trends

86 followers

Get insights into the bot activity for Twitter trends

86 followers

With Spot A Bot, you will be finally able to uncover real bot activity taking place on Twitter trends. Using advanced machine learning techniques we can now shed light on the impact bots have on our trends and how they try to shape the discourse.

Free

Launch tags:Twitter•Artificial Intelligence•Bots

Launch Team

Customer.ioAutomate Messaging Everywhere — Startups Get 12 Months Free

Promoted

Spot A Bot

Maker

📌

Greetings Product Hunt Community! 👋👋 We, the team of Spot A Bot, are excited to finally launch our project here on Product Hunt. We are already looking forward to reading your comments and feedback! 💙 As you may have noticed, there is a major uncertainty regarding bots and their mischievous activities on Twitter. Even Elon Musk is wondering about how many bots are really on his own platform. Tweets generated by bots can be difficult to recognize. Bots that are up to no good use their anonymity to manipulate, spread misinformation or hijack trends. 👾 Considering this we, four students from Austria, decided to take things into our own hands and searched for a way to show the presence of bots for current trends. That's why we developed Spot A Bot, a platform that uses advanced machine learning techniques to analyze currently trending topics on Twitter and identify accounts that are likely to be bots. 🤖 You can explore recent trends and participating accounts and vote on whether they agree or disagree with our classification of accounts as bots. Trending topics change every day, so there is always something new to explore. 🔍 We aim to be as transparent as possible with our results and methods. On Spot A Bot you can retrace the entire process employed by the system. Try it out and see the difference it can make for your social media experience, especially when it comes to identifying bot activity in current twitter trends. Let us know what you think! 💙💬

Report

4yr ago

FriendsQuest

Congrats on the launch! 👍🏼 Bot detection plays a very important role, especially in today's world.. Since when have you been keeping track on bot activity? Where there any interesting findings?

Report

4yr ago

Spot A Bot

Maker

Hey @markus_aly, thank you for your comment! We've been monitoring Twitter bots for the past 6 months. Last November Chinese bots spammed certain Twitter trends to burry news about the COVID-19 protests taking place there. Before official news reports emerged, these Chinese trends had already appeared in our analysis with over 90% bot activity. This was an alarming finding that demonstrated how bots can hijack organic trends.

Report

4yr ago

SphereOne

Really interesting. Private sector would love this. Have you been following any bot networks that you can infer are being orchestrated by any nation state threat actors (like the Lazarus group of our North Korea for example)? Also really cool that you're incorporating feedback loops from users to help refine machine learning algo.

Report

4yr ago

Spot A Bot

Maker

@cjries Thank you for your comment and interest in the topic. We really appreciate it! We have detected bot nets that were engaging in politically sensitive topics. That was the case in November of 2022 when Chinese city name trends were being flooded with inappropriate content, apparently to bury the discussions on those trends concerning the then ongoing COVID-19 protests across China. The trends reached a bot percentage of over 90% across many major Chinese cities. Upon manual review, we found many accounts linked to a private contractor who even advertised their bot net services over platforms such as Telegram. However, we cannot conclusively discern whether this activity was state-sponsored. Nevertheless, this case clearly shows how bots can be used to hijack sensitive trends.

Report

4yr ago

Congratulations on the launch! It's really interesting to be able to look at these statistics. One question that comes to mind is how do you distinguish between good and bad bots?

Report

4yr ago

Spot A Bot

Maker

@intellibill Thanks a lot, we appreciate the feedback! 😊 Regarding good or bad bots - we don’t make a distinction. Our goal is to give our users an unfiltered insight into Twitter bot activity and to show what bot tweets can and do look like. This is why we provide sample tweets for analyzed accounts. We want to empower you to form an informed opinion about what you see on social media.

Report

4yr ago

FriendsQuest

Absolutely loving the idea and design of your website and reports! 🤖 You currently have English as well as German trends in your analysis. Is a support for other languages planned (e.g. French or Spanish)?

Report

4yr ago

Spot A Bot

Maker

@nijo Thanks Jonas, we appreciate the kind words! When determining bot activity, the model works independent of language. It instead focuses more on behavior patterns and tweet similarity. We do plan on adding more trends from other countries in the future, though. So definitely stay tuned for updates! 🙂

Report

4yr ago

Carbone

It would be interesting to get an insight into bots during all-country elections and share it publicly. It's not only "bots" anymore, and it's against reality and truth by spreading false information. Are you impacted by Twitter enforcing API rules recently? We need more tools like "Spot a bot" for all social platforms/media. Definitely love your product @lener ! Keep it on 💪

Report

4yr ago

Spot A Bot

Maker

@steevep Absolutely, these large scale events and the misinformation that comes with them inspired us to approach this project in the first place. We are thinking about taking extra steps to examine specific trends more in-depth in the future, so stay tuned. 🙂 Regarding Twitter API rules, we weren't able to notice anything major different since the recent developments at Twitter HQ. Thanks for the Feedback!

Report

4yr ago