Botometer Pro
9.2 / 10
1,105ms
100%
N/A
Botometer Python Research Project for PhD
I am a fourth year Economics Ph.D. student at Binghamton University. I have historical Twitter data from 2016 that I plan to use Botometer on to investigate which tweets were potentially posted by bots. This investigation will potentially be part of a promising research project.
I have successfully had Botometer Python to work, and it runs fairly smoothly. The problem is that I have a very large set of tweets. I have around 200 million (but I can probably do something meaningful with around 3-million not less). At the speed at which I had this working, running the program on 3-million tweets will take months (about 10-15 tweets/minute).
I am curious if this is an issue with the rate-limit on the API? Is it possible to increase that at a rate that serves my need. Could you please advice on what is the best way to use Botometer on such a large data-set?
Join in the discussion - add comment below:
How does one go about analyzing historical Twitter data with Botometer? I have scraped data using TAGS in Google sheets and I need to identify which accounts are operated by bots. Can I use the Botometer API in Python this way?
Hi, if you are using the Botometer Pro API , the rate limit for the free tier matches the rate limit of the Twitter API with user authentication, and the rate limit of the paid tier matches the rate limit of the Twitter API with app authentication. Assuming you need to use the Twitter API to get data to pass to the Botometer API, that is a hard limit for you. If instead you have access to high volume Twitter data from GNIP, then you could go beyond these limits. For such cases, it is possible to license the Botometer software. However this is both more complicated and more expensive. If you are interested, we can put you in touch with the Indiana University Office of Commercialization, which handles licensing. HTH.