state-op-detector / README.md
curt-tigges's picture
Added contact info
9c50f3f
metadata
language:
  - en
tags:
  - classification
license: apache-2.0
widget:
  - text: >-
      Zimbabwe has all the Brilliant Minds to become the Next Dubai of Africa No
      wonder why is so confide | Invest Surplus yako iye into Healthcare that
      will save lives amp creat real Jobs in Healthcare Sector | To the African
      Diaspora in Americas Europe AsiaUK this i | If the Dictatorship in dnt see
      this vision Zambia can still impment the ideaamp have Zambia as the |
      Ndeyake Zimbabwe anoita zvaanoda nayo Momudini He Died for this countr | 
    example_title: Normal Tweet Sequence 1
  - text: >-
      Militants from Turkey to stay alive are preparing to leave settlements
      without a fight and surrender their weapons | Over thousand proTurkish
      militants left occupied territories in the southwestern part of Idlib | In
      early February during the humanitarian campaigns in the points of Mazlum
      of the province of Deir EzZor and Hazze of the province of Rif Damascus
      food packages were issued | Humanitarian assistance from the Russian
      military came from the first days of the Russian participation in the
      Syrian operation | Local residents of Khsham received food packages with a
      total weight of tons | The Russian Center for Reconciliation of the
      Parties held a humanitarian action in the village of Hsham DeirezZor
      Province | After asking for help from representatives of the Antiochian
      church the Russian reconciliation center delivered about two tons of food
      to Mahard | After mortar shelling of the village of Mahard residents were
      in a difficult humanitarian situation | The Russian military held a
      charity event in Maharde Aleppo Province handing out packages of products
      weighing more than two tons in total | The Russian military delivered a
      batch of humanitarian aid to the Christians of the city of Mahard in the
      Syrian province of Hama |
    example_title: Russian State Operator Tweet Sequence
  - text: >-
      Peru will sign a memorandum of understanding to join Chinas Belt and Road
      infrastructure initiative in coming days Chinas ambassador said on
      Wednesday despite recent warnings from the United States about the
      Beijings rise in Latin America peru latinamerica china us usa | People in
      Washington waved Turkish flags and displayed posters We remember victims
      of Armenian terror Stop Armenian aggression on Azerbaijan while Armenians
      held banners and chanted against Turkey over the socalled Armenian
      genocide armenia turkey | Putin reportedly told Kim that Russia supported
      his efforts to normalise North Koreas relations with the US KimJongUn
      putin northkorea russia | Israeli Prime Minister Benjamin Netanyahu said
      he will name a new settlement in the Golan Heights after US President
      Donald Trump israel us usa golan GolanHeights | Turkish FM urges sincerity
      in counterterrorism Mevlut Cavusoglu reiterates Turkeys fight against all
      terror groups of FETO PKK YPG Daesh turkey feto pkk ypg daesh | Sri Lanka
      declares April national day of mourning Decision taken during meeting of
      National Security Council chaired by Sri Lankan President Maithripala
      Sirisena SriLankaBlast SriLankaBombings SriLankaTerrorAttack | Kazakh
      President KassymJomart Tokayev on Tuesday secured veteran leader Nursultan
      Nazarbayevs backing to run in the June snap presidential election
      virtually guaranteeing Tokayevs victory nazarbayev tokayev Kazakhstan |
      Sri Lanka wakes to emergency law after Easter bombing attacks
      SriLankaBlasts SriLankaBombings SriLankaTerrorAttack SriLankaBlast |
      Libyan govt claims control of most of Tripoli airport Development follows
      clashes with forces loyal to eastern Libyabased commander Khalifa Haftar
      libya khalifahaftar tripoli haftar | Death toll from Philippine quake
      rises to Search and rescue work continues for people buried under
      supermarket that collapsed early Monday philippine quake | 
    example_title: Chinese State Operator Tweet Sequence
  - text: >-
      You live in a fantasy world Tim The real insurrection was a stolen
      election where the will of | Canada isnt importing drugs and slaves Just
      Globalism and oppression of its people | Our systems are corrupted Who is
      trying to fix them and rectify the immense damage this illegitimate
      administratio | Just the cars that run people down while killing the
      planet Maybe these ga | But teaching them that they can chop their Johnson
      off and be a girl in prek thats not Obscene | If youve been wondering if
      there is anyone or anything that Washington holds in higher contempt than
      Russia and Vl | Thanks Joe You SUCK | It seems like lately the right has
      been focused on protecting rights while the left is focused on leaving you
      with nothing left | This makes me a proud American | Youre welcome We are
      here and have been | 
    example_title: Normal Tweet Sequence 2

State Social Operator Detector

Overview

State-funded social media operators are a hard-to-detect but significant threat to any democracy with free speech, and that threat is growing. In recent years, the extent of these state-funded campaigns has become clear. Russian campaigns undertaken to influence elections are most prominent in the news, but other campaigns have been identified, with the intent to turn South American countries against the US, spread disinformation on the invasion of Ukraine, and foment conflict in America's own culture wars by influencing all sides as part of an effort to weaken America's hegemonic status.

Iranian and Chinese efforts are also well-funded, though not as widespread or aggressive as those of Russia. Even so, Chinese influence is growing, and often it uses social media to spread specific narratives on Xinjiang and the Uyghur situation, Hong Kong, COVID-19, and Taiwan as well as sometimes supporting Russian efforts.

We need better tools to combat this disinformation, both for social media administrators as well as the public. As part of an effort towards that, we have created a proof-of-concept tool that can be operated via browser extension to identify likely state-funded social media operators on Twitter through inference performed on tweet content.

The core of the tool is a DistilBERT language transformer model that has been finetuned on 250K samples of known state operator tweets and natural tweets pulled from the Twitter API. It is highly accurate at distinguishing normal users from state operators (99%), but has some limitations due to sampling recency bias. We intend to iteratively improve the model as time goes on.

Usage

You can try out the model by entering in a sequence of 1-10 tweets. Each should be separated by pipes, as follows: "this is tweet one | this is tweet two." The model will then classify the sequence as belonging to a state operator or a normal user.

Further Information

You can obtain further information on the data collection and training used to create this model at the following Github repo: State Social Operator Detection

Contact

You can reach me at projects@curttigges.com.