• Terms of Use
  • Privacy Policy
  • Contact
  • EnglishEnglish
    • EnglishEnglish
    • हिन्दीहिन्दी
    • PortuguêsPortuguês
    • Bahasa IndonesiaBahasa Indonesia
    • 中文 (中国)中文 (中国)
Sunday, June 8, 2025
  • Home
  • AI
  • Gaming
  • Health
  • Robotics
  • Security
No Result
View All Result
  • Home
  • AI
  • Gaming
  • Health
  • Robotics
  • Security
No Result
View All Result

How we are using machine learning to detect GOV.UK feedback spam

December 7, 2022

Watch On:

Summary

User feedback is one of the most direct ways that the Government Digital Service learns about user experience. These responses can dilute our insights, cause security concerns, and prevent real problems from being identified. There was also a security risk to consider, as individuals could attempt to negatively exploit feedback mechanisms to disrupt the usual workings of GOV.UK. With ML, we could use the probability score to demand a high level of confidence in our model’s predictions, reducing the mislabelling of legitimate feedback. Once we had made the decision, we focused on the tools and techniques that would help deliver a working solution as soon as possible. We then ranked individual feature importance to assess the features used in the model’s classification judgements, to understand which derived features had the most impact on the outcome of the model’s predictions. Tools such as PyCaret and DVC meant that we were able to focus on deploying a working solution at pace. We can run it on over a month’s worth of feedback data –around 40,000 responses– in less than five minutes, a fraction of the time it takes human reviewers. Now, a careful iteration process is important for combating spammers that adjust terminology to outwit filters and cause “model drift”.

Show Notes

Unfortunately, this also creates a lot of avenues for spam responses.
The problem with GOV.UK feedback spamAt GOV.UK, we received around 540,000 feedback responses from the public and other departments in 2021.
In early 2022, we saw spam responses surge due to a technical change on the front end, peaking at 12% of total feedback.
Why we used machine learningWe quickly recognised that colleagues needed to derive their insights quickly, without needing to manually filter out spam.
We saw the challenge as a great application for a machine learning (ML) model.

Source

https://dataingovernment.blog.gov.uk/2022/10/03/how-we-are-using-machine-learning-to-detect-gov-uk-feedback-spam/

Tags: emailGovernmentlearningspamtoolstrainingwebsite
Previous Post

BT Group Launches New Internal ML Operations Platform ‘AI Accelerator’

Next Post

A Prodigal Son Returns

Next Post

A Prodigal Son Returns

Latest

Artificial Intelligence and Machine Learning Empowers Healthcare in China

January 9, 2023

Play-to-Earn Crypto Games: What to Know

January 9, 2023

Mobile Gaming Platform BlueStacks Offers its Affiliate Program for Publishers Across 57 Countries

January 9, 2023

Netflix is ​​the #1 streamer for the most in-demand video game adaptations

December 29, 2022

Magic: The Gathering is getting Final Fantasy and Assassin’s Creed cards

December 29, 2022

American Business Council and Others gear up for the 2022 Cybersecurity Conference.

December 29, 2022
Load More
Currently Playing
    TikTok Youtube Instagram LinkedIn Twitter Facebook

    Facebook

    Instagram

      The Instagram Access Token is expired, Go to the Customizer > JNews : Social, Like & View > Instagram Feed Setting, to refresh it.

    Recent News

    Artificial Intelligence and Machine Learning Empowers Healthcare in China

    January 9, 2023

    Play-to-Earn Crypto Games: What to Know

    January 9, 2023

    Mobile Gaming Platform BlueStacks Offers its Affiliate Program for Publishers Across 57 Countries

    January 9, 2023

    © 2022Tech Flash News

    No Result
    View All Result
    • Home
    • AI
    • Gaming
    • Health
    • Robotics
    • Security
    • हिन्दीहिन्दी
    • PortuguêsPortuguês
    • Bahasa IndonesiaBahasa Indonesia
    • 中文 (中国)中文 (中国)

    © 2022Tech Flash News

    This website uses cookies. By continuing to use this website you are giving consent to cookies being used. Visit our Privacy and Cookie Policy.
    -
    00:00
    00:00

    Queue

    Update Required Flash plugin
    -
    00:00
    00:00