Comment moderation

How Arwen's AI moderation tech out-performs older techniques

Arwen breaks new ground in moderation technology, using AI technology to radically innovate the process of creating safer and more inclusive communities.


Social media platforms are an integral part of modern communication and have become a crucial platform for businesses and individuals alike. However, with the explosion of social media usage, the challenges of managing content have grown exponentially. In response, social media platforms have developed different strategies for content moderation. This has led to the evolution of moderation technology through three waves.

Wave one - human moderation

The first wave of social media moderation technology involved human moderators manually monitoring content via the platform's interface. This approach was labor-intensive and time-consuming, with moderators tasked with reviewing every piece of content uploaded to the platform. However, this approach proved to be effective in its ability to remove inappropriate content from the platform, including spam and hate speech.

Wave two - keyword moderation

The second wave of social media moderation technology was the introduction of keyword moderation technology. This technology was designed to automatically filter out specific words or phrases deemed inappropriate or offensive. Although it was an improvement over manual moderation, it had its limitations. For instance, keyword moderation struggled to differentiate between different contexts and could flag innocent posts.

Wave three - AI-enabled moderation

The third wave of social media moderation technology is AI-enabled moderation technology. This technology uses artificial intelligence to analyse and detect harmful content. AI-enabled moderation is a game-changer as it can detect and flag inappropriate content in real-time. It can also identify complex patterns and trends, enabling it to detect previously unknown harmful behaviour.

How Arwen has changed the game

Arwen's style of AI-enabled moderation is superior to previous solutions because it accurately detects and removes inappropriate content in real-time. It also detects the tone and context of a post, allowing it to identify sarcasm, irony, or humour that keyword moderation technologies cannot. This reduces the risk of removing posts that do not violate the platform's policies, known as "false positives", making the moderation process more accurate.

Moreover, AI-enabled moderation technology learns and adapts based on all past experiences, improving accuracy over time. This means that it can detect and remove new types of inappropriate content that emerge on different platforms.

Some specific comparisons

Social Listening tools

Social Listening platforms monitor how followers are interacting with your social content through actions (likes, RTs etc) and the sentiment of what they say. They don't have a specific focus on unwanted content and even less of a focus on tackling spam. They also don't include sub-second removal of offensive material, instead focusing on reports that help you understand your community.
 
In comparison, Arwen is a purpose-built AI that monitors each and every comment in full, automatically detecting 24 types of unwanted content across 29 languages. And when Arwen finds a comment that breaks your rules, it instantly hides it. We would say that Arwen is an "Active Social Moderation" tool, not a passive social listening tool.
 

Wave 2 moderation tech

Wave 2 moderation providers usually bring a combination of keyword filtering and human oversight. Keywords are a notoriously crude way to moderate social media, as they can't determine a positive or neutral use of a term eg black, from a toxic one. All comments with the word "black" are hidden, which damages engagement. As a result they tend to rely on costly human moderators to provide a 'second line' review, and to manually correct errors. This raises the overall cost and delays decision making, which is sub-optimal when social media moves at lightning speed.

In comparison, Arwen is a Wave 3 moderation tech solution. We use AI to continually detect 24 types of unwanted content, across 29 languages, in real time. Arwen looks at the entire context of the comment, not just key words. It's comparing each comment to every comment it has ever seen and making a rounded judgment on whether it's toxic, spam or another type of unwanted content. It then hides those offensive comments in under a second, without any need for human oversight. This keeps Arwen commercially competitive and also means we offer unrivalled moderation speeds.

As social media platforms continue to grow, the importance of effective moderation technology cannot be overstated. Arwen's AI-enabled moderation technology is a crucial step in making social media a safer, more inclusive and more enjoyable platform for everyone.

Similar posts

Get notified on new blog posts

Be the first to know about new blog posts and online safety insights.