Ally

Ally uses AI and Natural Language Processing (NLP) to identify the best and worst behaviours in your online communities and conversations. Ally identifies bad actors and helps keep online communities safe and healthy

REQUEST MORE INFO

– Ally allows moderators to understand community behaviour in near real time.
– AI-based automation identifies cause of community disruption, not just the individual words.
– Chat moderation no longer requires reviewing and moderating individual messages.
– For large communities, the number of messages sent per day can easily be in the tens of millions. Ally operates at large scales.
– Ally allows moderators to evaluate behaviour at the conversation and player levels.
– Faster, more effective moderation, means better mental health for your community moderators, as well as more time to nurture your best community members, rather than focusing on your worst.

Peter is a multi-skilled data scientist and engineer who has deep experience in the startup, corporate, and academic worlds. He brings together an academic backgound in neuroscience and artificial intelligence, along with many years of commercial insights into the production of natural language processing systems for social media and chat products. His personal interests include studying propaganda and other forms of disinformation, and generally exploring the flow of information in human social systems.

The key components that power
ALLY

Automated intervention

Automatic interventions via web hooks trigger muting, rule reminders, or suspensions. The most serious abuse is brought to the attention of members of your team who can take additional actions against the worst actors in your community.

Unique mix of AI & Machine Learning

Ally uses a combination of Ai, Machine Learning, Natural Language Processing (NLP), and Natural Language Understanding (NLU), to analyze any message stream. The insights surfaced by Ally have the power to transform the efficiency of your community moderation.

Behavioural predictive analysis

Toxic speech is only one aspect of negative behaviour. The algorithms that define Ally’s classification system have been trained on millions of messages. Ally recognizes and tags those conversational patterns and notify the community managers.

Customisable, contextual & adaptive

Detection algorithms are tailored to reflect your community norms. Ally’s targets and tags both the positive and negative behaviour that impacts your customers most.

Real-time customer intelligence

Don’t let a vocal minority be the whole story. Ally provides in depth analysis of your whole
community and represents the breadth of community sentiment. Bad actors are highlighted as well as the positive influencers that make your community an attractive and pleasant place to be.

Easy-to-use, adaptable & secure

All you need to do is push your message stream to Ally’s API. Ally’s insights into your community can then be accessed in the cloud via a web interface, an API, or deployed onto your own infrastructure via Docker containers.

A look inside the platform

Get a bird’s-eye view of the latest activity in your community. Seeing the changing patterns in the behaviour of your community members will help your moderation team find where to intervene quickly and efficiently.

It is critical to understand the context of a message when determining whether a conversation is positive or negative. Ally shows the whole conversation allowing your moderators to be confident that both Ally’s and your moderator’s interpretation of the participant’s intention is correct. We call these classified conversations “stories”.

With Ally, your moderation team can search and browse messages from the data stream. This lets them discover newly evolving patterns of behaviour and create new stories when needed. If these patterns persist, we can work with your customer service team and software to automate
the classification process.

The actions performed by your moderation team can be audited within Ally itself leaving you confident they are applying your community management guidelines rigorously and consistently.