Make faster, more accurate decisions

Our analytics framework gives you a simple way to use the context of resolved entities and relationships in scenarios, rules and models.

 

It builds on open interfaces to familiar tools and allows you to leverage existing models as well as get uplift from new features using techniques you already use. It provides models that are transparent, easy to deploy and to integrate into live services.

How context improves your analytics process

  •  Create entities and networks to build rich customer context that describes real-world relationships – providing more than just a graph of your raw data.
  •  Use this rich network to easily create the analytical components you need – whether they are simple or complex.
  •  Benefit from the full power of AI and machine learning with our best practice methodology, whilst still maintaining open and transparent models. Make them fully accessible to your end users and easily explained in a governance process.
  •  Empower your data scientists with network visualization to understand the context they’ve created, explain the results and validate that their model is working.
Analytics framework

The benefits of our analytics framework 

Our framework gives you complete ownership all of your analytics.

Total flexibility to manage analytical dependencies within complex flows of logic.

Data scientists can see why models are behaving the way they are, with fully transparent and accessible models.

Investigators can easily understand the decision-making process and each decision made by the model.

Adhere to strict governance and process requirements with a systematic and auditable approach for putting models into production.

Get stronger predictions by leveraging entities and networks to create context-based models.

Use open interfaces to the machine learning libraries you are familiar with.

Leverage behavioral profiling and dynamic segmentation for additional context.

Leveraging the expert knowledge within your organization using custom rules and features describing intuitive concepts.

Now, see how to visualize, understand and use the context you create

Learn more

Supporting the journey to supervised machine learning

 

Supervised machine learning (also known as predictive) uses historic cases to statistically identify patterns. However, certain use cases, including many anti-money laundering, have very few previously known examples.

 

Our proven methodology supports the journey to a supervised model for these use cases.

 

1

2

3

Expert driven scorecards

Using joint knowledge, Quantexa identifies the patterns expected to relate to interesting cases and aggregate the risk through a proven approach of building an expert-driven scorecard. This provides investigators with the cases they expect to find useful. Network features provide coverage even of less well understood risks.

2

Analytically tuned expert scorecards

With a reasonable number of alerts reviewed and many escalated, weights and thresholds can be tuned. Quantexa augments the business knowledge with reports on the observed outcomes, resulting in improved hit rates.

3

Supervised machine learning model combined with expert scorecard

When enough alerts have been reviewed and hundreds of escalations achieved, machine learning models can be trained. This optimizes alerting for hit rate, while still leveraging expert knowledge to cover all risks (including rare risks which methodologies relying solely on machine learning will miss).

Analytics brief

Using AI and machine learning with Quantexa

Find out how to you can leverage AI and machine learning to create the context needed to make better decisions – while maintaining an explainable and transparent approach.

 

Download now

Using AI and machine learning with Quantexa

Enhance the performance of your analytics framework

Speak to our team and learn how our platform lets you use context to improve your analytics processes and increase the value of data science in your organization.