At the heart of our explainable AI approach is an automated process that extracts statistically well-defined groups of data items, called patterns. A pattern is defined by a set of data items that perform unusually high or unusually low in terms of a user-specified target variable and at the same time are defined by a set of common feature value ranges.
For example, a pattern might be a set of investments that have similar dividend yields and are in the same set of sectors. The pattern browser dashboard on the left allows analysts to explore these patterns in terms of the various data attributes or features, and learn about the most important data features. Read more about it here.
We recently applied our software to predict the risk of certain US counties to experience larger than usual COVID-19 mortality. We designed the COVID-19 Risk Explorer shown on the left to allow users to explore the counties and their risk patterns, if any.
To start exploring the risk patterns of your county simply click here and a new tab or window with our dashboard will open. You can read more about the dashboard here.
AKAI KAERU – a quick Look under the hood
Akai Kaeru Simplifies Data Analysis by Combating Information Overload
A common headache for data analysts is how to pick a manageable set of meaningful attributes and features from a list of hundreds or more. Akai Kaeru's new approach to data analysis puts an end to this guessing game.
Akai Kaeru’s Pattern Miner automatically decomposes the data into a manageable set of statistically robust data patterns. These are groups of data items where each can be concisely described with just a few attributes.
Next, Akai Kaeru's Visual Causal Analyst distills correlations that exist among these patterns into a terse set of causal relations. This process eliminates all spurious correlations and makes it easy to discern the true interactions that exist in the data. This is illustrated in the figure on the right.
Finally, two visual interfaces are available for analysts to visualize both patterns and relations in an integrated and intuitive fashion: the Data Context Map visual layout as well as the Pattern Browser visualization dashboard. Analysis can freely interact with these two visual interfaces to explore the patterns and the relations among them.
Akai Kaeru Enables Extra Insight by Inviting the Human into the Analytics Loop
Humans are still far superior to robotic data analysis, especially when transforming creative hunches into novel strategies. Let our software do the heavy lifting. Channel its work to help you in carving out new actionable insight. Make use of its wide set of interactive visual tools to inspect and tune the data patterns and test new relations derived from them.
See the Product page to learn more.
