1. Motivation: Holistic Objects

The term “ObjectAnalytics” is justified whenever an object is to be analyzed as a whole, in particular in cases where you need to analyze events within an object in relation to each other. For example you may want to understand prescribed drugs of a patient relative to the diagnoses the patient has received, or relative to other prescriptions of the patient. Or you may want to understand failure events in a machine relative to previous usage of the machine, and what sequence of actions follows after those failures.

The XOE implements an interactive usability concept which emphasizes on that sort of “holistic analysis”. You may set selections on events in one part of an object and see how events in another part of the object are related to that. Thereby, there are different means to define how selections “propagate” between objects. “Aggregation dimension”, “relative time axis”, “rank aggregations” with fixed and with floating semantics …

… the XOE lets you visually explore and learn about those concept. By exploring features of the XOE you will soon understand motivation behind ObjectAnalytics in general. The XOE is an intuitive way to get accommodated to ObjectAnalytics and learn about what distinguishes ObjectAnalytics from other technologies.

Once you have understood the motivation behind ObjectAnalytics, you can go to the next level and start to use ObjectAnalytics not just in terms of click, drag and drop - you can start developing programmatically against an ObjectAnalytics server. The XOE helps you with that also. In the XOE, you may look up how artifacts (which you might have generated by click, drag and drop) are defined, and copy that definition to your program code. You may run scripts from within the XOE or from within other programming environments (e.g. from within Python). Once you have manipulated a “session” programmatically, you can any time switch back to the XOE, pick up the session and “visually” see the results.

Even if you do lot’s of development programmatically or if you develop your own web-based solution - nevertheless, the XOE will likely be a core tool to support this.

“Its just descriptive analytics” - you might say. Yes it is! Just on a very different level, which lets standard users explore complex data and complex objects.

… and there is a predictive element also. Xplain Causal Discovery is seamlessly embedded into the XOE and into its interactive usability concept. Our goal is to make predictive analytics intuitive and easy to understand, and by that we also want to “de-mystify” it. This simultaneously is the basis to take predictive analytics to a different level, as the algorithms also run on comprehensive objects and with that on data of a much higher complexity. The XOE lets you also easily learn about that.