Using the app is very simple. Users just need to upload the data and define the target. The attractive user interface guides users step by step through the definition of the analysis. When all the details have been provided, the data is analyzed quickly and fully automatically. The result is provided in the form of a report in HTML format.
Users benefit in particular from the fact that no data preparation is required in terms of data cleaning and synchronization or pre-selection of relevant data. Once the data has been uploaded, a wide variety of parameters and questions can be tested without additional effort.
The app currently offers the applications “Find signal dependencies” and “Find influencing factors”. The former determines how well a signal can be predicted live over other signals in time series data using a model. The Insights report also includes the signals that contribute to the prediction. “Find Influencing Factors” identifies how well a property can be predicted from other properties based on tabular data. The results report lists the properties relevant to the prediction.
These two applications already allow users to quickly and easily find deep relationships and dependencies in their data that help them better understand and improve processes.
The range of applications is being expanded continuously. In the near future, it will be possible to determine the causes of certain events as well as the properties relevant for the achievement of certain key performance indicators (KPIs).
Fabian Rüchardt, CEO of PerfectPattern, puts the aivis technology into context: “We have been hearing a lot about the democratization of artificial intelligence: AI should no longer remain ruling knowledge, but become a commodity. For our customers, however, the use of AI is not in the foreground – they have a clear, data-driven question arising from an industrial context and desire usable results. The fact that this requires machine learning and AI is interesting for our customers, but this is of secondary significance.”
According to Rüchardt, the PerfectPattern team has created a system that allows data owners in the industry to quickly gain insights from their data, even without data science expertise. Rüchardt continued: “Important aspects for us were also usability and user experience. It’s not enough to develop cutting edge technology, it also has to be convenient to use.”
“Practical relevance is also crucial”, says Robert Meissner, Product Manager for aivis: “There are already apps with similar basic features, but thanks to our technology we are much closer to the respective problem and its solution. We don’t bother our users with machine learning questions, but instead focus on an application-oriented approach and usable results.”
Everyone who registers with the app initially receives a free subscription. This already includes large parts of the functionality. If this is not enough, users can take out a PRO subscription for 299 euros a month to access the full range of features.