26. June 2024

Standalone Automated Machine-learning software platform

divis intelligent solutions releases ClearVu Analytics, a standalone Automated Machine-learning software platform



The ClearVu Analytics (CVA) software platform enables no-code automated machine learning for structured datasets. Using CVA, products can be developed and optimized faster, process data can be analysed and predictive models can be created – offering advantage to application areas such as quality and maintenance optimization, manufacturing, engineering and more.

Our bespoke platform is approach-agnostic. This is to say that when undergoing the modelling process, CVA automatically compares various Automated Machine Learning (AutoML) approaches, and selects the best one given the dataset the user is modelling. It does this through a series of cross-validation, hyperparameter optimization and statistical tests. The user is given the choice of what pool of methods they want to compare or can select a specific one to use if they want to leverage a certain approach (e.g. a neural network). Each modelling approach is given a quality evaluation, allowing a user to easily assess the effectiveness of different approaches.

Once a model is created, it can be used within CVA for optimisation, i.e., for searching for input parameters that maximise or minimize the model output.

Visualizations and graphs can also be created within CVA, meaning that visualisations of the solution space, dataset, model, and even comparisons of different modelling approaches can be easily created.

And finally, our design of experiments functionality helps the user understand how best to enrich the data, either through physical experiment or direct simulation. This can be extremely useful particularly for datasets with a wide variety of input and output parameters.

Leveraging best-in-class AutoML, CVA can be used to improve a variety of business, manufacturing processes within an organization, making AI accessible and useful in data-driven business.

For more information on CVA – and to arrange a demo – please see our website and to discover how the divis CVA platform can improve the way you interact with, model, visualize and leverage your datasets.

For any further questions, please contact us at divis. https://divis-gmbh.de/contact/