The following tables contain all features of ClearVu Analytics, sorted by groups.

+49 231 97 00 340 contact@divis-gmbh.com

(CVA)

The following tables contain all features of ClearVu Analytics, sorted by groups.

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Graphical user interface | Visualization, simple workflow, integration of all components, flexible configuration | ✓ | ✓ | ✓ |

Parallelism | Parallel calculation of models | ✓ | ||

Easy installation | Minimal requirements to the software environment | ✓ | ✓ | |

Intuitive project management | – Persistence of the projects – Project representation by one file | ✓ | ✓ | ✓ |

Clipboard selection | Easy clipboard selection (copy/paste) of graphics and tables from CVA to other applications | ✓ | ✓ | ✓ |

Customized configuration | Customization options for the corporate design for the generation of graphics and reports | ✓ | ✓ | ✓ |

Documentation | – User manual – Online help – Tutorial | ✓ | ✓ | ✓ |

Data import | Supported formats: Excel, csv, ods | ✓ | ✓ | ✓ |

Graphic export | Raster image, svg, emf-format | ✓ | ✓ | ✓ |

Modeling interface | Interface to “R” for the usage of external modeling algorithms | ✓ | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Quality measure for variables | ✓ | ✓ | ✓ | |

Various graphical visualizations | Histograms, correlation plots, scatter plots | ✓ | ✓ | ✓ |

Filters for variables | ✓ | ✓ | ✓ | |

Variable transformation | Automatic suggestions for variable transformations to enhance the model quality | ✓ | ✓ | ✓ |

Variable groups | Declaration of restrictions in groups (e.g., selection of two out of five variables to form a group) | ✓ | ✓ | ✓ |

Outlier detection | Automatic outlier detection based on a standard deviation measure | ✓ | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Automatic variable selection | Hierarchic clustering for complexity reduction | ✓ | ✓ | ✓ |

Linear models | ✓ | ✓ | ✓ | |

Variable selection for linear models | Forward and backward selection | ✓ | ✓ | ✓ |

Rule based fuzzy models | Model representation as fuzzy rules | ✓ | ✓ | ✓ |

Support vector machine | ✓ | ✓ | ✓ | |

Decision trees | ✓ | ✓ | ✓ | |

Kriging | ✓ | ✓ | ✓ | |

Random forests | ✓ | ✓ | ✓ | |

Neural networks | ✓ | ✓ | ✓ | |

Automatic meta modeling | Automatic selection of the best model, given the data set at hand | ✓ | ✓ | ✓ |

Automatic adjustment of the models to the data | ✓ | ✓ | ✓ | |

Calculation of different quality criteria for model quality | ✓ | ✓ | ✓ | |

Integrated hyperparameter optimization | ✓ | ✓ | ✓ | |

High dimensional model outputs | Represented as a set of meta models | ✓ | ✓ | ✓ |

Measures for the importance of influencial variables | ✓ | ✓ | ✓ | |

Sensitivity analysis | Interactive 2D and 3D surface plots and ternary diagrams | ✓ | ✓ | ✓ |

Contour plots | ✓ | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Standard designs | Factorial design | ✓ | ✓ | |

Plackett-Burman | ✓ | ✓ | ||

D-Optimal design | ✓ | ✓ | ||

Latin hypersquares | ✓ | ✓ | ||

Space filling designs | Considers already existing data points | ✓ | ✓ | |

Enables linear restrictions | ✓ | ✓ | ||

Considers restrictions of variable groups | ✓ | ✓ | ||

Formulation design | Considers filling substances | ✓ | ✓ | |

Miscellaneous | Editor for new variables | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Integration with the modeling | Objective functions of the optimization are based on models | ✓ | ✓ | |

Editor for mathematical objective functions | Objective functions can be formulated mathematically by using model outputs | ✓ | ✓ | |

Utilization of multiple target functions | Full integration of multi criteria optimization including visualization of the Pareto front | ✓ | ✓ | |

Constraints editor | Constraints on the target functions can be formulated in a general way, also under consideration of algebraic constraints | ✓ | ✓ | |

Optimization visualization | Display of optimization progress and Pareto front | ✓ | ✓ | |

State-of-the-art optimization methods | Advanced evolution strategies for identifying the best solutions in high dimensional multimodal search spaces | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Slider bar analysis | Exploration of models by interactive parameter variations | ✓ | ✓ | |

Easy, visual and fast evaluation of planned experiments | ✓ | ✓ | ||

Visualization of the sensitivity of specific parameters | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Modeling add-in | Application of the models as Excel add-ins (i.e. cells in Excel can calculate model based predictions based on CVA models) | ✓ | ✓ | |

Calculation of the models in Excel | ✓ | ✓ | ||

Model exchange between CVA and Excel | ✓ | ✓ |

Feature | Description | Basic | Standard | Professional |
---|---|---|---|---|

Command-line features | Usage of all components (modeling, optimization) by means of command-line calls (without GUI) | ✓ | ||

Design of experiments in batch mode | ✓ | |||

Modeling in batch mode | ✓ | |||

Optimization in batch mode | ✓ |

Visualize data and gain insights

This module of CVA supports the exploratory data analysis, i.e., the visualization and the first analysis of the data. For this process the following methods are available:

- Easy data import from common data formats (csv, Excel)
- Definition of value ranges and constraints
- Scatterplots
- Boxplots for analyzing the data distribution
- Outlier detection
- Calculation and visualization of correlations
- Suggestions for variable transformation to support the modeling
- Extensive reporting functions (export of charts)

For many applications, for example formulation optimization, additional functions are implemented. Furthermore, CVA is able to handle mixed-integer data as well as categorical data (class variables) and partially missing values without problems.

Generate the best model automatically

The modeling module in CVA provides the possibility to automatically generate models from data. To find the best model CVA applies multiple (mostly nonlinear) modeling algorithms on the data because it is impossible to determine the best one beforehand.

CVA optimizes the learning parameter for all methods through so-called hyperparameter optimization. This approach guarantees the configuration and selection of the best modeling algorithms and the generation of the best model. The target criterion for the models is their ability to generalize, i.e. the prediction quality of the models is evaluated. Using a so-called cross-validation approach avoids overfitting, a common phenomenon in modeling processes.

A traffic light system tells you how successful the modeling is and which model performs best. So you can directly proceed and apply the models for prediction, sensitivity analyses or optimization.

Experts have the possibility to influence the learning parameters of the methods. However, experience shows that this is rather unnecessary because of the excellent quality of the automatic modeling.

Although CVA provides all state-of-the-art methods of nonlinear data analysis, the generation of the optimal model is very easy for the user—only computation time is needed. The result is presented in an intelligible way. Among the available modeling algorithms are:

- Linear models
- Support vector machine
- Fuzzy models
- Decision trees
- Random forests
- Neural networks (MLP, feed-forward)
- Gaussian process
- Partial least squares regression
- Principal component regression

Understanding and using models

The best model which was generated during the optimization can be applied and analyzed in CVA in multiple ways:

- Model-based prediction
- One- and two-dimensional sensitivity analysis
- Ternary diagrams for three-dimensional sensitivity analysis
- Model inspection for certain modeling algorithms (non-black-box methods)
- Determination of parameter importance via Sobol-indices
- Reduction of the parameter dimensionality with dendograms
- Various visualizations

Take prediction as an example: for new parameter settings (process setting, product configuration, etc.) the model output (quality, stability, costs, etc.) can be predicted. Moreover, you can conduct sensitivity analysis to find out what effect the variation of one or two parameters has on the model output. With the help of ternary models you can also examine the influence of three variables which sum up to a constant value (often used in formulation optimization).

In model analysis, it is possible for some model types to inspect the model on your own. For example, for linear models as well as fuzzy models it is possible to inspect the rules for obtaining a deeper understanding of the interrelations within the system.

Further analysis methods include the determination of the importance of particular parameters (e.g., by means of so-called Sobol indices) and the usage of parameter dendrograms to reduce the dimensionality of the parameter space. Both methods can be used intuitively, since intuitive visualizations are provided.

Contact details

divis intelligent solutions GmbH

Joseph-von-Fraunhofer-Str. 20

44227 Dortmund, Germany

+49 231 97 00 340

contact(at)divis-gmbh.de

Newsletter

Our free newsletter informs you about news and dates.

Here you can subscribe to our newsletter: