Synthetic test data
For many purposes nowadays you need data that is not available. We offer to generate data according to your specifications for the evaluation of a new CRM, your database or your product/application. These can include names, valid address, numeric attributes like age, binary variables and much more according to your preferences. In addition, it is often necessary to place outliers in such data in order to test systems correctly. During testing, it should always be analyzed how the system handles correct and incorrect inputs. This can be integrated into a generated data set after consultation.
Imputation
Imputation includes all statistical methods to artificially fill missing values in a data set. These methods cannot be used for all types of data, but imputation should be considered and applied prior to any analysis or modeling.
With our imputation methods, we place a lot of emphasis on working closely with you, as only a solid understanding of the data will result in flawlessly populated data.
Simple example of an imputation
Input Data
Name | Age | ( Other attributes ) | Is interested |
Wenger | 55 | … | Yes |
Müller | 34 | … | |
Meier | 18 | … | No |
Bauer | 27 | … |
Imputated data
Name | Age | ( Other attributes ) | Is interested |
Wenger | 55 | … | Yes |
Müller | 34 | … | Yes |
Meier | 18 | … | No |
Bauer | 27 | … | No |