Data analysis and data mining
Statistical analyses
Data contains information - we are happy to help you make this information visible so that you can make the right decisions for your product or process. Whether your data comes from planned tests, quality assurance or the production process, we will take care of selecting the right statistical analysis method, performing the statistical tests and preparing the results for you. Our employees are experts in statistical data analysis and have years of experience in handling data material. In addition, we can flexibly use just about any common software package.
Planning and evaluation of surveys
After carefully designing the questionnaire and conducting the survey, the next important task is to evaluate the questionnaire. This evaluation is done with professional statistical software. We start with simple graphical and evaluations and tables, which can be sufficient in many cases. For more complex questions, more complex statistical models are often required, which we master and are also happy to interpret together with you. The results of the survey are documented and finally presented and discussed with you.
Data Mining
Data mining is a synthesis of statistics and computer science. Explicitly, statistical methods are used here to generate knowledge from large amounts of data - Big Data. A distinction is made between two main areas of application:
Unsupervised Learning" tries to find patterns/structures in large amounts of data. This is helpful in many areas. For example, business decisions can be made based on the insights gained - as is done in the field of business intelligence (BI) - or research can be carried out into anomalies in medical data in order to discover risk factors.
Supervised learning" is also used in the BI environment. Here, the attempt is made to generate forecasts that are as precise as possible. These can then influence decisions in the context of business intelligence or directly solve practical problems, such as text recognition (OCR). Typical methods used here include decision trees (CART, MARS), neural networks and ensemble models.
Data Mining Tools
Even though the science of data mining evolved in part from statistics, this is where many statistical programs reach their limits. Most data mining/business intelligence tools are stand-alone programs that usually master the two applications that are important for business intelligence - pattern recognition and forecasting. Especially for the visualization of processes and large amounts of data, programs, quite excellent.
Of course, there are also exceptions to the rule. In the course of the increasing relevance of BI topics, the most important statistical tools (e.g. SPSS, SAS, R) now also offer offshoots or additional modules to be able to answer the questions of data mining.
Business Intelligence
Business Intelligence (BI) is a generic term for the systematic collection, processing and analysis of data. The goal of Business Intelligence is to optimize business processes with the information gained in order to improve quality, minimize costs or reduce risks. The greatest support is provided by a variety of tools that support the BI process and integrate it into the existing software landscape.