Discovering the value and use of data
“How valuable is my data? Which benefits can be obtained?” Such questions can often only be answered after initial exploration and analyses in the course of the project. We use tools like R and Python on test data (e.g. samples) to investigate which performance figures can be derived from your data pool and which predictions can be made with them. In this way a prototype for data analysis is developed, the first inquiries are sounded out and the results prepared for clear presentation.
Combining your own data with external data
The mixture is decisive – data from your own company are not the only source for big data solutions. In many instances, benefits and expediency only arise from the combination with data of open data pools – for example data from public authorities and open sourcing initiatives – or data of third parties. Data exploration is therefore a matter of identifying the right data sources on the basis of an inquiry and combining them correctly.
Gaining insights and findings – small to large data volumes
Data mining (DM) and procedures revolving around Knowledge discovery in data bases (KDD) like statistical methods and Machine learning algorithms are not only useful for data exploration. They assist in the recognition of patterns in data sets, in generating forecasts and thereby support decision making. In connection with scaling considerations, the insights from the exploration phase can also be transferred to larger data volumes.