After the general business value of the unstructured data warehouse is established, it is a good idea to look carefully at the source data. Chapter 1 has shown that there are many different forms of unstructured text.
The first step in the integration of unstructured data is that of physically accessing (that is, reading) the unstructured data. Stated differently, if you cannot physically access unstructured data, then you will never be able to integrate the data. Figure 4.2 contains a more detailed version of the textual integration process from Figure 3.3.
Figure 4.2 The ‘E’ of ETL (Extract)