Reporting The Results
todo: full update …
In this third group of blogs, we finally turn our attention back to the concrete tasks we need to perform to accomplish our objectives. Reusing the generalizable mechanisms and results from the previous section, we outline our now English textual input preparation steps. We also specify the format of our outputs, optimized on purpose for flexibility and versatility during future processing.
The effective product of all the activities presented in this section is the “conflict graph”, a condensed summary of the now “understood” meaning of our input texts. This graph is then interpreted and used to synthesize various metrics and statistics about the semantic context of our data. We describe the graph in more detail and supply a set of custom operators on it.
We also outline an introductory sampling of analytics. As the end-results are all reproducible, traceable and verifiable numeric quantities, standard data-mining and “big-data” practices can be applied. We end the blogs by presenting just such a “data-mined” report, based on actual filed documents and Court proceedings.