Supervised data refining engine. Really experimental.
By a given subject(enumeration of tokens) specified in link_dump.py should traverse various resources on the web and accumulate a list of usable links to be processed.
gather_info.py will gather all text considered useful from the files.
At this point all raw data has been accumulated.
The gathered data is passed to structure.py which would try to formulate meaningful sentences based on the raw data, from which the user refines the results interactively.