Mt search grammar is the trial and error outcome of reverse engineering the engine database structure.
The assumption in my case is that it is all one semantic net until the detail is minimal enough to support a large dictionary. Thus is determined my search grammar.
My search grammar is the same, from general to specific I traverse a known graph until I iot a known, specific and large dictionary, then end my search phrase with specific terms within the known dictionary.
My search is right to left as I descend a semantic tree. Then list specific keywords. It is like having prepends to get the proper framework. And postpends to find sets within the framework. A very simplified version of the written grammar we are taught. It is almost an exact copy of 1) Find the book 2) Go to the index.
The latest Moore's Law shortened step 1 and added look up power to step 2. The search engines have index tables for many sorts of specialized activities.
No comments:
Post a Comment