Welcome! Log In Create A New Profile

Advanced

another way of indexing and searching

Posted by Jan 
Jan
another way of indexing and searching
December 17, 2010 02:31PM
A major difference between a book index and an index generated by a search machine is that the book index just contains entries that are relevant to the subject and acomputer generated index contains all words from document except some words that are listed to this purpose. Looking in the keyword table it is clear that the majority of the keywords listed are irrelevant and yet use space.

I think there is a way of making a table with relevant keywords.

The first time a searching term is typed in the program searches for that word in the pages (following links as is done when indexing) After going through the site places were the term has been found are given. At the same time this word is placed as a keyword in the keyword table. All in the same way is done when indexing plus mentioning the result. If the same term is asked again the program first looks in the keyword table, gives the results, looks for pages that have come after the earlier searching procedure and, if present, adds the data to the database. In this way a relevant index is built up gradually.

Of course it is wise if the owner of the site introduces himself the first few of most relevant terms when starting. Later, when a new term is asked a message can come like 'This is the first time this term is searched for, please be patient.' As usually sites grow. If a term already present is asked, the first result can come quickly, as the database is filled already. So the a message can come like 'Here are some results, more may come.'

If this is a bad idea, please make clear to me where I make a mistake.

Best wishes,
Jan
Sorry, only registered users may post in this forum.

Click here to login