This text will comprehensively describe how to use the TR-WWW search features.
This TR-WWW setup is configured to combine multiple terms with a Boolean "and." This means the search engine will make multiple passes through the document set and find the logical intersection of all those documents containing the search term(s).
TR-WWW searches to not support nested queries nor truncation (stemming), but searches are case-insensitive.
Search results will be presented you in "ranked relevance" order. This means the documents appearing towards the beginning of the list are considered more relevant than the items appearing later in the list. This relevance ranking is computed considering the length of the searched documents and the number of times the search terms are located within the documents.
- Context versus relevance searches
- This radio button input option allows you to choose between context searches and relevance searches. When context searches are selected, then the search results will list lines of text containing your search term(s), search terms in context. You then have the option of retrieving the located document in its entirety by clicking on the highlighted term. Relevance searches are the default option of simple searches. These searches return lists of documents in "relevance ranked" order.
- Maximum hits to return
- If your search type is "context", then you can specify the maximum number of hits to return. Enter an integer between 10 and 200 in the text input field.
- Term relationship operator
- If you enter multiple terms in the text input field, then you can specify how you would like to have your terms logically combined. The choice are:
- Or
- This option creates the logical union of all the located documents containing your search terms.
- And
- This creates the logical intersection of all the documents containing your search terms. This is the setting for simple searches.
- Phrase
- You can use this option to locate search terms next to each other. For example, if the search terms are "Apple Corporation" and the Boolean relationship is Phrase, then you should find documents containing the phrase "Apple Corporation."
- Near
- This option allows you to specify that the search terms should be close together within the located document(s).
- Notnear
- This is the opposite of the Near option and specifies that the search terms should be distant from each other within the document.
- Nearness factor
- If you selected "Near" or "Notnear" as you term relationship operator, then use this option to specify how near or not near the search terms should be located. Enter an integer representing 10 to 200 words.
Eric last edited this page on September 26, 1995. Please feel free to send comments.