Advanced Level

GeoWorlds Project, Distributed Scalable Systems Division
University of Southern California Information Sciences Institute

Back Home Up

 

Advanced Level
  1. How can I perform a geo-located query?
  2. How can I exchange documents and/or categories among users?
  3. What is the advanced functionality available for manipulating category folders in the Category Editor?
    Compare
    Cross
    Intersect
    Subtract
    Add
    Duplicate
    Flatten paths
    Domain names
    Leaf nodes
    Categorize by property value
    Sort
    Sort recursively
    Prune out empty categories
    Count documents
    Clear property
    Set property
  4. How can I customize the software to my own needs?
    Set the maximum number of web sites visited
    Set the wrapper time out
    Ignore same domain URLs
    Set the web browser command
    View original source
    Automatically translate on download
    Set up the translation server
    Set up the summarization server
    Display document properties
    Automatically take snapshot on download

How can I perform a geo-located query?

The query tool can automatically fill out geo-locational fields in search engines, such as the GTE SuperPages. Also, it can add the geo-locational names as additional Boolean constraints to search engines. Also, see How can I get the latest place names exported from the Map manager.

Show me!

Back to Top

How can I exchange documents and/or categories among users?

Clipbook servers provide a way to exchange information among remote clients. The information exchanged can be a partial category, an entire category, or a window panel.

Also, see What is the difference between the Clipboard and the Clipbook.

Show me!

Back to Top

What is the advanced functionality available for manipulating documents in the Category Editor?

See the functionality listed below.

Back to Top

Compare

Use a 3D bar chart to compare the content of two category structures (A and B). The structures can be select from either the Category Editors window or the Document Analysis window. Three comparison metrics are provided:

The Similarity metric measures the closeness of two categories. In the 3D bar chart the height of the bar equals to the size of intersection of two categories divided by the size of the union of the two categories times 100. A value of 100 indicates the two categories contain exactly the same set of documents. A value of 0 indicates that they have no documents in common. This tabbed pane is useful in exploring ways to merge documents sets that are similar.
The Subdivide by A metric measures how categories in A partition categories in B. The height of the bar equals to the size of intersection of two categories divided by the size of the category in B times 100. A value of 100 indicates the B category is completely contained in category A. A value of 50 indicates that the B category is partitioned in half by category A. The tabbed pane is useful in exploring ways to subdividing categories in B by categories in A.
The Subdivide by B metric, which is the opposite of Subdivide by A, measures how categories in B partition categories in A. The height of the bar equals to the size of intersection of two categories divided by the size of the category in A times 100. A value of 100 indicates the A category is completely contained in category B. A value of 50 indicates that the A category is partitioned in half by category B. The tabbed pane is useful in exploring ways to subdividing categories in A by categories in B.

Show me!

Back to Top

Cross

Compute the cross product of two categories. This operation is conceptually depicted by the following picture:

cross.jpg (8245 bytes)

Show me!

Back to Top

Intersect

Compute the intersection of two categories (A intersect B) at the document level, i.e., the position of the document in the category structure is ignored for the purpose of determining document equality. This operation is equivalent to removing all the documents in A that do not occur in B.

Show me!

Back to Top

Subtract

Compute the subtraction of two categories (A - B) at the document level, i.e., the position of the document in the category structure is ignored for the purpose of determining document equality. This operation is equivalent to removing all the documents in A that  occurs in B.

Show me!

Back to Top

Add

Compute the addition of two categories (A + B) at the document level, i.e., the position of the document in the category structure is ignored for the purpose of determining document equality. This operation is equivalent to adding from B to A documents that do not occur in A.

Show me!

Back to Top

Duplicate

Make a clone of a category.

Show me! (first portion of subtract Show me!)

Back to Top

Flatten paths

The two flatten path operations are available. Flatten Paths collapses a multi-level category into an one level category, i.e., a root node with children as leaves. The names of the collapsed node are pre-appended to the name of the leaves.    Flatten Paths 1 adds a intermediate level of nodes for a two-level category.

Show me! (Flatten Paths)       Show me! (Flatten Paths 1)

Back to Top

Leaf nodes

Similar to Flatten Paths except that names of the collapsed nodes are not prepended to the name of the leaves.

Show me!

Back to Top

Domain names

Reclassify documents of a category based on the URL domain names of the documents.

Show me!    

Back to Top

Categorize by property value

Each category is associated with a list of property name/value pairs. Reclassify   categories based on their property values.

Show me!

Back to Top

Sort

Alphabetically sort a category's children based on their name.

Show me!

Back to Top

Sort recursively

Alphabetically sort a category's descendents of  based on their name.

Show me!

Back to Top

Prune out empty categories

Delete categories that do not contain any children with documents.

Show me!

Back to Top

Count documents

For each category, count the number of descendents it has.

Show me!

Back to Top

Clear property

Each category is associated with a list of property name/value pairs. Delete one or more property name/value pairs from a category.

Show me!

Back to Top

Set property

Set a  property name/value pairs of a category.

Show me!

Back to Top

How can I customize the software to my own needs?

See below for the available customization.

Show me!

Back to Top

Set maximum number of web sites visited

Set the maximum number of web sites to visit. Currently, this option is only used by the Yahoo directory search.

Show me!

Back to Top

Set the wrapper time out

Set the number of second the system is willing to wait before deciding that a Web server is down. Currently, this option is only used by the Yahoo directory search.

Show me!

Back to Top

Ignore same domain URLs

For the URL List command,  ignore URL links with the same domain name as the parent. This is sometimes useful in weeding out links back to top of page, home pages, and commercial pages.

Show me!

Back to Top

Set the web browser command

Set the command used by the OS to execute the Web browser.

Show me!

Back to Top

View original source

View the original document, instead of the cache version.

Show me!

Back to Top

Automatically translate on download

Automatically translates the document on download.

Show me!

Back to Top

Set up translation server

Select which translation server to use. The rows represent the language to translate from. The columns represent the server to use. Each row can have at most one server selected.

Show me!

Back to Top

Set up summarization server

Select which summarization server to use. The rows represent the input document language and the output summarization language. The columns represent the server to use. Each row can have at most one server selected.

Show me!

Back to Top

Display document properties

Select which properties to display and in what style in the category editor. The radio buttons on the left provide a list to available display styles. The check buttons on the right provide a list of available properties.

Show me!

Back to Top

Automatically take snapshot on download

Automatically take a snapshot of the content of the document  imported through the browser. Normally, only the URL of document is imported.

Show me!

Back to Top
 

(C) Copyright 1998-2003 USC Information Sciences Institute. All Rights Reserved.
For problems or questions regarding this web contact [[email protected]].
Last updated: July 02, 2001 .