System and method and computer program product for ranking logical directories

• US 7,464,076 B2
• Filed: 05/15/2004
• Issued: 12/09/2008
• Est. Priority Date: 05/15/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Tomlin, John Anthony, McCurley, Kevin Snow, Eiron, Nadav
• Primary Examiner(s): ABEL JALIL, NEVEEN
• Application Number:
  US10/847,143
• Publication Number:
  US 20050256887A1
• Time in Patent Office: 1,669 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation

Abstract

A logical directory ranking system ranks documents or web pages utilizing logical directories. The present system groups together compound documents as a single information node with one or more leaves, constructing a logical directory graph. URLs can be grouped at a level of granularity below an individual directory. For example, the URLs may be grouped together on the basis of hostname, domain, or any level of the hierarchy of the URLs. Edges in the logical directory graph are formed by links between the logical directories. Edges have weights corresponding to the number of links between logical directories. Nodes have weights corresponding to the number of web pages or leaves represented by a node. A ranking level is determined for each node as a function of the node weight and the edge weight. The ranking level is then applied to each URL that the node represents.

Claims

1. A processor-implemented method for ranking a plurality of nodes in a ranked logical directory graph with a superimposed hierarchy of logical directories, the method comprising:

accessing a plurality of uniform resource locators;

forming a logical directory graph from said plurality of uniform resource locators which have been accessed, wherein each of said uniform resource locators comprises a plurality of elements with at least one argument specification to a dynamic uniform resource locator separated by a delimiter, and represents one of;

i) a link to a logical directory in said hierarchy of logical directories, or ii) a link to a web page, file, or iii) document contained in said logical directory comprising;

identifying each node in each of said uniform resource locators corresponding to said logical directory at each logical directory level in said logical directory graph, wherein each said logical directory is represented by information found in said logical directory'"'"'s uniform resource locator up to a last standard delimiter in said uniform resource locator, forming said logical directory graph comprising at least two nodes thereby providing a greater level of granularity than just a root node;

identifying all leaves by uniform resource locators which match said node'"'"'s uniform resource locator up to said last standard delimiter, wherein each leaf comprises information found after said last standard delimiter of said leaf'"'"'s matching uniform resource locator;

forming a single group node in said logical directory graph by;

i) grouping together said uniform resource locators for all nodes that match up to said last standard delimiter ii) removing at least one of the plurality of elements from the dynamic uniform resource locator, and iii) grouping a plurality of uniform resource locators with respect to at least one of a plurality of variable names;

forming edges between nodes in said logical directory graph having a link between said each other; and

aggregating together into a single edge, all multiple edges found to exist between any two nodes in said logical directory graph;

assigning an edge weight to each of said single edges in said logical directory graph, said edge weight corresponding to a number of links found to exist between said any two nodes;

assigning a node weight to each of said nodes in said logical directory graph, said node weight corresponding to a number of leaves associated with that node;

assigning a ranking level to at least some nodes in said logical directory graph, said ranking level based on said node weight and said edge weight;

applying said ranking level to at least some leaves associated with at least some of said nodes which have been assigned a ranking level; and

generating a ranked logical directory graph based upon said ranking level which have been applied to said leaves.

2. The method of claim 1, wherein the logical directory is identified using a structure of a uniform resource locator for a page on the world wide web.

3. The method of claim 1, wherein the plurality of elements comprise a host name and a file path specification.

4. The method of claim 1, wherein the plurality of elements comprise at least one directory.

5. The method of claim 1, wherein the node weight is assigned a predetermined value of one, reducing a processing time of the logical directories.

6. The method of claim 1, wherein the edge weight is assigned a predetermined value of one, reducing processing time of the logical directories.

7. The method of claim 1, wherein the ranked logical directory graph is sent to PageRank processor.