SYSTEM AND METHOD FOR GENERATING AND RETRIEVING DIFFERENT DOCUMENT LAYOUTS FROM A GIVEN CONTENT
A document search and retrieval system and program product therefor. Search requests are provided to the system through a user interface. A document decomposer decomposes documents into individual document components. Document components and corresponding searchable indices for each are stored in a Component Library. A search unit searches stored document components responsive to search queries. A results validator compares document hitlists with a document type identified in a search query to select valid hitlists entries for a final hitlist. A document view assembly module collects identified document components and assembles them into a document for view at the user interface.
Latest IBM Patents:
- Shareable transient IoT gateways
- Wide-base magnetic tunnel junction device with sidewall polymer spacer
- AR (augmented reality) based selective sound inclusion from the surrounding while executing any voice command
- Confined bridge cell phase change memory
- Control of access to computing resources implemented in isolated environments
The present application is a divisional application of allowed U.S. patent application Ser. No. 10/159,373, (Attorney docket No. YOR920020124US1) entitled “A SYSTEM AND METHOD FOR GENERATING AND RETRIEVING DIFFERENT DOCUMENT LAYOUTS FROM A GIVEN CONTENT” to Gregory T. Brown et al., filed Jun. 3, 2002, now issued as U.S. Pat. No. 7,254,571, which is assigned to the assignee of the present invention and incorporated herein by reference.
BACKGROUND OF THE INVENTION1. Field of the Invention
The present invention is related to searching and retrieving electronic documents over a computer network and more specifically, to dynamically assembling electronic documents at retrieval, based on the document type most suitable for the user context.
2. Background Description
Normally, someone wishing to find information over the Internet uses a search engine to identify and retrieve relevant documents. Documents available over the Internet normally have a fixed document type (e.g. Download, Hints & Tips, White Paper, etc.) with static content layout. Specialized search engines may filter search results based on document type, filtering out all documents not matching the specified document type or types.
Unfortunately, very often this typical system 100 may not provide an answer/solution to the query, especially, when the correct answer is embedded in a document that does not match the requested document type/layout. In another example, to find downloadable video driver for product A, a prior art system may limit the search scope to Download documents only. So, the search engine may overlook relevant information that appears in a Hints&Tips document instead for example. So, the search result is somewhat limited by a document layout or type that is normally once and forever determined by the document provider. Typically, unless the same document is stored in multiple formats, the searcher cannot choose content layout. So, typical state of the art search engines are restricted by the static nature of available documents. Thus, navigating through document storage to find relevant information often requires a level of familiarity with the document type schema. Document organization may hamper searching. Different content providers cannot choose suitable content and layout for particular local portals. So, users must live with whatever documents are stored and available.
These search constraints are especially troublesome in corporate technical support systems, typically a complex hierarchical schema of document types combined with a product taxonomy tree. Usually corporate-wide documents are standardized to provide a unified document view through the corporate technical support portal. These constraints make retrieving information from a corporate technical support system a challenging task especially if the document storage system contains heterogenous document collections.
Thus, there is a need for a way to select document presentation according to the needs of a particular user or presentation context.
SUMMARY OF THE INVENTIONIt is a purpose of the invention to facilitate finding relevant information regardless of the format of documents containing the information;
It is another purpose of the invention to present such information in a selectable document type and/or layout that may not match the format of the original document containing the information;
It is yet another purpose of the invention to choose a most suitable document content layout.
The present invention is a document search and retrieval system and program product therefor. Search requests are provided to the system through a user interface. A document decomposer decomposes documents into individual document components. Document components and corresponding searchable indices for each are stored in a Component Library. A search unit searches stored document components responsive to search queries. A results validator compares document hitlists with a document type identified in a search query to select valid hitlists entries for a final hitlist. A document view assembly module collects identified document components and assembles them into a document for view at the user interface.
BRIEF DESCRIPTION OF THE DRAWINGSThe foregoing and other objects, aspects, and advantages will be better understood from the following non limiting detailed description of preferred embodiments of the invention with reference to the drawings that include the following:
According to a preferred embodiment of the present invention, content components are extracted from original documents and stored in a component library. When a query calls for information found in the content components, a context is selected in which retrieved information is viewed. Documents of the chosen type/layout are constituted on the fly from stored document components. More relevant documents may be identified and rendered in a selected context/layout.
The Document Decomposer 122 distills individual components from each of the documents 128, 130, 130 and the components are stored in the Component Library 120. For example, a typical FAQ 128 might include a Title, a Problem Statement, a Solution and, optionally a reference link to additional information. Further, these document components can be collected and assembled to form an FAQ document. Likewise, a Download document may include a Title, a Solution, an Attachment and a reference. Table 1 is an example illustrating typical document components for several document types. Typically, each of these components is tagged by a section subtitle in the original document. Each document type has its own set of sections according to predefined corporate templates. The Document Decomposer 122 locates each tagged component, extracts each located component and stores extracted components in the Component Library 120. Then, the individual components are indexed in the Textual Index Unit 116, making each indexed component available for full text search.
A search is initiated with a query that specifies both search terms and preferred document type passing through the User Interface 112 to Search Unit 114. Search Unit 114 searches component indices in Textual Index 116 and retrieves a hitlist for specified search terms. Results Validator 126 checks the hitlist and identifies candidates that include all of the components needed to constitute a document in the selected format, e.g., FAQ format. The Results Validator 126 returns a list of remaining documents that can be constituted into the selected format. Each request also passes through User Interface 112 to Document View Assembly Module 124 which retrieves and assembles components into a document in the selected format. The assembled document is returned for viewing through User Interface 112.
In addition to document components, the Component Library 120 contains a table of document type masks for every supported document type. Table 2 shows an example of a document type mask table for the above example of four identified document types. Each document type mask defines a set of components constituting a particular document type.
In another preferred embodiment of the present invention, a document search is constrained such that the search result hitlist includes only documents that can be rendered in the requested viewing context. So, for example, while search results may identify numerous documents in each of the document types, the search results hitlist would list only those documents that can be constitute a FAQ type layout, i.e., FAQ and APAR type documents.
Thus, search result documents are provided in a user selected document type based upon the user request. Documents of a requested type are assembled dynamically from a given content. The document with an answer/solution for the user's question/problem can be found, even if its static document type does not match the document type requested by the user. Advantageously, the number of available document types for a given content is supplemented from previously unavailable documents.
While the invention has been described in terms of preferred embodiments, those skilled in the art will recognize that the invention can be practiced with modification within the spirit and scope of the appended claims.
Claims
1. A system for searching document components and retrieving search result documents, said system comprising:
- a document decomposer decomposing documents into individual document components;
- a component library storing said document components;
- a search unit searching stored said document components responsive to search queries; and
- a results validator analyzing document hitlists of documents identified by said search unit against a document type identified in a search query to select valid hitlists entries for a final hitlist.
2. A system as in claim 1 wherein search requests are provided as said search queries including search terms and a preferred document type to said system through a user interface and search results are provided from said system through said user interface.
3. A system as in claim 1 wherein the document decomposer locates document components within an identified document, extracts each located component and stores extracted components in said component library.
4. (canceled)
5. A system as in claim 3 wherein said results validator identifies documents in said hitlist containing components constituting a selected document type.
6. A document search and retrieval system as in claim 5 wherein said document decomposer decomposes documents of document types including frequently asked questions (FAQ), downloads and authorized program analysis reports (APAR).
7. A system as in claim 1 further comprising:
- a textual index unit containing a searchable index of said components stored in said component library.
8. A system as in claim 7 wherein said textual index unit includes an indexer indexing components in said component library to create a textual index for each component.
9. A system as in claim 1, further comprising:
- a document view assembly module collecting identified said document components and assembling collected said document components into a document for view at said user interface.
10. A document component search system operating on one or more computer systems, the computer systems having one or more central processing units (CPUs), one or more memories, and one or more network interfaces connecting the computer system to one or more computer networks, said document search system comprising:
- a document decomposer extracting text subportions from documents to distill document components, said documents being defined by one or more document type models, document components being indexed for full text searching;
- a document search unit identifying relevant documents of static document type and validating search results, valid search results being documents that include all components of a requested document type; and
- a document retrieval unit collecting components for a selected document type from a selected valid document and dynamically assembling said document components according to said selected document type for display.
11. A document component search system as in claim 10, said document decomposer comprising:
- identification means for identifying document components according to document type;
- storage means for storing documents for decomposition and storing document components in a component library; and
- text indexing means for indexing document component content for text searching.
12-13. (canceled)
14. A computer program product for searching and retrieving document components and presenting documents constituted from retrieved components for view, said computer program product comprising a computer usable medium having computer readable program code thereon, said computer readable program code comprising:
- computer readable program code means for entering search requests and presenting search results for view;
- computer readable program code means for decomposing documents into individual document components;
- computer readable program code means for storing said document components;
- computer readable program code means for searching stored document components responsive to a search query and listing documents identified as responding to said search query in a hitlist; and
- computer readable program code means for analyzing document hitlists against a document type identified in said search query to select valid hitlists entries for a final hitlist.
15. A computer program product as in claim 14, wherein decomposable document types include frequently asked questions (FAQ), downloads and authorized program analysis reports (APAR).
16. A computer program product as in claim 14, wherein the computer readable program code means for entering search requests accepts queries including search terms and a preferred said document type.
17. A computer program product as in claim 14, further comprising:
- computer readable program code means for storing a searchable index of said document components stored in said component library.
18. A computer program product as in claim 17, wherein said computer readable program code means for decomposing documents locates document components within an identified document, extracts each located component and stores extracted components in said component library.
19. A computer program product as in claim 17, wherein said computer readable program code means for storing a searchable index includes computer readable program code means for indexing components in said component library to create a textual index for each stored said document component, said textual index being stored as a searchable index.
20. (canceled)
21. A computer program product as in claim 17, wherein said computer readable program code means for analyzing document hitlists identifies documents in said hitlist containing components constituting a selected document type.
22. A computer program product as in claim 14, further comprising: computer readable program code means for collecting identified said document components and assembling collected said document components into a document for view.
Type: Application
Filed: Jul 10, 2007
Publication Date: Jan 17, 2008
Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION (Armonk, NY)
Inventors: GREGORY BROWN (ROCKMART, GA), THOMAS COFINO (RYE, NY), YURDAER DOGANATA (CHESTNUT RIDGE, NY), YOUSSEF DRISSI (OSSINING, MA), TONG-HAING FIN (HARRISON, NY), MOON KIM (WAPPINGERS FALLS, NY), LEV KOZAKOV (STAMFORD, CT), JOHN MILLER (DULUTH, GA)
Application Number: 11/775,514
International Classification: G06F 17/30 (20060101);