Retrieving, organizing, and utilizing networked data using databases

Abstract

A networked catalog search, retrieval, and information correlation and matching system is disclosed. The system allows suppliers to publish information in electronic catalogs, structure the information in an object oriented representation distributed across a network of computers, for example, the Internet. The system also enables customers to search and retrieve information on products and suppliers that match dynamically specified customer requirements. Through retrieving protocol compliant pages or other information, a search engine forwards retrieved pages to a database which sorts received information by the information's internal organization structure. By searching the information as stored in the knowledge base, a user may quickly retrieve the stored information as highly tailored to the user's search strategy. Also, additional tools operate on the stored information to, inter alia, organize design trades between selected elements where additional design and selection options are provided to each user. Further, the system supports product creation as formed of constituent parts. These products may then be used as parts to be formed into higher-level products.

Claims

We claim:

1. A portal for retrieving information from a supplier, said supplier providing information arranged in a predefined protocol to said portal, said portal comprising:

a storage for storing said information from said supplier in catalogs, said storage comprising a database, said information being stored in said database in accordance with organizational terms defined by the protocol, said protocol including an ontology for organizing said information into an ordered form; and

a processor for querying said storage in response to searches from users.

2. The portal according to claim 1, wherein said database is a relational database.

3. The portal according to claim 1, wherein said database is an object oriented database.

4. A portal for retrieving information from a supplier, said supplier providing information arranged in a predefined protocol to said portal, said portal comprising:

a storage for storing said information from said supplier in catalogs, said storage comprising a database, said information being stored in said database in accordance with organizational terms defined by the protocol; and

a processor for querying said storage in response to searches from users,

wherein the protocol defines a domain and a class, and said database is a relational database that includes a domain table and a class table.

5. A system for accessing catalogs as stored by a portal in a database, said system comprising:

a terminal generating a search request and transmitting said search request to said portal,

said terminal receiving the results of a search of said database,

wherein the portal performs said search of information stored in said catalogs as stored in said portal and wherein said catalogs are created using organizational terms from suppliers, said organizational terms being part of an onotolgy.

6. The system according to claim 5, wherein said database is an object oriented database.

7. The system according to claim 5, wherein said database is a relational database.

8. A system for accessing catalogs as stored by a portal in a database, said system comprising:

a terminal generating a search request and transmitting said search request to said portal; and,

said terminal receiving the results of a search of said database,

wherein the portal performs said search of information stored in said catalogs as stored in said portal and wherein said catalogs are created using organizational terms from suppliers,

wherein said database is a relational database and wherein said relational database includes a hierarchical structure.

9. A method for retrieving information from a portal, said method comprising the steps of:

transmitting a search request to said portal, said portal having a database having a hierarchical structure of classes and entries in which information from suppliers is stored based on organizational terms present in the information from said suppliers; and

receiving a search result from said portal, said search result referencing the hierarchical structure of said database.

10. The method according to claim 9, wherein said database is a relational database.

11. The method according to claim 9, wherein said database is an object oriented database.

12. A system comprising:

a supplier computer having data organized in accordance with an ontology;

a portal that receives said data, said portal having a storage that stores said data based on said ontology, said ontology having a hierarchical structure of classes and entries in which information from at least one supplier is stored based on organizational terms present in the information from said at least one supplier; and

a user computer that transmits queries to said portal and receives results from said queries, said queries relating to at least information stored in said storage.

13. The system according to claim 12, said storage further comprising a relational database.

14. The system according to claim 12, said storage further comprising an object oriented database.

Description

The following description includes some copyrighted material. While Applicants do not object to the copying of this specification for patent related purposes, Applicants reserve all copyrights to themselves and/or the assignee of the present invention.

BACKGROUND OF THE INVENTION

1. Technical Overview

The present invention relates to a networked, information gathering and delivery system. More specifically, the present invention relates to a network based (intranet-based, Internet-based, extranet-based) cataloging system that provides a user with a plurality of methods of gathering and analyzing information relating to a subject of interest.

2. Related Art

Manufacturing and service industries exist to provide products and services to end users while making a profit at the same time. In this day and age, many businesses do not control the entire manufacturing process from collecting raw materials to outputting a high level end product. Rather, most businesses take existing components and incorporate the components into larger products. An example includes engineering design systems in which engineers take existing components and design new products using these components. To reduce costs, these businesses generally attempt to maximize profits through finding the most affordable materials and processes that satisfy their needs. Simultaneously, to meet the performance or quality objectives for the end product, business must find materials and processes that satisfy a variety of other criteria. Unfortunately, depending on the industry, finding the lowest cost materials or services which meet specified criteria can become very costly and time consuming.

Currently, the designers selecting the various components for a final design can quickly become overwhelmed by the available data as many lower end suppliers may produce similar products. Sorting the available products and attempting to match potential products into an overall design scheme quickly becomes a formidable task when a single end product may contain tens to hundreds to thousands of individual components with each component having its own set of associated data (for example, operating tolerances, power consumption, size constraints, etc.). Further, the manufacturing process, which produces the final design, may affect the end product in ways unconsidered by the product development team. Because of the significant burden placed on designers to handle design while another set of engineers generally handle the realization of the design through a manufacturing process, the end product may suffer from poor integration of the design and manufacturing processes. To this end, product support also suffers as the team handling product support generally has minimal information regarding the actual design and manufacturing processes and how these processes affect the end product. At least one cause of the failures of the current design, manufacturing, and product support model is the requirement that all steps are handled by an engineer with minimal to no automated support for accommodating the demands of the current model. In fact, most information has to be manually integrated and considered in the formation of a final product.

Another failure of the current model is the lack of integrated tools available to a designer. Currently, a designer must wade through technical information and sequentially use disparate, unintegrated electronic tools to perform a variety of tasks such as requirements definitions, computer aided design (CAD), structural analysis, costing, manufacturing planning, etc. Accordingly, a need exists to automate and integrate the design function as much as possible.

The problems associated with the current design and manufacturing model exist in other industry sectors as well. For example, the real estate industry uses as its primary search and indexing tool the multiple listing service (MLS). This service allows real estate brokers to list properties as well as search for properties using a simple Boolean search function. However, the real estate agent is then required to peruse through the list of retrieved entries until a match is found to many diverse buyer requirements. Further, there is no ability to link to other sources of other information (for example, financial institutions providing loan information, etc.) that may be relevant to the information contained within the MLS. Moreover, MLS is limited to real estate agents only. Independent sellers and buyers cannot list and retrieve the listings available in MLS.

Unlike the restricted nature of MLS, the Internet provides easy access to a variety of information. Search engines (such as Lycos.TM., AltaVista.TM., and other search engines) exist that repeatedly scan the Internet (a.k.a., the World Wide Web or, simply, the Web) for content. Conventional search engines retrieve and store the textual content of HTML pages of the Web in large indexes. A user may later pass simple Boolean queries to databases created by the search engines and retrieve HTML pages relating to the submitted queries. The problem with such Boolean queries for these large indexes of stored information is that actually getting the desired information is cumbersome. Also, current systems must treat the found content as flat, unstructured data; for there is no way to create or find associations between the elements stored on HTML pages. Further, the appearance of subject-oriented searches fails to satisfy the needs of designers and like individuals interested in a narrow field of industry. To this end, current search systems are not directed to the needs of specific users.

A metadata framework has been proposed by Sakata et al. in Metadata Mediation: Representation and Protocol. Computer Networks and the Systems 29 (1997). Pp. 1137-1146. Sakata et al. discloses an ontology in a metadata framework. However, all navigation of the metadata appears to direct an end user back onto the original sources of the data on the Internet to receive requested information. Difficulties with constantly pursuing information located on multiple sites over the Internet include the occasional problems of the inability to contact desired sites due to communication breakdowns or bottlenecks. Accordingly, while Sakata et al. provides a framework for creating and expanding a metadata protocol, it relies on continuous real-time access/availability to the distributed information located across the Internet for the underlying information for the end users and is, thus, open to delays and related problems.

SUMMARY OF THE INVENTION

The present invention relates to a computer-based information organizing, searching, retrieving, and exchanging platform. Through gathering organized information, a consumer accesses the gathered information to create an end product in a shorter period of time. The present invention uses a predefined common language and format (for example, hypertext markup language (HTML) extended to implement an underlying organizational structure or XML and related schema) for organizing information placed on the network of computers. When retrieved the organized information is readily sorted and cataloged into a repository of similar cataloged information.

When placing the information in a form readily accessible by other computers (for example, on the Internet), a supplier provides the information in a format including tags which characterize the formatted information. The format may include HTML, XML, and other mark-up languages. These tags may include a "class" identifier with each class having various "attributes" and "methods". The various attributes are contemplated to include "features" (comprising character strings), "parameters" (comprising numerical strings), and "index" tags. Through listing an item as belonging to a class and having various attributes, a supplier may fully identify the item using this structured language. Also, each class may be a nested item of another class. So, each component may be broken down into its elemental parts through classifying each of its parts individually. Notably, through the combination of various parts, products may be compiled. A similar approach can be applied to processes or services and their sub components. At this point, the consumer who compiled the product may place the product on the Web as specified in a page with the appropriate tags so that a later consumer may choose to use the previous consumer's product. In other aspects, the product information may be provided to a consumer using XML packaged with envelopes such as ICE (Information Content Exchange), WSDL (Web Services Descriptor Language), WAP and other information formats.

To this end, each consumer may also be called an integrator as each consumer may integrate parts into products for later integrators.

Also, some classes may have methods or tags associated with them as well which provide procedures for invoking simple or complex algorithms or computations and compute performance or behavior of an item in the class. Because the behavior of an item typically depends on its specific use and environment, these computations require information provided by the integrator and not knowable in advance by the product supplier. Thus, method tags provide access to procedures invoked by the integrator when he evaluates possible uses of the item. Method tags contain pointers to computational algorithms, and to the "signature" of the method. A signature defines what input and output information the algorithm needs and provides, respectively.

A scanning engine scans the computers having accessible pages to locate all pages having the predefined organizational structure as including class, attribute, and methods identifiers. Through organizing the retrieved information on the basis of the class, attribute, and methods information, the retrieved information is cataloged for later retrieval. The information may conform with various protocols (including but not limited to HTML and/or XML tags) that permit easier cataloging. Catalogs are collections of the retrieved information as stored in at least one computer system referred to as a "portal." The catalogs are stored and updated as needed. Instead of storing all possible information from each page or set of information from the remote information sources (as including graphic files, text files, Java scripts, and the like) or transmitted collection of information, the catalog, for example, may only store the class, attribute, and method identifier information that allow a user to search for and identify the accessed item. The pages or sets of information may include HTML web pages and other sets of information (including information marked up with the XML or related-markup language protocol). The remainder of the accessible information is left back on the originating site for later retrieval. Through this organizational scheme, high level descriptive information (for example, information that may satisfy an initial search) may be stored in a computer that interfaces between an end user and the information source. In this instance, the more detailed information may be left back at the originating source. Thus, the most detailed information, which is the most likely to change is readily updated at the source of the information, minimizing the information which has to be provided to keep each end user, as well as each interfacing computer, current.

As described above, in one embodiment, the cataloged information is stored in at least one local computer system interfacing between the end user and the information source. For purposes of simplicity, the local computer system as storing categorized information is hereinafter referred to as a "portal" as the portal acts as a primary interface to the organized information residing on the network of computers. When a portal is located on a user's local site, the user has quick access to the variety of information stored at the site.

To this end, embodiments of the present invention contemplate the sector portal maintaining a list of pages or content at each supplier's site that comply with a predefined protocol (HTML, XML, or the like) so that each portal need only retrieve a new list of supplier sites for compliant information, rather than search the Web as a whole for compliant information. All references herein to the protocol and to the placement of the protocol on pages are intended to be inclusive in that the protocol may include, XML, HTML, and other mark-up languages that provide information about the content of the transmitted information.

Also, for simplicity, the network of computers is referred to as the Web or the Internet. However, it is understood that any networked group of computers storing organized information is included within the scope of the invention. Further, while Web pages are referenced as one form of providing information to users, other techniques of providing information to users are specifically included. For example, one may use a device that does not display HTML pages (for example, some Internet-ready telephones as are known in the art) but may display text resulting from a web-clipping service. Also, devices may display received information having been marked up using XML (extensible markup language).

The various users of the present invention may be categorized as information suppliers and information consumers. As will be described in greater detail below, as the information consumers in turn supply information to other upstream information consumers, every information consumer may be considered a supplier to other consumers.

To enable each supplier to provide requisite information on its pages, a "sector" portal establishes common terms (class, attribute, and method names) for the suppliers and consumers to use. The sector portal is so named because each industry sector is contemplated to have at least one governing portal from which all other portals in that industry sector derive their common vocabulary, taxonomy, or ontology (collectively, referred to herein as organizational terms). For reference, an ontology (as used herein) relates to a set of terms that are specified in some way to establish a conceptualization of a concept, specifying the names of the objects and characteristics of the concepts and indicating the hierarchical relationships of its pieces as is known in the art. Each sector portal has the ability to update its own list of accepted classes, attributes, and methods (generally referred to as descriptors). So, when a supplier begins to use new classes, methods, and attributes, the sector portals selectively update their ontology to include the new classes, methods, and attributes according to a predetermined scheme. For example, a predetermined scheme may include updating the ontology only after a new class, attribute, or method is used a predetermined number of times (e.g., ten times). An advantage of not updating the ontology for every possible new class, attribute, or method is that 1) mistakes will be made by suppliers in not using the predefined ontology so that the ontology should not necessarily include the mistakes of suppliers and 2) an ontology expanding too quickly may fail to minimize the number of terms used to describe a single item. To this end, the ontology should not necessarily include multiple listings of the same class or attribute when a single one may be used across suppliers.

Alternative definitions of an ontology are known and may be used in various implementations.

Once the catalogs are established at a portal, a consumer may use at least one of three search strategies to retrieve the stored information including Boolean searches, hierarchical searches (for example, where a consumer navigates through a hierarchy of classes), and parametric searches (which may include fuzzy searches).

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows an organization diagram of various suppliers and consumers according to embodiments of the present invention.

FIG. 2 shows the configuration of an industry sector portal according to embodiments of the present invention.

FIG. 3 shows a low-end supplier with browser access according to embodiments of the present invention.

FIG. 4 shows a high-end supplier with a page generator according to embodiments of the present invention.

FIG. 5 shows a consumer according to embodiments of the present invention.

FIG. 6 shows an architectural representation of sector and local portals, according to embodiments of the present invention.

FIGS. 7A-B show relationships between items in an ontology according to embodiments of the present invention.

FIG. 8 shows relationships in a database model according to embodiments of the present invention.

DETAILED DESCRIPTION

The present invention will now be described in detail with reference to the accompanying drawings. While the present invention is described in the context of a communications network including a specific number of components, the cataloging search and retrieval system and integrated tool suite according to the present invention may be incorporated into networks of many structures and sizes. The drawings are intended to provide one example of a network configuration in which a cataloging search and retrieval system may be implemented and are not intended to limit the applicability of the present invention to other network configurations.

FIG. 1 shows a high level functional arrangement of the contributors and users of the organized information. The network of computers, in this example, is shown as the Internet or Web 101. Portal 102, in one aspect of the present invention, functions as an entry way into the stored catalogs of information. As will be described in greater detail below, portal 102 may reside at a number of locations. Also multiple portals may exist for an industry. In the case of a single portal 102 for an industry, the portal 102 is referred to as the sector portal 102. In the case of multiple portals 102 (for example, each company having its own portal 102), then one of the portals 102 may be considered the sector portal. The sector portal may maintain control over the accepted ontology. Portal 102 is contemplated to include at least one processor and at least one storage device. Through the implementation of stored program modules, the portal (or portals) 102 may control the collection and dissemination of information as described in greater detail with respect to FIGS. 2-6.

Also shown in FIG. 1 are suppliers (104 and 105) of information. The suppliers (104 and 105) of information provide information to portal 102 for use by user 103. User 103 uses the information stored in portal 102 to make decisions on the information as well as retrieve greater amounts of information, as needed from suppliers 104 and 105. Further, user 103 includes a tool suite which uses the information in portal 102 and at providers 104 and 105 to provide additional functions to the user including, inter alia, the availability of alternative decision trades, the likelihood of success of selected components meeting an overall design criteria, etc. To provide information to portal 102, suppliers 104 and 105 make their information available for retrieval via the Web 101 by making available such information through Web pages in HTML, XML, or other markup language format or other Web-accessible storage media using similar formats on a server connected to the Web 101. As only some of the entities (business, organizations, etc.) which access the Web 101 have the ability to provide the server technology to Web-enable their information, suppliers 104 and 105 are separated into two groups: suppliers (104) who do not have the capability to provide Internet access and the suppliers (105) who do have the capability. As for suppliers (104) who cannot Web-enable their information, the system allows suppliers 104 to upload their product information (and other information) to an ISP for page hosting purposes. Accordingly, suppliers 104 are considered low-end suppliers. The functionality of uploading pages from suppliers 104 is treated in greater detail with respect to FIG. 3. As for suppliers (105) who can Web-enable their information, the system accommodates their ability as well. The functionality of Web-enabling from suppliers 105 is treated in greater detail with respect to FIG. 4.

FIG. 1 further shows user 103 using information stored in portal 102. Embodiments of the present invention contemplate portal 102 not storing every piece of available information in portal 102 as the amount of storage space required may quickly become overwhelming. So, portal 102 stores some information that may answer some initial questions for user 103 and point to suppliers 104 and 105 for additional information. As user 103 consumes the information provided over Web 101, user 103 may be referred to as an integrator as it integrates the available information. Notably, the resulting data sets combined by user 103 may be placed in another portal 102 for another user's access. In this instance, user 103 becomes another supplier. For simplicity, the groupings of the entities using the disclosed system are referred to collectively as "players". The labels applied to the various players in the system (supplier, user, consumer, integrator, etc.) are intended to be relative to the surrounding players and not intended to be an absolute definition of the player.

To provide information to portal 102, suppliers 104 and 105 (and user 103, when applicable) encode their pages using a predefined protocol. Use of the protocol encourages placing available information in an organized format. The protocol may include tag codes which describe the information contained therein. For example, an HTML-based tag protocol may include the following information:

    <UC domain="product" class="gimbals/shanks/widgets">
    <UC_ENTRY name="standard widget" realization state="available">
    <UC_FEAT name="Supplier Name" value="Acme Corporation">
    <UC_FEAT name="Manufacturer Name" value="Acme Corporation">
    <UC_ID name="part number" value="1234">
    <UC_PAR name="Price" value="$100">
    <UC_PAR name="Length" value="2.5" units="inches">
    <UC_PAR name="Width" value="3.5" units="inches">

    <?xml version="1.0"?>
    <?xml-stylesheet type= ' text/xsl ' href= ' /UCLP.xsl ' ?>
    <UC domain="Products" version="2.0" class="Fuze/Proximity/Stand
    Alone/RF/Guided Missile" status="current">
    <UC_ENTRY name="6665005" realization_state="available">
    <UC_ID Name="Product Name" value="Target Sensor"/>
    <UC_FEAT name="QualifiedTo" value="ISO9001"/>
    <UC_ID name="Supplier Number" value="94990"/>
    <UC_PAR name="Price" value="Proprietary" units="Dollars"/>
    </UC_ENTRY>
    </UC>

    Subclass:        Rectifiers
    Rating:             IF AV         VRRM         IFSM      TJ
    Desired Values:       5            75           215      180
    Relative            very         little        very     most
    Importance:
    Ranking Criteria: more is better more is better less is better exact

        Ensemble    A product which includes this product;
        Ancestor    Previous version of this product;
        Descendent  Updated version of this product;
        Component   A component used in this product;
        Accessories Other products which go the with the product; and,
        Alternatives Other products which might also meet requirements.

• Inventors
  Chipman, Richard R.; Mankofsky, Alan; Karandikar, Harshavardhan M.; Warren, Gary; Laskey, Kenneth J.;
• Assignee
  Science Applications International Corporation (San Diego, CA)
• Published
  Sep-28-2004
• Current US Classes:
  707/1
  707/2
  707/3
  713/201
• Application #
  953228
• International Classes
  G06F 017/30; G06F 011/30
• Field of Search
  707/1-3 707/6-7 707/8 707/10 707/102 707/103 713/200-201 705/75-76
• Examiner
  Kindred; Alford
• Agent
  Banner & Witcoff, Ltd.
• US Patent References:
  5649186
  5694523
  5694593
  5710887
  5737592
  5832494
  5848410
  5890175
  5895470
  5913214
  5926798
  5940807
  5946665
  5953716
  5963965
  5970475
  5970490
  5974412
  5978847
  5995943
  6009413
  6014639
  6016504
  6023683
  6026429
  6101515
  6112201
  6573907