Automatic search method

• US 20040030692A1
• Filed: 08/15/2003
• Published: 02/12/2004
• Est. Priority Date: 06/28/2000
• Status: Active Grant

• Current Assignee: ZAMA Innovations LLC
• Original Assignee: Thomas Leitermann
• Inventors: Leitermann, Thomas
• Granted Patent: US 7,080,068 B2
• Time in Patent Office: Days
• US Class Current: 707/4
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: Y10S 707/959 Network
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

An automatic search method for documents stored in a data processing system is presented. The method involves preselecting large amounts of data automatically thereby enabling the user to find relevant information easily. The task of comparing search string and document content is assigned to the user who does not need to formulate any query statement but browses the documents by comparing their representatives and choosing the ones which fit the query statement. Search terms comprise representatives of the documents, extracts of the documents, or synonyms of the extracts which function as nodes of the semantic network wherein the edges are set up by various relations between the search objects, as for example, detailing relations, thesaurus relations, and combining relations. Movements along the edges are simulated by retrievals of the data storage resulting in new nodes presented to the user in a viewing device of the system.

Claims

1. Method for automatic searching and synchronous content display of a predefined set of documents in a displaying device in form of a list, characterized in that for each document

encapsulated text extracts of various length are taken from the documents or synonyms of the said text extracts in form of natural language,

the said text extracts or their synonyms are stored as data objects or—

entities in a persistent data store e.g. as records in a database,

detailing relations are created defining couples of the said objects wherein one object states more precise the meaning of the other,

the said detailing relations are mapped by attributes of the objects containing references to longer, enclosing and therefore more precise text extracts (shorter, enclosed and therefore less precise text extracts respectively), thus creating chains of text extracts from the mapped detailing relations which start for example with words extracted from documents and end with headlines and

assigning each of the text extracts of the said chains to a longitudinal class enabling queries of the data store by a computer displaying a text extract assigned to a longer class after selection of a text extract assigned to a shorter class and thus gradually displaying an increasing part of the document and representing its content more and more detailed.

2. Method of claim 1, characterized in that the detailing relations to text extracts belonging to a certain longitudinal class are represented by certain attributes which are each specific for the longitudinal class and which represent all detailing relations to objects of that class, including thus

enabling the computer to query for each text extract all available detailing relations to extracts of the shortest longitudinal class, reading the related text extracts from the persistent data store and displaying them on the viewing device, and

displaying the same link of all said chains of text extracts and therefore representing all documents equivalently to the same degree of detail,

enabling the procedure to reach the end of all chains after a couple of queries at the same time.

3. Method of claim 1, characterized in that—

besides of the said detailing relations (connections with longer text extracts, with the documents and its profiles)—

also thesaurus relations (connections with superordinated, associated and subordinated text extracts) and combining relations (connections with text extracts appearing respectively not appearing in the same document) which are as well represented as references each by a specific attribute of the data objects to be queried by the computer.

4. Method of claim 1 or 2, characterized by an attribute of the data object mapping a relation between the longest link of the said chain of text extracts and one or more documents or document profiles so that any text extract is assigned to a document or document profile directly or indirectly by a text extract of the longest longitudinal class.

5. Method of claim 3, characterized in that the user does not need to formulate the query but only has to choose text extracts and actions in a displaying device, wherein the underlying data base queries are generated automatically, wherein the text extracts which have to be selected by the queries from the data store are connected with the chosen text extracts by any relation defined in claim 1 to 3 corresponding to the chosen action.

6. Method of claim 1 to 5, characterized in that the same text extracts or their synonyms appearing in various documents are only once displayed, wherein all connections of text extracts include connections of synonyms.

7. Method of claim 1 to 6, characterized by displaying text extracts connected by the detailing relation in form of a list close together resulting in an ergonomically perceivable order of the text extracts

8. Method of claim 1 to 4, characterized by a classification grouping documents by document classes which can be

connected logically with the data objects of the text extracts representing the document so that a connection of a data object with a document class corresponds to the logical relation with the documents contained therein

or assigned directly by one or more data attributes to data objects representing documents or document profiles of a certain document class.

9. Method of claim 8, characterized in that the document classes define non-thematic characteristics in a certain combination defining constraints for the search for documents, document profiles and text extracts

10. Method of claim 1 to 9, characterized in that the relations between the data objects can be either local or global depending on the assigned document classes;

global relations are valid for all documents, whereas local relations are only applied to documents assigned to a certain document class so that only text extracts are displayed at a query connected with a chosen text extract by a relation applied to the restriction.

11. Method of claim 1 to 9, characterized in that parts of the retrieved document sets are transfered to other logical or physical data storages (e.g. at archiving on a client system) together with their logically related data objects so that the search features are kept intact.

12. Method of claim 1 to 11, characterized in that all display tasks (comprising displaying information in a result list) are performed by document profiles containing text extracts of the document, document-related characteristics, system-related characteristices and a link to the document (reference) so that the document is only loaded if required by the user.

13. Method of claim 1, characterized in that document profiles can additionally be retrieved by classification structures, especially by directories in form of matrices and network plans being related logically to the document profiles.