Remote access of heterogeneous data

• US 7,882,122 B2
• Filed: 03/17/2006
• Issued: 02/01/2011
• Est. Priority Date: 03/18/2005
• Status: Expired due to Fees

• Current Assignee: State Capital Resources Limited
• Original Assignee: Capital Source Far East Limited
• Inventors: Wong, Kwok Kay
• Primary Examiner(s): Mahmoudi; Tony
• Assistant Examiner(s): Uddin; Md. I
• Application Number:
  US11/909,023
• Publication Number:
  US 20090234823A1
• Time in Patent Office: 1,782 Days
• Field of Search: 707/708, 707/713, 707/718, 707/719, 707/765, 707/722, 707/759, 707/760, 707/763, 707/769, 707/770, 707/771, 707/776, 707/802, 715/700, 715/255
• US Class Current: 707/760
• CPC Class Codes: G06F 16/9577 Optimising the visualization of content, e.g. distillation of HTML documents

Abstract

A method and system for providing remote access to heterogeneous information sources from computing devices including but not limited to personal computers, laptop computers, PDAs, cellular telephones is presented. A middle-tier server is provided in a multi-tier environment for allowing a data-centric approach to accessing, modifying and updating content located at remote data sources. The middle-tier server provides functionality for a mobile (or similar) user to perform a search query, request for document and/or data, update remote data, and perform commands on remote data sources. In addition, the middle-tier server manages the authentication of the user in sessioned or sessionless mode. In particular, search query results are processed such that the results answer a search query. That is, rather than simply providing a list of document matches, the middle-tier server extracts information from the document that answers the search query most appropriately. Finally, the middle-tier server, using user preferences, device capabilities and settings, transforms the result, document or data into an appropriate format that is suitable for the device, user and communication network. This allows the user to easily process and view the document on a (possibly limited capacity/functionality) device.

Claims

1. A computer implemented method for accessing information in a distributed networked environment having multiple clients, a middle tier server, and one or more remote data repositories, the method comprising:

allowing access to said middle tier server for client queries;

processing client queries, including converting said queries to native query languages supported by the appropriate remote data repositories, and dispatching said native queries to said remote data repositories;

processing of native query results, including integrating said native query results and converting said native query result into one or more data views, where a data view is a single presentation of heterogeneous information content in a homogeneous form;

wherein said native query result comprises a list of entries, wherein converting said native query result into one or more final data views comprises;

identifying the form of said native query result, wherein said form describes the presentation format of said native query result in terms of its layout and structure;

identifying the form of the final result, wherein said final result is based on the requirements of said client profile, preferences, settings and parameters of the client mobile device;

determining the number of said final data views, wherein said number of data views is the minimal number of distinct data views required to display all the said final results;

transforming said native query result into said form of the final result; and

collating said final results in said final data view;

wherein said collation involves ranking and duplicate elimination;

wherein each entry is ranked using its said weighted value;

wherein said transforming of said native query result comprises;

for said native query results that are not of embedded form, a shallow transformation of each entry from said native form to said final result is performed;

wherein said embedded form is a form with a list of forms embedded in said form;

wherein said shallow transformation involves changing said annotation of the type of native result form from the original to the target form;

for said native query results that are of embedded form, wherein each entry does not have the required search term data type, a shallow transform of each entry from said native form to final result is performed; and

for said native query results that are of embedded form, wherein each entry has the required search term data type, said embedded entries are decomposed based on said data type and a shallow transform of each decomposed entry from said native form to final result is performed.

2. The method of claim 1, further comprising:

determining client specific information, including but not limited to, display configuration, list of installed software and their configuration and user profiles and configuration, of said client;

accessing and maintaining client specific information in said middle tier server.

3. The method of claim 2, further comprising transforming said data view to an appropriate data view format based on said client specific information in said middle tier server.

4. The method of claim 2, further comprising:

adaptively compressing said data view based on said client specific information and network bandwidth in said middle tier server; and

if necessary, decompressing said data view in said client.

5. A method of claim 2, further comprising display and layout optimization;

wherein said display and layout optimization is a client program residing on said client;

wherein said display and layout optimization includes the following steps;

transforming said final result from said middle-tier server to an appropriate format based on said mobile device display, user settings and preferences; and

caching and storing the history of queries, links and results that were requested.

6. The method of claim 1, wherein:

said distributed networked environment includes a mobile network environment having multiple mobile devices;

or

said middle tier server and said client reside in a single computer system;

or

said middle tier server and said remote data repositories reside in a single computer system;

or

said middle tier server, said remote data repositories and said client reside in a single computer system;

or

said remote data repositories and said client reside in a single computer system;

or

said middle tier server and said remote data repositories reside in each client;

wherein said distributed networked environment is a peer to peer network;

or

said distributed network environment further comprises an authentication server.

7. The method of claim 1, further comprising preprocessing said client query;

wherein said preprocessing comprises;

parsing said client query into a list of search terms;

caching said client query;

stop word filtering of said parsed client query;

stemming and query expansion of said parsed client query; and

annotating each search term in said parsed client query, wherein said annotation represents the type and format required for said search term.

8. The method of claim 1, wherein said query dispatching and search comprises:

determining target remote data repositories;

performing a metadata search on said target remote data repositories;

manual or index search on filtered target remote data repositories; and

post processing and caching of said native search results.

9. The method of claim 8, wherein:

said target remote data repositories are determined by specifying said keyword search for selecting data repositories;

or

said metadata search performs a search on the document name, database name, author name or creation date of the contents of said data repositories;

or

said metadata search performs a search on the field names of tabular data repositories.

10. The method of claim 8, wherein said manual search comprises:

identifying the search methods for each target remote data repository;

performing a tabular search for said target remote data repository with tabular form;

performing a free form search for said target remote data repository with free form; and

performing a loosely structured search for said target remote data repository with loose structure.

11. The method of claim 8, wherein said post processing further includes annotating each said term in said native search result with said repository identifier, query identifier, data type, display format and editable field;

wherein said query identifier is the native command used to perform the search;

wherein said data type and display format is obtained from said type and format definitions file;

wherein said editable field identifies whether said term can be updated or modified by a user.

12. The method of claim 10, wherein said tabular search is one or more queries on each table;

wherein said query is only performed on the columns of the tables that have matching search term type names.

13. The method of claim 10, wherein said free form search includes the following steps:

identifying the minimal list of minimal text regions;

wherein said text regions are sentences or paragraphs, wherein said text regions contain at least one match corresponding to each said search term; and

expanding said text region to search for matching said search term data types, if said search term data types are not found within said text region.

14. The method of claim 10, wherein said loosely structured target remote data repository is internally represented as a tree hierarchy, wherein each node in said tree hierarchy represents data of a basic form, wherein said form identifies the presentation format of a piece of data in terms of its layout and structure, further comprising:

traversing said tree hierarchy to identify match regions;

wherein said match regions represent the minimal subtree;

wherein said minimal subtree has descendant nodes that include at least one result node corresponding to each search term.

15. A method of claim 1, wherein said data view is augmented with direct links, wherein said direct links enable navigation to the data in said data repository, wherein said data are the results of said query.

16. The method of claim 1, further comprising enabling client commands to be executed from said client, wherein said client commands are maintained in a list of mappings, matching a unique keyword to one or more API calls, modules or commands that can be executed by said middle tier server;

wherein said client commands include browsing the contents of a data repository, user account administration, file management and saving search results locally.

17. The method of claim 1, further comprising data transformation of said data view or said data in said data repository to a format that can be presented by said client, wherein said data transformation comprises the steps of:

specifying transformation clusters, wherein said transformation cluster includes document formats and links between said document formats, wherein said links represent the ability to perform transformations between said formats;

identifying said transformation cluster corresponding to said original format;

identifying said transformation cluster corresponding to said required format;

determining the set of transformations required;

wherein said set of transformations can be described by an ordered list of said links to be traversed in order to convert the data from said original format to said required format.

18. A method of claim 1, further comprising processing client check-ins wherein said client check-in is one of:

updating a data value, including updating the value of a portion of said data in said data repository;

saving an associated version, wherein said associated version is stored in said data repository as a modified version of the existing data in said data repository;

overwriting existing data in said repository in the format of the check-in data;

overwriting existing data in said repository in the native format of said data.

19. A method of claim 1, wherein said collating of said final results for said final data view of tabular form further comprises:

identifying the number of search terms that have data types;

identifying the number of data types in each entry result;

calculating the number of columns required in said final result; and

partitioning each entry result into said number of required columns.

20. A method of claim 1, further comprising type coercion and data translation for the said data views, wherein said type coercion and data translation converts each entry in said data view to a consistent format across entries;

wherein the most common display format or preferred display format is the target display form for each data type.