Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-10-02
2001-09-25
Hong, Stephen S. (Department: 2176)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000
Reexamination Certificate
active
06295542
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to a method and apparatus for cross-referencing text, and particularly, but not exclusively, to a computer program for automatically inserting hypertext links into legal documents, for example.
BACKGROUND OF THE INVENTION
Many forms of legal document such as joint venture agreements, construction contracts, licences or materials supply agreements are created by companies in the conduct of their business. For example, The Pooling and Settlement Agreement for England and Wales which governs the trading of electricity has over 7000 clauses, 5900 cross-references, and 2000 definitions.
Frequently the documents may be interrelated, such as those which deal with different aspects of a new power station project, for example: joint venture, power purchase, fuels supply, construction, and operation and maintenance agreements for which consistency in the usage of defined terms, dates, liabilities and so forth across the agreements is important.
It is a huge task to prepare and check all the documentation, often involving many people. Usually, there is time pressure in the preparation of the documents especially during negotiations. The documents may need to be redrafted many times to accommodate the changes, any of which may lead to errors.
Only now are systems coming to market which are starting to help contract draughtsman. Although word processors have been in use for quite a few years, the level of machine checking has been limited to the use of a “spell checker”, and “document comparison” to mark changes between different versions.
With the advent of the Internet and the worldwide web, the availability and popularity of a formatting language called HTML has increased dramatically. Many current commercially available programs, such as Word Processing and Internet Browser Software sold by Microsoft®, can generate and read HTML.
HTML is an acronym of Hypertext Mark-up Language and is a particular form of a general mark-up language. The HTML language employs a series of standardized codes inserted into unformatted (typically ASCII) text. These codes, being standardized, can be read and understood by a wide variety of programs, operating on many different platforms. The text may thus be displayed in a formatted manner according to the codes embedded within the text.
One particular feature of HTML is the ability to create and view multi-dimensional electronic documents, which may be entered at many points and which may be browsed in any order by interactively choosing words or by phrases as search parameters for the next text or image to be viewed.
The words or key phrases which permit the user to jump quickly and easily to related text in other documents (or indeed within the document itself) are known as “hypertext links ”or “hyperlinks”. As with the formatting of the text itself, these hyperlinks are inserted using HTML mark-up tags.
Legal documents are naturally very rich in cross-references which can be made into hypertext links.
As will be familiar to those skilled in the art, hyperlinks are created in HTML by “tagging” chosen words or phrases with code. This code, which is of course hidden (i.e., not shown in the displayed text), defines the address of, for example, a document held on another server elsewhere, or that of a document held in the same directory as the current document, or even another position within the same document. In the latter case, the code associated with the displayed text forming the hypertext link is given a name by the user. When the hypertext link is activated (usually by clicking a mouse button), the programme reading and displaying the HTML code searches for a marker elsewhere within the document, the marker being the name given by the user to the hypertext link.
All current commercial authoring tools for creating documents with hyperlinks require the author to insert the links by hand. For contracts with many hundreds of possible links it is impractical to undertake this task manually.
Prior work on the automatic creation of hypertext links in hypermedia has been concerned with creating links between concepts contained in different documents. The problem has then been to analyse each document to find the key words or phrases which describe what the document is about, then to find similar documents for linking to.
As with, for example, speech recognition, sophisticated artificial intelligence is required to allow a program to understand ambiguous words or phrases in context and thereby correctly insert links. Currently, progress in this field is slow and is largely academic.
Anonymous Research Disclosure No. 350009, published Oct. 6, 1993, discloses a technique for creating automatic hyperlinks for reference documents. The hypertext generation program is written to read a referenced document and convert as many of its footnotes and bibliography references to multimedia link markers as possible. The program expects the footnote and bibliographical format used to be input along with the name of the document to be converted to multimedia hyperlinks. The program scans the document looking for footnote references or bibliographical references.
When a reference is found, a search is commenced with the reference passed on to a multimedia library system. If the library system finds the reference, it passes back the information to a hypertext generation program. The program then saves the information that allows the reference to be later hyperlinked. In addition, the conversion program saves information that allows the reference (footnote marker, bibliographical description) to be highlighted so as to indicate to a researcher (user, reader, etc.) that the reference is activated and available for immediate linking.
It is an object of the present invention to provide an improved technique for insertion of automatic hypertext links.
SUMMARY OF THE INVENTION
In a first aspect, the present invention accordingly contemplates a software program for cross-referencing text within a document, the program carrying out the steps of:
(a) locating label strings within said document, each label string labelling an associated part of said document;
(b) converting each said located label string into standard format such that the said associated parts of said document are each unambiguously labelled by said standard format;
(c) locating cross-reference strings within said document, said cross-reference strings referring to parts of said document labelled by said label strings;
(d) converting each said cross-reference string into said standard format such that said converted cross-reference strings refer unambiguously to one or more labelled parts of said document;
(e) collating said standard format cross-reference strings with said standard format label strings; and
(f) providing a program link between said cross-reference strings and those labelled parts of said document to which they refer.
The conversion of label strings into a standard format allows previously ambiguous label strings, such as section headers for example, to be uniquely identified. By also converting cross-references into a standard form, labels and cross-reference strings may be matched up with one another so that hypertext links between the two, for example, may be created automatically. To mark up a lengthy document with hyperlinks manually can take many man hours, and the time saving provided by the present invention is significant.
The program preferably locates the label strings and cross-reference strings by parsing the input document into constituent character groups using a predefined set of rules. Specific character strings are sought for locating particular types of label strings, for example paragraph headers, which can act as hypertext link anchors. These are recognized by seeking patterns of characters that comply with predefined rules.
According to a second aspect of the present invention, there is provided a software program for cross-referencing text within a plurality of documents, the program carrying out the steps of:
(a) locating label string
Haynes and Boone LLP
Hong Stephen S.
National Power PLC
LandOfFree
Method and apparatus for cross-referencing text does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for cross-referencing text, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for cross-referencing text will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2467417