EP1002284A1 - A method and apparatus for authoring of customizable multimedia documents - Google Patents
A method and apparatus for authoring of customizable multimedia documentsInfo
- Publication number
- EP1002284A1 EP1002284A1 EP98939453A EP98939453A EP1002284A1 EP 1002284 A1 EP1002284 A1 EP 1002284A1 EP 98939453 A EP98939453 A EP 98939453A EP 98939453 A EP98939453 A EP 98939453A EP 1002284 A1 EP1002284 A1 EP 1002284A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- document
- ofthe
- class
- annotations
- html
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/221—Parsing markup language streams
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/55—Rule-based translation
- G06F40/56—Natural language generation
Definitions
- This invention relates to a method and apparatus for the authoring of customizable multimedia documents and the adaptive generation of versions thereof for particular uses.
- Natural Language Generation is a young but growing research field, whose goal is to build computer systems that automatically produce fluent and effective texts in various human languages.
- NLG systems have used knowledge databases containing general world knowledge and specific domain knowledge, together with various linguistic resources (e.g., lexicons, grammars, discourse relations), to produce texts with limited variation in word choice, sentence and discourse structure, and virtually no variation in rhetorical style or pragmatic purpose.
- a master document may refer to the complete superset of instructions to direct the actions of a robot on an assembly line.
- this process of adaptive document generation should be easily implementable on a computer system at minimum possible cost and maximum possible ease of use to both the author ofthe master document and the user ofthe generation system.
- IDAS The IDAS project (Reiter, Mellish, and Levine 1995) recognized the need to tailor both textual and non-textual information, including visual formatting, hypertext input, and graphics output. IDAS also tried to address the need for explicit authoring tools in the adaptive document generation process, but here the focus was on authoring at the knowledge-base level (i.e., at the level of a computer system's internal representation), while there still exists a need to provide an authoring tool that may be used by a non-computer-programmer professional writer who could compose the master document at the level of ordinary English, with additional markup as required (e.g., HTML markup to support an HTML presentation format for a resulting customized version ofthe document).
- knowledge-base level i.e., at the level of a computer system's internal representation
- additional markup e.g., HTML markup to support an HTML presentation format for a resulting customized version ofthe document.
- IDAS relies mainly on canned texts and aims to provide the user with a means of navigating through the whole "hyperspace" of possible (canned) texts. There is however a need to provide for a much finer-grained degree of tailoring than the IDAS implementation. While IDAS relies mainly on canned texts, other adaptive generation systems do use more-dynamic text generation: the Migraine system (Carenini, Mittal, and Moore 1994) uses an approach to text planning that adaptively selects and structures the information to be given to a particular reader. However, Migraine relies on a large number of context-sensitive and user-sensitive "text plans" (i.e., text schemas) so that its methods of tailoring must of necessity be very specific to its particular domain.
- context-sensitive and user-sensitive "text plans” i.e., text schemas
- the PEBA-II system uses more-general text plans, as well as text templates, that it can choose from to adapt information to the individual reader, but the tailoring done is very specific, focussing on the user's familiarity with a topic.
- the PIGLET system (Cawsey, Binsted, and Jones 1995) also uses a combination of text plans and text templates, but its tailoring is also quite specific in nature, mainly concerned with emphasizing material that is relevant to a particular patient.
- the ILEX-0 system (Knott, Mellish, Oberlander, and O'Donnell 1996) is similar to the PIGLET model in its anticipation of all the possible texts that might be generated, but also includes annotations (e.g., a condition on a piece of canned text) to allow some local customization.
- annotations e.g., a condition on a piece of canned text
- very free and flexible use of annotations could lead to problems of repetitive text and inappropriate use of referring expressions in the resulting document, requiring textual repair.
- the system should be able to support either an adaptive generation system with full facilities for selecting and repairing texts, as described by DiMarco, Hirst, Wilkinson, and Wanner (1995) and Hirst, DiMarco, Hovy, and Parsons (1997), or a simpler version ofthe system, based on "generation by selection only", i.e., with no facilities for textual repair, an implementation of which (called “WebbeDoc”) is described by DiMarco and Foster (1997).
- an author of a customizable document needs to be able to describe the variations of a document, which may be both textual and non-textual, at various levels ofthe document structure, together with the conditions for selecting each variation.
- the author then needs a means of selecting all the appropriate variations for a particular purpose or audience, re-assembling the selected variations into a coherent document, and producing an appropriately customized version ofthe document, in potentially many different levels of representation (e.g., surface English, a deep syntactic or semantic representation for use in textual repair) and presentation formats (e.g., HTML, LaTeX).
- levels of representation e.g., surface English, a deep syntactic or semantic representation for use in textual repair
- presentation formats e.g., HTML, LaTeX
- This invention seeks to provide a computer system for customizing an initial master document containing information for a multiplicity of versions ofthe document intended for the different purposes or different users, for a specific purpose or for a specific user.
- a computer system for customizing an initial master document in accordance with a user-defined set of purpose parameters.
- a data structure i.e., the customizable, "master”, document
- a master document will therefore contain all the information that the system might need to include in any particular customized version ofthe document, together with annotations giving the selection conditions as to when each piece of information is relevant and other annotations giving linguistic and formatting information, including for multimedia elements ofthe document.
- a further aspect ofthe invention provides for a method and apparatus for reading said data structure into a form implementable on a suitably programmed processor such that the implementation ofthe data structure can store both the form and content of a master document, i.e., all the elements ofthe document and their variations, and can also act as the process for selecting the relevant variations ofthe document, according to given values of input parameters specifying the intended purpose or intended user, and then generating the appropriately customized version of the document.
- Figure 1 is a schematic diagram showing the architecture ofthe system
- Figure 2 shows a generalised form of a data structure for specifying a customizable, master, document according to an embodiment ofthe present invention
- Figure 3(a) shows a generalised form of a data substructure for use within the main data structure for specifying linguistic or presentation format information for a component of the master document
- Figure 3(b) shows a generalised form of a data substructure for use within the main data structure for specifying hypertext links to parts ofthe main data structure or to other data structures ofthe form as specified in Figure 2 and provided in other source datafiles;
- Figure 4 is a flowchart ofthe overall process of reading-in as input the data structure and generating as output a final customized version ofthe document;
- Figure 5 is a graph showing the resolution process for a customized document generated according to an embodiment ofthe present invention.
- the system comprises: a datafile 12 including a data structure 13 for defining a master document; a parser 14 for reading in the contents ofthe datafile 12 and for creating instances ofthe document-class data structures 18 in accordance with the general definitions of document-class data structures 16; a user input interface 20 for reading new values of purpose parameters 22 which are input by a selection engine 18.
- the selection engine 18 uses the current values ofthe purpose parameters to select the relevant variations of each component ofthe document and to generate appropriately customized versions 26 of a document, which may also include hypertext links to new documents 28, which may themselves be customizable documents ofthe form illustrated in Figure 2.
- purpose parameter means a parameter used in evaluating a selection condition associated with a particular variation of a document structure, where this parameter can be used in defining either a particular intended purpose or use of a customized version ofthe document or a particular intended user or audience for a customized version.
- FIG 2 an embodiment ofthe data structure 13 according to the present invention is described, which shows the blocks, sub-blocks, fields, and subfields ofthe data structure.
- the data structure 13 has the following main blocks of information: 1. Identification of purpose parameters and representation-level parameters, and their possible values.
- Block 1 The purpose parameters.
- the first block ofthe data structure identifies the purpose parameters, or user parameters, together with their possible values, that can be used in forming the Boolean expressions that give the conditions for selecting each variation of a document.
- the first block also identifies the representation-level parameters, together with their possible values, that can be used in forming the Boolean expressions that give the conditions for selecting each desired level of representation ofthe sentences in the master document during the process of generating a customized version ofthe document.
- Block 2 The toplevel object.
- the toplevel object identifies the document-class instance which is the "root" element ofthe entire document. It is with this object that the resolution process for generating a customized version ofthe document begins.
- Blocks 3-9 The program structures. Blocks 3-9 describe the program structures, the classes that implement the substructures ofthe data structure that specify the form and content of a customizable document. The program structures are related in the following manner:
- a datafile describing a set of program structures is a particular example of a customizable document created for various uses.
- the datafile may be divided into various parts.
- the data structure may be divided into components, or elements, referred to as the classes Document, Section, Topic, Sentence, and Lexical, which each implement a substructure ofthe data structure that defines a component of a customizable document.
- Each such substructure ofthe data structure also includes the variations of a component and the conditions for selecting the appropriate variation of a component.
- the data structure contains basic components which are instances ofthe classes Word and Annotation, and other components, which are instances ofthe class External, for linking to other datafiles.
- Block 3 The Documents.
- Block 3 describes the instances ofthe class Document.
- Each Document description must specify the following properties: - A list of its variations.
- Each variation must be an instance of the class
- Annotation An instance of the class Annotation can specify both textual and non-textual properties of a document or a component of a document in terms of a particular document-layout format, structure, or linguistic representation.
- an Annotation object could specify multimedia elements ofthe document's layout, such as alignment of text, font size, background colour, text colour, and graphics; other Annotation objects could specify linguistic information such as discourse relations or coreference links.
- Each variation of a Document class is then described as an instance ofthe class DocumentVariation.
- Each DocumentVariation description specifies the following properties:
- the condition for selecting this variation must be a Boolean expression composed from pairs of purpose parameters and their allowable values.
- Block 4 The Sections. Block 4 describes the instances of the class Section. Each Section description specifies the following properties:
- Each annotation must be an instance of the class Annotation.
- Each variation of a Section class is then described as an instance ofthe class SectionVariation.
- Each SectionVariation description specifies the following properties:
- the condition for selecting this variation must be a Boolean expression composed from pairs of purpose parameters and their allowable values.
- Block 5 The Topics. Block 5 describes the instances ofthe class Topic. Each Topic description specifies the following properties:
- Topic-Variation A list of its annotations. Each annotation must be an instance ofthe class
- TopicVariation Each variation of a Topic class is then described as an instance ofthe class TopicVariation.
- Each TopicVariation description specifies the following properties: - The condition for selecting this variation. The condition must be a
- Boolean expression composed from pairs of purpose parameters and their allowable values.
- Block 6 The Sentences. Block 6 describes the instances ofthe class Sentence. Each Sentence description must specify the following properties: - A list of its variations. Each variation must be an instance ofthe class
- the condition for selecting this variation must be a Boolean expression composed from pairs of purpose parameters and their allowable values. - A list ofthe components of this variation.
- Each annotation must be an instance ofthe class Annotation.
- Each SentenceRepLevel description must specify the following properties:
- the condition for selecting this sentence representation must be a list of one or more representation-level parameters.
- an instance ofthe class SentenceRepLevel may be a character string, with any Lexical components identified by surrounding reserved characters. This is a simplification made for ease of testing system prototypes, and does not limit the scope ofthe invention.
- Block 7 The Lexicals. Block 7 describes the instances of the class Lexical. Each Lexical description must specify the following property:
- LexicalVariation Each variation of a Lexical class is then described as an instance ofthe class LexicalVariation.
- Each LexicalVariation description must specify the following properties:
- the condition for selecting this variation must be a Boolean expression composed from pairs of purpose parameters and their allowable values.
- a component is a Sentence, Topic, or Section, then it is treated as a set of variations of a separate piece of a document, which will be resolved to select the appropriate version.
- a component is an External object, then it is treated as a whole complete customizable document. In this way, hypertext links to other customizable documents can be handled within the system.
- Block 8 The Words. Block 8 describes the instances ofthe class Word. Each Word description must specify its associated string and its associated annotations. Each annotation must be an instance ofthe class Annotation.
- Block 9 The Annotations.
- Block 9 describes the instances ofthe class Annotation that will be used to insert all the relevant linguistic and formatting information into the customized version ofthe document to be output. A description of block 9 is not included in figure 2. Instead, the general structure of an Annotation class object is shown in figure 3(a).
- the Annotation objects can be grouped into several distinct sub-taxonomies, one for each type of linguistic or formatting annotations that will be attached to the main master-document data structure.
- one Annotation sub-taxonomy might specify details of HTML layout for the overall document and each component ofthe document; another Annotation sub-taxonomy might record properties ofthe discourse structure ofthe overall document and each component ofthe document, such as rhetorical relations and coreference links.
- Each Annotation object has a property "parent" to reference its immediate ancestor in its (sub-)taxonomy.
- Block 10 The Externals.
- Block 10 describes the instances ofthe class External , which will be used to create hypertext links to other customizable documents specified in other datafiles. Each External description must specify the following attributes: The name ofthe file containing the external customizable document.
- a user profile The user profile is a list of parameters that describe the user or audience for whom the document customization is being performed. The general structure of an External class object is shown in figure 3(b).
- the data structure 13 allows an author to describe the structure of a customizable document (i.e., a master document).
- the data structure has a recursive and object-oriented form and can be implemented using an object-oriented programming language so that a customizable document described in the form ofthe data structure can be implemented as an object-oriented computer program.
- the elements ofthe data structure are related by both part relationships and by inheritance relationships, so that the relationships between the elements of a customizable document described in the form ofthe data structure and implemented as a corresponding object-oriented computer program can be recognized and maintained by the object hierarchy and inheritance mechanism of an object-oriented computer program.
- an object-oriented computer program that implements the data structure is both the form and content of a customizable document and the process for selecting and generating an appropriately customized version ofthe document.
- the data structure is generic in the sense that it can implement any customizable document given in the form ofthe data structure.
- the data structure 13 describes the corpus text of a specific customizable document (i.e., a master document) having elements and structure, as shown in figures 2(a) , 2(b) and 2(c) following.
- the form ofthe data structure is defined by the set of general class objects 16 describing the elements and structure of a customizable document in terms of object- oriented program structures. These class objects are related by both part relationships and by inheritance relationships to be explained below.
- the operation ofthe system 10 (the “tailoring engine") may be explained as follows:
- the parser program 14 reads in and parses the master-document datafile 12 to recognize its structure and then maps the contents ofthe input datafile into class objects of name according to class names specified in the input data structure. Properties ofthe classes are also recognized according to information contained in the data structure.
- the parser program 14 also acts as a document-class instantiator program which uses the parsed contents ofthe input datafile to dynamically create instances of the program structures identified by the general class objects described above. Properties ofthe classes are also dynamically assigned according to information contained in the data structure.
- An integration of data structures with the main process ofthe system comprises the following:
- a feature ofthe system is that given the current values ofthe purpose parameters, the instances ofthe program data structures, that is to say, the instances of the class objects describing the elements ofthe master document, execute themselves to select and generate the appropriate customized version ofthe document.
- the core ofthe system that is, the integration of program data structures with the selection process, is generic in terms ofthe following properties:
- the system core is independent ofthe application: the only items that need be re-defined for a new application are the input datafile and the interface for reading the current values ofthe purpose parameters. This is discussed further below.
- the system core is currently implemented in the Java programming language, but is independent ofthe underlying programming language, i.e., processor, to the extent that the programming language used must provide an object-oriented paradigm and a semantics for property inheritance that is consistent with the specification ofthe resolution process used in the system for generating a customized version of a document from the instances ofthe program data structures.
- the process of generating a customized version of a document from the instances ofthe program structures is referred to as resolution and will be discussed later with reference to Figure 5.
- a customized version of a document can be generated in any number of different levels of representation of its content (e.g., surface English; a syntactic or semantic representation to be used by a text-repair facility; and so on).
- the different representations to be generated for any given application must be indicated by the representation-level parameters in the description ofthe document given by the data structure in the input datafile. This information is specified in block 1 ofthe data structure as described with reference to figures 2(a), 2(b), and 2(c) above.
- Each different representation ofthe content of a customized version ofthe document will be generated along with a list of all the relevant annotations to the content. These annotations provide information on the multiple forms in which the document may subsequently be presented to a reader, and the linguistic information that may be used to guide subsequent repair ofthe customized document.
- the customized document that is generated will also include all annotations to the content concerning the External objects that can be used to provide hypertext links to other customizable documents or to other applications ofthe system 10 (the "tailoring engine") from figure 1.
- the process of generating a customized version of a master document is shown in Figure 4. There are two main stages to this process, the initial setup and the main program loop. In the initial setup, the input datafile is read in and parsed, and the appropriate instances ofthe various document classes described above are created.
- the parser program 14 reads in the input datafile 12, which contains the data structure 13 giving the specification of a customizable document.
- the driver program also acts as a class instantiator to create instances of all relevant document classes according to the input data structure.
- Program links, i.e., references, are created between these class instances via the setting of their properties and assignment of their property values.
- new values ofthe purpose parameters are read in, and a customized version ofthe document is generated as output for each specified level of representation.
- a user interface in the form of a reader program 20 obtains the new values ofthe purpose parameters. In the latter instance, the parameter values may be entered interactively or may be read in from previously compiled profiles of user preferences stored in computer databases.
- a selection engine 18 resolves the document instances created in the setup stage according to the current values ofthe purpose parameters to generate the appropriately customized version ofthe document.
- the customized document is output in all specified levels of representation with all relevant linguistic and formatting information attached to each component of the document If there are no more new purpose-parameter values to read in, the main program loop terminates.
- FIG. 5 is a graph showing the resolution process for a customized document generated according to an embodiment ofthe present invention.
- the pseudocode for the Resolve procedures, which implement the resolution process, is as follows:
- An instance of DocumentObjectSet i.e., a set of references to resolved DocumentObjects, with one set member for each desired level of representation of this Variation instance
- This property is a list of all the Annotation objects that apply to this document-class instance and must contain one Annotation object for each ofthe distinct sub- taxonomies in the overall class of Annotation objects.
- a user interface in the form of a reader program to obtain new values of purpose parameters is provided.
- This reader program returns as output an instance of WorkingCondition which is a set of purpose parameters and their values (implemented as a hashtable in one embodiment ofthe invention).
- the main program first reads in the datafile, then calls the reader program to obtain a new instance of WorkingCondition. The main program will then start the resolution process for the toplevel Document object by passing it the WorkingCondition.
- the output of each iteration ofthe main program will be a customized version ofthe document in all the levels of representation specified in the input datafile using the representation-level parameters. Each representation ofthe customized document is output for possible later processing by the application system.
- the first example is for the customizable home page ofthe HealthDoc Project at the University of Waterloo (Waterloo, Canada).
- the second example is for a master document giving basic health information on diabetes.
- top-level object toplevel Document.webbedoc // The Documents and DocumentVariations Document webbedoc
- Sentence sent-compliance-2
- SentenceRepLevel sent-compliance-3 a-english
- Sentence sent-compliance-4
- Lexical lexSynonymsl I variations lexSynonyms 1 a lexSynonyms 1 b& annotations ⁇
- condition-()& componentList-Topic.topic 1 & annotations html-sec-diabetes-subsec 1 - 1 a discourse-sec-diabetes-sec 11
- repLevel english& componentList ⁇ Lexical.lexDiab Lexical.lexis Lexical. lexa Lexical.lexgroup Lexical. lexof Lexical. lexconds Lexical.lexin Lexical.lexwhich Lexical. lexglucose Lexical. lexlevels Lexical.lexare Lexical.
- Sentence sent3a-l
- condition ()
- condition ()
- the program objects are:
- BasicObject classes e.g., BasicDocument, BasicSection, etc.
- VariationContainer classes e.g., Document, Section, Topic
- VariationContainer classes are extensions of ResolvableObject and therefore have a Resolve method.
- An instance of List (a list ofthe desired representation levels to output)
- An instance of DocumentObjectSet i.e., a set of references to resolved DocumentObjects, with one set member for each desired level of representation of this Variation instance
- Toplevel Procedures These are the toplevel procedures used to read in a datafile containing a master document, create instances ofthe document-class objects, then loop to read in new values ofthe purpose parameters and generate an appropriately customized version ofthe document, with all appropriate annotations, at each level of representation as specified by the representation-level parameters given in the datafile.
- the line is a comment or a blank line then skip over it else if the line specifies the toplevel DocumentObj ect then set the "toplevel” variable to reference this object else if the line specifies the purpose parameters and their values then set the "purposeParameters” variable else if the line specifies the possible representation levels then set the "repLevels” variable else instantiate the specified document object
- Algorithm call an application-specific interface to:
Abstract
Description
Claims
Applications Claiming Priority (7)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB9716986 | 1997-08-11 | ||
GBGB9716986.6A GB9716986D0 (en) | 1997-08-11 | 1997-08-11 | A method and apparatus for authoring of castomizable multi,edia documents |
GBGB9720133.9A GB9720133D0 (en) | 1997-09-22 | 1997-09-22 | A method and apparatus for authoring of customizable multimedia documents |
GB9720133 | 1997-09-22 | ||
CA002230367A CA2230367C (en) | 1997-08-11 | 1998-02-24 | A method and apparatus for authoring of customizable multimedia documents |
CA2230367 | 1998-02-24 | ||
PCT/CA1998/000771 WO1999008205A1 (en) | 1997-08-11 | 1998-08-11 | A method and apparatus for authoring of customizable multimedia documents |
Publications (1)
Publication Number | Publication Date |
---|---|
EP1002284A1 true EP1002284A1 (en) | 2000-05-24 |
Family
ID=27170640
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP98939453A Withdrawn EP1002284A1 (en) | 1997-08-11 | 1998-08-11 | A method and apparatus for authoring of customizable multimedia documents |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP1002284A1 (en) |
AU (1) | AU8795998A (en) |
WO (1) | WO1999008205A1 (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6981214B1 (en) * | 1999-06-07 | 2005-12-27 | Hewlett-Packard Development Company, L.P. | Virtual editor and related methods for dynamically generating personalized publications |
US7302430B1 (en) | 1999-08-31 | 2007-11-27 | International Business Machines Corporation | Workflow management system for generating output material based on customer input |
US7028303B2 (en) | 1999-09-17 | 2006-04-11 | International Business Machines Corporation | Method, system, and program for processing a job in an event driven workflow environment |
CA2285877A1 (en) | 1999-10-15 | 2001-04-15 | Ibm Canada Limited-Ibm Canada Limitee | System and method for capturing document style by example |
US7171373B2 (en) | 1999-10-21 | 2007-01-30 | International Business Machines Corporation | Database driven workflow management system for generating output material based on customer input |
AU2047401A (en) * | 1999-11-24 | 2001-06-04 | Sun Microsystems, Inc. | System, method and computer program product for publishing web page content having a uniform predetermined format and features |
US7246315B1 (en) | 2000-05-10 | 2007-07-17 | Realtime Drama, Inc. | Interactive personal narrative agent system and method |
EP1170672A1 (en) * | 2000-07-04 | 2002-01-09 | OKS GmbH | Automatic generation of publishing documents on the internet |
AU2001289595A1 (en) * | 2000-09-22 | 2002-04-02 | Vizion Factory E-Learning A/S | Electronic document system |
-
1998
- 1998-08-11 WO PCT/CA1998/000771 patent/WO1999008205A1/en not_active Application Discontinuation
- 1998-08-11 EP EP98939453A patent/EP1002284A1/en not_active Withdrawn
- 1998-08-11 AU AU87959/98A patent/AU8795998A/en not_active Abandoned
Non-Patent Citations (1)
Title |
---|
See references of WO9908205A1 * |
Also Published As
Publication number | Publication date |
---|---|
AU8795998A (en) | 1999-03-01 |
WO1999008205A1 (en) | 1999-02-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6938203B1 (en) | Method and apparatus for authoring of customizable multimedia documents | |
Wilcock | Introduction to linguistic annotation and text analytics | |
Androutsopoulos et al. | Generating natural language descriptions from OWL ontologies: the NaturalOWL system | |
Bouayad-Agha et al. | Natural language generation in the context of the semantic web | |
Ko et al. | Barista: An implementation framework for enabling new tools, interaction techniques and views in code editors | |
Forbes-Riley et al. | Computing discourse semantics: The predicate-argument semantics of discourse connectives in D-LTAG | |
Jackendoff et al. | Relational morphology: a cousin of construction grammar | |
de Souza et al. | Semiotic engineering principles for evaluating end-user programming environments | |
DiMarco et al. | HealthDoc: Customizing patient information and health education by medical condition and personal characteristics | |
Hornstein | The minimalist program after 25 years | |
Witt et al. | On the lossless transformation of single-file, multi-layer annotations into multi-rooted trees | |
EP1002284A1 (en) | A method and apparatus for authoring of customizable multimedia documents | |
Androutsopoulos et al. | Source authoring for multilingual generation of personalised object descriptions | |
Byamugisha | Ontology verbalization in agglutinating Bantu languages: a study of Runyankore and its generalizability | |
CA2230367C (en) | A method and apparatus for authoring of customizable multimedia documents | |
Kapetanios et al. | Simplifying syntactic and semantic parsing of NL-based queries in advanced application domains | |
Ranta | Gf: A multilingual grammar formalism | |
DiMarco et al. | Generation by selection and repair as a method for adapting text for the individual reader | |
Steel et al. | Generating human-usable textual notations for information models | |
Dimitriadis et al. | How to integrate databases without starting a typology war: The Typological Database System | |
Branting et al. | Integrating discourse and domain knowledge for document drafting | |
Bouayad-Agha et al. | Natural language generation and semantic web technologies | |
EP0358702A4 (en) | Tailored text generation. | |
Theune et al. | Generating varied narrative probability exercises | |
Becker | Natural language generation with fully specified templates |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20000225 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE |
|
17Q | First examination report despatched |
Effective date: 20000803 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20020924 |
|
PLAO | Information deleted related to despatch of communication that opposition is rejected |
Free format text: ORIGINAL CODE: EPIDOSDREJ1 |
|
PLCK | Communication despatched that opposition was rejected |
Free format text: ORIGINAL CODE: EPIDOSNREJ1 |