Xerox Corporation

Palo Alto Research Center Inc.

Kirk John Ocke

Michael David Shepherd

Dale Ellen Gaucas

Rong Zhou

Minh Binh Do

Barry Glynn Gombert

Philip Crane Rose

Reuven J Sherwin

Frank Michael Goetz

Dale Ellen Gaucus

Michael D Shephard

The present disclosure provides for categorization of lists of words. The method comprises querying DBpedia to find the resources related to the given list of words. Once the resources are found, the corresponding media Wikipedia categories can be retrieved, as well as their ancestors, generating a graph of categories. A number of graph analysis algorithms can then be applied to the graph, each returning a selected category. For each algorithm a classifier is trained to decide whether the output of the algorithm is indeed the 'best' category. An ensemble weighted majority voting can then be used to select the best category based on votes cast by each classifier. The disclosure demonstrates a more accurate selection of the best category and can include an ensemble majority rated voting algorithm comprising all voting members initially casting one vote; i.e., highest frequency, most frequently occurring word, least common ancestor and centrality measures.

System for categorizing lists of words of arbitrary origin

A computer-implemented method and system for creating data sources for a variable data publishing (VDP) marketing campaign pattern. The method and system can be utilized to generate a data source from external sources, such as Encyclopedic knowledge-bases, when desired data is not locally available. A user can specify a conceptual identity, and the method and system automatically map the conceptual identity to an appropriate database, generate semantic queries, and query the database to develop the data source for use in the campaign.

Knowledge-assisted approach to dynamically create data sources for variable-data marketing campaigns

A system and method for semantically classifying numerical data includes using semantic classification techniques on 'nearby' non-numerical data to identify a context whereby opaque data sets of numbers can be semantically classified inside of that context. An Electronic Knowledge Base is used to query against the context and determine the semantics of the opaque numeric data sets.

Method for semantic classification of numeric data sets

An improved automated planning method and system includes the use of an automated planner for validating a workflow plan with respect to device capabilities. As such, the present invention provides for an improved method and system for implementing and managing workflow plans utilized in printing environments such as, for example, print shops. A plurality of device capability descriptions can be converted into a data indicative of facts and operators for utilizing in a knowledge-based reasoning system. Next, a workflow plan can be converted into properties indicative of a goal state. The goal state and the data indicative of facts and operators can then be utilized in association with a domain-independent or domain-dependent planner to determine if said workflow plan is valid and executable in order to thereby validate said workflow plan.

System and method for automatically validating a workflow plan using an automated planner

Methods and systems herein receive a selection of at least one variable data printing (VDP) pattern from a first user to compose an initial sample design. The methods receive image, text, and graphics into the initial sample design from the first user and the user tags such image, text and graphics with VDP pattern elements. The processes of receiving the image, text, and graphics and tagging alter the initial sample design into the tagged sample design. Further, the methods and systems can automatically convert the tagged sample design into at least one different format file. The methods and systems can automatically provide the different format file to at least one second user. Additionally, the second user can include data components into the different format file to support the composition of a targeted marketing campaign sample.

System and method for creating variable data print samples for campaigns

An improved automated planning method and system includes the use of an automated planner for generating a plan template and a final plan generator to produce a fully parameterized workflow plan. Such an approach provides for the generation of fully parameterized workflow plans utilized in rendering environments. A plurality of device capability descriptions can be converted into a data indicative of facts and operators for use with an automated planner, which can be either domain-independent or domain-dependent. Next, a formal product description can be converted into properties indicative of a goal state. The goal state and the data indicative of facts and operators can then be utilized with the automated planner to synthesize a plan template. The plan template is then provided to a final plan generator, along with a formal product description, to produce a fully parameterized workflow plan.

System and method for automated generation of a fully parameterized workflow plan

A method and user interface for eliciting and processing user feedback includes displaying on a computing device display a digital asset, wherein the digital asset has an associated list having at least one keyword; displaying on the computing device display a user feedback interface, the user feedback interface comprising at least one input component configured to receive user generated input; receiving a keyword replacement pair via the at least one input component; storing the keyword replacement pair; analyzing the stored keyword replacement pair; and producing results indicting potential changes to the digital asset based upon the analyzing. Based upon the results of the analysis, an updated version of the digital asset may be produced including any user suggested changes.

Elicitation method for custom image preferences using keywords

Systems and methods for automatically assigning items to printable media are disclosed. A computer readable medium is in communication with a computing device and includes programming instructions for electronically obtaining a first template that includes a plurality of first content items, each including a first content area. For each first content item, first content area properties and first content item properties are identified. A second template including a plurality of second content areas is electronically obtained. For each second content area, second content area properties, second content item properties and content rules are automatically identified. For each first content item, it may be determined whether a first content item corresponds to a second content area based on the rules. The first content item may be automatically assigned to the second content area. A representation of each assigned content item may be printed.

System and method for a design with intent metadata

A method and system provides one or more product type descriptions, and constructs a map between the sub-conditions of the product type descriptions and constructs in a Web Ontological Language (OWL). The method converts the product type description rules into new OWL classes by creating a new OWL class for each corresponding rule; adding the new OWL class to the OWL ontology; for each condition of the corresponding rule, extracting sub-conditions; for each the sub-condition extracted; creating new constructs for the new OWL class using the map; and adding new class restrictions to the new OWL class, wherein each of the new class restrictions relates to a corresponding condition of the corresponding rule.

Converting rules in a rule-based system to ontological information

A method and system for automatically identifying an existing workflow to manufacture a given product type. A workflow description of a workflow can be transformed into a product type description utilizing particular facts associated with the workflow. An inferred product type description can then be at least one product type among a group of product type descriptions representing known product types. A mapping can then be derived of the workflow with respect to product types, in response to comparing the inferred product type description to the product type description(s) among product type descriptions representing known product types. The workflow description can be converted into the product type description utilizing a rule-based expert system. The particular facts can be one or more of the following: processing steps, sequences of processing steps, types of operations to be performed, and outputs produced.

Method and system for automatically identifying an existing workflow to manufacture a given product type

A method of semantically classifying a data set of open class nouns and a system for executing the method. The method includes loading, by a processing device, a data set comprising one or more open class nouns from a computer readable medium operably connected to the processing device; extracting, by the processing device, the one or more open class nouns from the data set; for each open class noun, querying, by the processing device, one or more application programming interfaces (APIs) to produce one or more results; deriving, by the processing device, a confidence score for the data set based upon the one or more results; and determining, by the processing device, a classification for the data set based upon the derived confidence score.

Open class noun classification

A rule based system can produce a formal product description when only partially specified processing instructions and page description language (PDL) files are available. The PDL files contain product content data and can contain product description data. The partially specified processing instructions can be obtained from PDL file metadata or from other sources such as a Job Definition Format (JDF) Graybox. A knowledge based reasoning module can use the product description data and partially specified processing instructions to produce a formal product description from which a process network can be automatically produced.

Transforming partially specified processing instructions and PDL data into a product description

A method and system for automatically classifying a product description into a product type based on a classification system is disclosed. Real product descriptions representing a given product type can be compared with a pre-defined set of characteristics that are relevant to the classification system. The relevant information regarding the classification system is extracted from each of the real product descriptions. The extracted information can then be coalesced and transformed into a new set of classification rules for a new product type. The new classification rules can be used to automatically classify the product descriptions that represent the new product type. The classification system can be a rule based classification system and/or semantic web classification system.

Method and system for automatically adding new class definitions to a classification system

A method provides a knowledgebase having a set of existing print product classes into which print product descriptions are classified. The method receives the new print product class from a user and maps the representation of new print product class conditions to be semantically consistent with the representation of existing class conditions. The method evaluates the new print product class to produce user feedback. The user feedback is output to the user. The new print product class is added to the knowledgebase only if the new print product class is valid and free of errors and semantically consistent with the definitions of already existing print product classes.

Semantically validating print product definitions

A method and system for adding processes to print production workflows utilizing asset metadata and automated reasoning system is disclosed. The asset metadata can be extracted from the resources associated with a print product description such as PDF files and image and artwork files. The asset metadata can be processed through an automated reasoning system in order to infer additional metadata. The inferred metadata can be utilized to add and parameterize a process node in order to build a more effective and useful process network. The reasoning system can be a rule based reasoning system and/or ontology based reasoning system.

Growing community of inventors

Ontario, NY, United States of America

Kirk John Ocke