Abstract: A non-transitory computer readable storage medium has instructions executed by a processor to invoke an image processing module to ingest a digital invoice. An evaluation module derives metrics from the digital invoice. A semantic document processing module forms entity extracts from the digital invoice, where each entity extract from the digital invoice has a potential mapping to a trained machine learning model element. An entity extraction correction module overrides the potential mapping to the trained machine learning model element when user feedback from a similar entity extract from a previously processed digital invoice exists to produce a processed digital invoice with a user feedback element inconsistent with the potential mapping to the trained machine learning model element. The processed digital invoice is delivered to an accounting module for final disposition of the digital invoice.

Abstract: In one aspect a computerized method for detecting anomalies in expense reports of an enterprise includes the step of implementing a semantic analysis algorithm on an expense report data submitted by an employee, wherein the expense report data is provided in a computer-readable format. The method includes the step of, with one or more machine learning algorithms, detecting an anomaly in expense report data. The method includes the step of obtaining an augmentation of the expense report data with a set of web scale data. The method includes the step of verifying receipts associated with an expense report. The method includes the step of determining that the employee or any employee has previously claimed an expense in the expense report data. The method includes the step of identifying an inappropriate expense in the expense report data.

Abstract: In one aspect a computerized method for detecting anomalies in expense reports of an enterprise includes the step of implementing a semantic analysis algorithm on an expense report data submitted by an employee, wherein the expense report data is provided in a computer-readable format. The method includes the step of, with one or more machine learning algorithms, detecting an anomaly in expense report data. The method includes the step of obtaining an augmentation of the expense report data with a set of web scale data. The method includes the step of verifying receipts associated with an expense report. The method includes the step of determining that the employee or any employee has previously claimed an expense in the expense report data. The method includes the step of identifying an inappropriate expense in the expense report data.

Abstract: In one aspect a computerized method for detecting anomalies in expense reports of an enterprise includes the step of implementing a semantic analysis algorithm on an expense report data submitted by an employee, wherein the expense report data is provided in a computer-readable format. The method includes the step of, with one or more machine learning algorithms, detecting an anomaly in expense report data. The method includes the step of obtaining an augmentation of the expense report data with a set of web scale data. The method includes the step of verifying receipts associated with an expense report. The method includes the step of determining that the employee or any employee has previously claimed an expense in the expense report data. The method includes the step of identifying an inappropriate expense in the expense report data.

Abstract: A computer implemented method includes receiving a document with line item textual entries and an attachment containing images of different objects characterizing different transactions. The images of the different objects are split into individual image objects. Attributes from the individual image objects are extracted. The line item textual entries are matched with the individual image objects to form matched image objects. The matched image objects include ambiguous matches with multiple individual image objects assigned to a single line item textual entry or a single individual image object assigned to multiple line item textual entries. An assignment model is applied to resolve the ambiguous matches. The assignment model defines priority constraints, assigns pairs of line item textual entries and individual image objects that meet highest priority constraints, removes highest priority constraints when ambiguous matches remain, and repeats these operations until no ambiguous matches remain.

Abstract: A computer implemented method includes receiving a document with line item textual entries and an attachment containing images of different objects characterizing different transactions. The images of the different objects are split into individual image objects. Attributes from the individual image objects are extracted. The line item textual entries are matched with the individual image objects to form matched image objects. The matched image objects include ambiguous matches with multiple individual image objects assigned to a single line item textual entry or a single individual image object assigned to multiple line item textual entries. An assignment model is applied to resolve the ambiguous matches. The assignment model defines priority constraints, assigns pairs of line item textual entries and individual image objects that meet highest priority constraints, removes highest priority constraints when ambiguous matches remain, and repeats these operations until no ambiguous matches remain.

Abstract: A data lineage tracking system may include a memory storing a module comprising machine readable instructions to obtain trace log entries representing an interaction with, a manipulation of, and/or a creation of a data value. The data lineage tracking system may further include machine readable instructions to select the trace log entries that are associated with commands performed by an application, cluster similar trace log entries from the selected trace log entries, and analyze mappings between the clustered trace log entries to determine data lineage flow associated with the data value.

Abstract: A corpus of textual listings is received and main concept words and attribute words therein are identified via an iterative process of parsing listings and expanding a semantic model. During the parsing phase, the corpus of textual listings is parsed to tag one or more head noun words and/or one or more identifier words in each listing based on previously identified main concept words or using a head noun identification rule. Once substantially each listing in the corpus has been parsed in this manner, the expansion phase assigns head noun words as main concept words and modifier words as attribute words, where possible. During the next iteration, the newly identified main concept words and/or attribute words are used to further parse the listings. These iterations are repeated until a termination condition is reached. Remaining words in the corpus are clustered based on the main concept words and attribute words.

Abstract: A data enrichment system may include an attribute relevance module to measure relevance of an attribute to a data object to be enriched. The data object may include the attribute including a known or an unknown value. An output value confidence module may calculate a confidence of an output value of a source used for enrichment of the data object. The output value may represent the known and/or unknown values of the attribute. The system may use the measured relevance of the attribute and the calculated confidence of the output value to determine assignment of the known or unknown values to the attribute.

Abstract: A document analysis, commenting, and reporting system provides tools that automate quality assurance analysis tailored to specific document types. As one example, the specific document type may be a requirements specification and the system may tag different parts of requirements, including actors, entities, modes, and a remainder. However, the flexibility of the system permits analysis of any other document type, such as instruction manuals and best practices guides. The system helps avoid confusion over the document when it is delivered because of non-standard terms, ambiguous language, conflicts between document sections, incomplete or inaccurate descriptions, size and complexity of the document, and other issues.

Abstract: In one aspect a computerized method for detecting anomalies anomalies in expense reports of an enterprise includes the step, of implementing a semantic analysis algorithm on an expense report data submitted by an employee, wherein the expense report data is provided in a computer-readable format. The method includes the step of, with one or more machine learning algorithms, detecting an anomaly expense report data. The method includes the step of obtaining an augmentation of the expense report data with a se of web scale data. The method includes the step of verifying receipts associated with an expense report. The method includes the step of determining that the employee or any employee has previously claimed an expense in the expense report data. The method includes the step of identifying an inappropriate expense in the expense report data.

Abstract: In one exemplary aspect, a contextual user engagement system for automatically engaging in a conversation with a user is provided. A context manager manage, with at least one processor, the context of the conversation with a user. The context manager receives a user utterance, wherein the context manager calls the other modules of the contextual user engagement system to obtain information with respect to the user utterance. The context manager maintains a context object for a conversation with a user. The context manager automatically updates the context object based on an information detected by the other modules of contextual user engagement system. A content knowledge database stores a content knowledge of the contextual user engagement system. The content knowledge includes a history of interactions with the user. An information extraction engine extract information from a specified source based on a set of rules.

Abstract: A computer-implemented method identifies textual statement relationships. Textual statement pairs including a first and second textual statement are identified, and parsed word group pairs are extracted from first and second textual statements. The parsed word groups are compared, and a parsed word score for each statement pair is calculated. Word vectors for the first and second textual statements are created and compared. A word vector score is calculated based on the comparison of the word vectors for the first and second textual statements. A match score is determined for the textual statement pair, with the match score being representative of at least one of the parsed word score and the word vector score.

Abstract: A method includes identifying a document structure instance that includes structure components. A selected structure component is chosen from the structure components. The method includes generating, with a processor, a hierarchical graph based on the selected structure component. The hierarchical graph includes classes, a class definition relationship, and an instance identifier that identifies the selected structure component. The hierarchical graph is output.

Abstract: A corpus of textual listings is received and main concept words and attribute words therein are identified via an iterative process of parsing listings and expanding a semantic model. During the parsing phase, the corpus of textual listings is parsed to tag one or more head noun words and/or one or more identifier words in each listing based on previously identified main concept words or using a head noun identification rule. Once substantially each listing in the corpus has been parsed in this manner, the expansion phase assigns head noun words as main concept words and modifier words as attribute words, where possible. During the next iteration, the newly identified main concept words and/or attribute words are used to further parse the listings. These iterations arc repeated until a termination condition is reached. Remaining words in the corpus arc clustered based on the main concept words and attribute words.

Abstract: A corpus of textual listings is received and main concept words and attribute words therein are identified via an iterative process of parsing listings and expanding a semantic model. During the parsing phase, the corpus of textual listings is parsed to tag one or more head noun words and/or one or more identifier words in each listing based on previously identified main concept words or using a head noun identification rule. Once substantially each listing in the corpus has been parsed in this manner, the expansion phase assigns head noun words as main concept words and modifier words as attribute words, where possible. During the next iteration, the newly identified main concept words and/or attribute words are used to further parse the listings. These iterations are repeated until a termination condition is reached. Remaining words in the corpus are clustered based on the main concept words and attribute words.

Abstract: A method includes identifying a first term in a document structure instance that matches a first permissible term in a glossary and identifying a second term in the document structure instance that matches a second permissible term in the glossary. A determination is made whether the document structure instance establishes an interaction between the first term and the second term. The method also includes generating, with a processor, a relationship map. The relationship map includes a first object that represents the first term, a second object that represents the second term, and an interaction object that represents whether the instance establishes the interaction between the first term and the second term.

Abstract: Embodiments of the claimed subject matter provide a process and system for designing composite application systems. One embodiment of the claimed subject matter is a process of designing a composite application system to meet an application requirement. The process includes defining a use case scenario performed by a composite application system to meet the application requirement, integrating the use case scenario with an execution tool, and specifying an underlying infrastructure for the system design, wherein, the entire process is performed within an integrated software application.

Abstract: A design assistant tool may automatically generate a functional design diagram representative of a system based on a processed requirements specification including a plurality of processed requirement statements and use cases. The design assistant tool may apply heuristic rules to the processed requirement statements and/or the processed use cases to determine various components of a functional design diagram. The design assistant tool may generate a technical design module based on the functional design diagram. The design assistant tool may allow various aspects of the functional design diagram to be modified to generate the technical design diagram.

Abstract: A data enrichment system may include an attribute relevance module to measure relevance of an attribute to a data object to be enriched. The data object may include the attribute including a known or an unknown value. An output value confidence module may calculate a confidence of an output value of a source used for enrichment of the data object. The output value may represent the known and/or unknown values of the attribute. The system may use the measured relevance of the attribute and the calculated confidence of the output value to determine assignment of the known or unknown values to the attribute.