600 results about "Entity type" patented technology

A system and method lets a user create or import ontologies and create databases and related application software. These databases can be specially tuned to suit a particular need, and each comes with the same error-detection rules to keep the data clean. Such databases may be searched based on meaning, rather than on words-that-begin-with-something. And multiple databases, if generated from the same basic ontology can communicate with each other without any additional effort. Ontology management and generation tools enable enterprises to create databases that use ontologies to improve data integration, maintainability, quality, and flexibility. Only the relevant aspects of the ontology are targeted, extracting out a sub-model that has the power of the full ontology restricted to objects of interest for the application domain. To increase performance and add desired database characteristics, this sub-model is translated into a database system. Java-based object-oriented and relational application program interfaces (APIs) are then generated from this translation, providing application developers with an API that exactly reflects the entity types and relations (classes and methods) that are represented by the database. This generation approach essentially turns the ontology into a set of integrated and efficient databases.

A spelling system derives a language model for a particular domain of structured data, the language model enabling determinations of alternative spellings of queries or other strings of text from that domain. More specifically, the spelling system calculates (a) probabilities that the various query entity types—such as STREET, CITY, or STATE for queries in the geographical domain—are arranged in each of the various possible orders, and (b) probabilities that an arbitrary query references given particular ones of the entities, such as the street “El Camino Real.” Based on the calculated probabilities, the spelling system generates a language model that has associated scores (e.g., probabilities) for each of a set of probable entity name orderings, where the total number of entity name orderings is substantially less than the number of all possible orderings. The language model can be applied to determine probabilities of arbitrary queries, and thus to suggest alternative queries more likely to represent what a user intended.

A method and apparatus for segmenting text is provided that identifies a sequence of entity types from a sequence of characters and thereby identifies a segmentation for the sequence of characters. Under the invention, the sequence of entity types is identified using probabilistic models that describe the likelihood of a sequence of entities and the likelihood of sequences of characters given particular entities. Under one aspect of the invention, organization name entities are identified from a first sequence of identified entities to form a final sequence of identified entities.

Embodiments discussed in this disclosure can be configured to provide for including a reference link in an instant message by a sender for a recipient includes receiving an identifiable character string in the instant message, the identifiable character string being associated with an entity, determining an entity type that is associated with the entity, and retrieving information related to the entity. Other embodiments of the method include creating a reference link that associates the identifiable character string with the information related to the entity and including the reference link in the instant message. Other methods and devices are also provided.

Methods and apparatus to transform attribute data about assets in a source system data model into attribute data about the same assets in a target system data model. The first step is to extract the necessary attribute data from attribute data collected about inventory assets of a business entity needed to populate the attributes in objects representing those inventory assets in a target system data model. Transformation rules are written which are designed to make all conversions necessary in semantics, units of measure, etc. to transform the source system attribute data into attribute data for the target system which has the proper data format. These transformation rules are executed on a computer on the extracted attribute data and the transformed attribute data is stored in an ER model. In the preferred embodiment, the transformation rules are object-oriented in that transformation rules for subtypes can be inherited from their parent types or classes. An export adapter which is capable of invoking the application programmatic interface of the target system CMDB is then used to export the transformed attribute data stored in the ER model to the target system CMDB. A heuristic method to create self-consistent data blocks without exceeding a maximum size limit involves loading instances of entity types and all related instances in the order of decreasing connectivity metric.

A system, method and computer program product for conducting questions and answers with deferred type evaluation based on any corpus of data. The method includes processing a query including waiting until a “Type” (i.e. a descriptor) is determined AND a candidate answer is provided. Then, a search is conducted to look (search) for evidence that the candidate answer has the required Lexical Answer Type (e.g., as determined by a matching function that can leverage a parser, a semantic interpreter and/or a simple pattern matcher). Prior to or during candidate answer evaluation, a process is provided for extracting and storing collections of entity-type pairs from semi-structured text documents. During QA processing and candidate answer scoring, a process is implemented to match the query LAT against the lexical type of each provided candidate answer and generate a score judging a degree of match.

A hierarchical packaging model of self-describing plugin modules and packages of plugin modules. Identifiers are assigned to each package of plugin modules in a hierarchical relationship so that adjacently identified packages are backward compatible. The package identifiers are maintained internally to the package. Similarly, identifying data of a plugin module is maintained internally within the namespace of the respective module. Interdependency of plugin modules is determined by comparison of data maintained in the namespace of each module.

Automatic relationship extraction is provided. A machine learning approach using statistical entity-type prediction and relationship predication models built from large unlabeled datasets is interactively combined with minimal human intervention and a light pattern-based approach to extract relationships from unstructured, semi-structured, and structured documents. Training data is collected from a collection of unlabeled documents by matching ground truths for a known entity from existing fact databases with text in the documents describing the known entity and corresponding models are built for one or more relationship types. For a modeled relationship-type, text chunks of interest are found in a document. A machine learning classifier predicts the probability that one of the text chunks is the entity being sought. The combined machine learning and light pattern-based approach provides both improved recall and high precision through filtering and allows constraining and normalization of the extracted relationships.

A preferred embodiment of the invention pertains to a method of comparing first and second databases (DB1, DB2) that are each comprised of a plurality of entities (A to E) having one or more characteristics, said entities being grouped into a plurality of data classes in each said database each representative of a particular entity type; the method comprising: (i) for each said data class of said first and second databases, compiling a list representative of the entities occurring within that class and the attributes for each said entity; (ii) identifying and comparing corresponding data classes for each of said first and second databases; and (iii) identifying on the basis of said comparison differences between corresponding entities of said corresponding data classes. A computer program, and storage medium are also disclosed.

The disclosed technology can identify indicia associated with different entity types that interact within an industry, identify one or more relationships (e.g., contractual provisions) that can affect interactions between such entity types, and identify transactions associated with one or more of the interactions. Further, the identified transactions can be organized into one or more transaction sequences. The identified indicia, the one or more identified relationships, and the one or more transaction sequences can then be associated to form a semantic network. An instance of the semantic network can be formed in response to the execution of at least some of the transaction sequence and can serve, at least in part, as the basis for processing requests associated with the entities. The requests can correspond to interactions associated with the entities and may be represented in a natural language format, exhibiting a fixed context and a fixed grammar.

A terminal for interacting with a physical entity includes a controller capable of selecting a physical entity, and thereafter receiving data from the entity at a terminal at least partially over an air interface. The controller can also identify a type of the entity based upon the data. Thereafter, the controller can be capable of performing at least one action based upon the entity type, an application operating on the terminal and a selection event. The application can comprise any of a number of applications, such as an application associated with the type of entity or an alternative application other than the associated application. Likewise, the selection event can comprise any of a number of different predefined events, such as “select and get information,”“select and open,” and/or “select drag and drop.”

A system has a processor coupled to access a document database that indexes keywords and instances of entities having entity types in a plurality of documents. The processor is programmed to receive an input query including one or more keywords and one or more entity types, and search the database for documents having the keywords and entities with the entity types of the input query. The processor is programmed for aggregating a respective score for each of a plurality of entity tuples across the plurality of documents. The aggregated scores are normalized. Each respective normalized score provides a ranking of a respective entity tuple, relative to other entity tuples, as an answer to the input query. The processor has an interface to a storage or display device or network for outputting a list including a subset of the entity tuples having the highest normalized scores among the plurality of entity tuples.

The invention discloses a multi-strategy fused method and device for identifying a named entity. A first identification result is obtained by utilizing a first identifying model to identify the named entity in obtained linguistic data, the first identifying model can update and extend a corpus in the method, accordingly the newly produced named entity in the linguistic data can be identified, thus a first identifying result has higher accuracy rate, the named entity in the linguistic data is identified by utilizing a multi-identifying-model fused method to obtain a second identifying result, the first identifying result and the second identifying result are fused to obtain a third identifying result, then a semantic mining system is utilized to conduct role assignment on the named entity, the named entity having a role is output, accordingly it is achieved that the named entity is reliably identified under the situations that data is massive, the entity type is diversified and new words emerge in endlessly, and role assignment is conducted on the identified named entity.

The invention relates to an event trigger word extraction method based on auto-encoder fusion document information, which comprises the following steps: generating a training set by using unlabeled free text corpora, and training a GRU model to construct an auto-encoder; pre-processing and labeling the training corpus, and extracting to-be-identified words; obtaining a document vector of the to-be-identified word in a document where the to-be-identified word is located by the auto-encoder, and taking the document vector as a global feature of the to-be-identified word; expressing word vectorsand entity types of the to-be-identified words in a distributed manner to serve as local features of the to-be-identified words; carrying out vector splicing on the global feature and the local feature to obtain context features of the to-be-identified word; inputting the context features to the Bi-GRU model to perform multi-classification to identify whether the to-be-identified word is an eventtrigger word and the corresponding event type of the to-be-identified word.

A method and system for socket API call emulation facilitates communication between an application and non-networked I/O devices. The present invention intercepts and modifies IP-aware socket API calls en route from the application to non-networked I/O devices so that such calls are received by non-networked I/O devices as IP-unaware I/O access calls. The invention allows the application to use socket API calls to communicate with all entity types and thereby relieves an application developer of the burden to write and debug/edit separate code and documentation for non-networked I/O device types with which an application is expected to communicate. Socket API call emulation is in some embodiments performed by an emulation module within a socket API operatively coupled between an application and a device driver associated with a non-networked I/O device.

The invention provides a Chinese and English-named entity identification method and a system based on a conditional random field (CRF). The method comprises the following steps: (101) converting inquiry voice of a user into a text; (102) separating text information into Chinese characters and English letters on the basis of a finite state machine; (103) extracting the characteristics of a text of separated vocabularies; (104) performing entity identification on the text by adopting a training CRF model according to a characteristic extraction result, and marking an entity type, wherein the CRF model is a conditional random field model of a linear chain structure. The step (102) further comprises the following steps: (102-1) performing character separation on Chinese and English; (102-2) identifying English word strings by using the finite state machine, namely, combining adjacent English letters, blank spaces and symbols in English; (102-3) performing word segmentation on the English word strings.

Methods and systems allow access to information in an enterprise environment that stores information in data silos. Entity type metadata, relations between entity types and access control information is extracted from the data silos and represented in a data virtualization system. Metadata information representing security information extracted from multiple data silos is combined to construct global security information for the enterprise. Security roles are combined to generate global security roles and access control lists are combined to generate globalized access control lists. The global security information can be modified by system administrators. Security information is refreshed from the data silos for each session created by the user and is applied to all data access requests created using the session.

A system and method for system and method data binding to type(s) of object(s) is provided. The system can be employed in a software application development environment to facilitate data binding to types of entity(ies) (e.g., object(s), web service(s) and/or a hierarchy of objects) rather than instance(s) of the entity(ies). With the ability to bind to types, the system provides the ability to abstract schema by which the control(s) are bound from the source of data. The system allows developer(s) and/or architect(s) to specify their own base classes for their own object(s). The system includes a binding source component that supports the ability to set its data source to a type, rather then a specific instance. For example, when the data source is changed, the binding source component can evaluate whether the data source (e.g., object) can be instanced and whether the data source (e.g., object) supports necessary designer interface(s). If the type isn't supported by the designer, using reflection, the binding source component can interrogate the public properties of the type. Additionally, the binding source component can analyze complex object(s).

The invention is applied to the technical field of data extraction, and provides a multi-type entity recognition multi-task deep learning model training method and device. The method comprises data pre-processing: a skip-gram neural network model is established to convert pre-processed text corpus data into vectors; a sample data set is constructed according to to-be-recognized-and-extracted entity types; word segment features of sample data are constructed; multi-type entity recognition multi-task deep learning model is established. According to the method, the extraction of common features of entities of related types is achieved in a parameter sharing manner, marking of the entities is completed by using an independent model, so that the model has better generalization capability on therecognition and extraction of multiple types of entities of text data, namely the overall correctness of entity recognition is improved, in addition, only one model is trained, the common features only need to be trained for once in one iterative procedure, and training time can be greatly reduced.

The invention discloses a medical entity relationship extraction method based on feature fusion, and the method comprises the steps: enabling entities in a knowledge base to be aligned to medical corpora through a remote supervision and rule combination method, and constructing an entity pair sentence set; performing word-level vector coding on the sentences based on a convolutional neural networkmodel to obtain overall feature vector representation of the sentences; extracting features in left and right subtree directions on the shortest dependency path of the sentences by using a recurrentneural network respectively, and performing splicing operation; and fusing the sentence overall features and the dependency syntax features which are extracted from the two parts respectively, and performing final relation extraction on the obtained fusion features. According to the method, on the premise that a dependency syntax structure is utilized; entity type characteristics capable of expressing entity relationship types among entities are introduced; the position features and the overall features of the sentences are integrated with the dependency syntactic features, the semantic relationship between the sentences is better learned, the interference of noise data on medical entity relationship extraction is reduced, and the accuracy of medical entity relationship extraction can be improved to a certain extent.