33 results about "Process semantics" patented technology

A system for electronically distilling information from a business document uses a network scanner to electronically scan a platen area, having a business document thereon, to create a bitmap. A network server carries out a segmentation process to segment the scan generated bitmap into a bitmap object, the bitmap object corresponding to the scanned business document; a bitmap to text conversion process to convert the bitmap object into a block of text; a semantic recognition process to generate a structured representation of semantic entities corresponding to the scanned business document; and a document generation process to convert the structured representation into a structure text file. The semantic recognition process includes the processes of generating, for each line of text having a keyword therein, a terminal symbol corresponding to the keyword therein; generating, for each line of text not having a keyword therein and absent of numeric characters, an alphabetic terminal symbol; generating, for each line of text not having a keyword therein and having a numeric character therein, an alphanumeric terminal symbol; generating a string of terminal symbols from the generated terminal symbols; determining a probable parsing of the generated string of terminal symbols; labeling each text line, according to a determined function, with non-terminal symbols; and parsing the business document information text into fields of business document information text based upon the non-terminal symbol of each text line and the determined probable parsing of the generated string of terminal symbols.

Embodiments of the present invention are directed to acquiring information from the worldwide web, organizing information acquired from the worldwide web, and using the acquired and organized information to facilitate web-page searching, web-page browsing, and other worldwide-web-based activities. In one embodiment of the present invention, a database of concept objects is created from an initial set of semantic objects and from hyperlink information obtained from web pages by one or more web crawlers. The initial set of semantic objects is processed using hyperlink based objects created by the web crawler. The processed semantic objects are then associated with additional hyperlink-based objects to create a concept-object database. In certain embodiments of the present invention, the concept-object database can be further refined and supplemented in an automated fashion by additional web crawling, subsequent association of hyperlink-based objects with concept objects, and creation of new concept objects as well as by user input to, and editing of, the concept-object database. The concept-object database may be employed, in various embodiments of the present invention, to facilitate web browsing, web-page searching, and other worldwide-web-base activities.

The invention discloses an image description method and system based on a vision and semantic attention combined strategy. The steps include utilizing a convolutional neural network (CNN) to extract image features from an image whose image description is to be generated; utilizing a visual attention model of the image to process the image features, feeding the image features processed by the visual attention model to a first LSTM network to generate words, then utilizing a semantic attention model to process the generated words and predefined labels to obtain semantic information, then utilizing a second LSTM network to process semantics to obtain words generated by the semantic attention model, repeating the abovementioned steps, and finally performing series combination on all the obtained words to generate image description. The method provided by the invention not only utilizes a summary of the input image, but also enriches information in the aspects of vision and semantics, and enables a generated sentence to reflect content of the image more truly.

A method, apparatus and program storage device for delimiting words is provided and requires adding an extensible vocabulary markup to a plurality of markup languages, wherein the plurality of markup languages comprises a knowledge representation, and processing a provided term from one or more vocabularies for use in the knowledge representation.

Embodiments of the present invention are directed to acquiring information from the worldwide web, organizing information acquired from the worldwide web, and using the acquired and organized information to facilitate web-page searching, web-page browsing, and other worldwide-web-based activities. In one embodiment of the present invention, a database of concept objects is created from an initial set of semantic objects and from hyperlink information obtained from web pages by one or more web crawlers. The initial set of semantic objects is processed using hyperlink based objects created by the web crawler. The processed semantic objects are then associated with additional hyperlink-based objects to create a concept-object database. In certain embodiments of the present invention, the concept-object database can be further refined and supplemented in an automated fashion by additional web crawling, subsequent association of hyperlink-based objects with concept objects, and creation of new concept objects as well as by user input to, and editing of, the concept-object database. The concept-object database may be employed, in various embodiments of the present invention, to facilitate web browsing, web-page searching, and other worldwide-web-base activities.

The invention provides a synonym screening method which includes the steps: training the word vector of a big corpus word; mining synonyms of the big corpus word, and particularly, acquiring a candidate synonym set; updating synonym similarity; performing screening to obtain a synonym list. Compared with the prior art, the synonym screening method, the synonyms obtained by big corpus training are wide in coverage, synonyms with good timeliness can be found by adding new big corpuses, the synonyms obtained by screening according to the principle of the requirement of the synonyms for synonymy are higher in quality, and a forceful tool for natural language processing semantic comprehension is added. The synonym screening method is applied to a chatting robot, sentences expressing the same meaning with different words by a user can be more effectively recognized, and the sentence understanding level of the robot is improved.

The invention relates to a Chinese sentence meaning structure model automatic labeling method based on CRF ++, and belongs to the technical field of computer science and natural language processing semantic analysis. Firstly, linguistic data in a BFS-CTC Chinese labeling linguistic database are used, a predicate recognition model, a word relation recognition model and a semantic case type recognition model are obtained through training; secondly, recognition on an original sentence is conducted by using the predicate recognition model, the word relation recognition model and the semantic case type recognition model so that predicate information, work relation information and semantic case type information in the sentence are obtained; and finally, a Chinese sentence meaning structure model is obtained according to collocation rules of the predicate, the work relation and the semantic case types. The method is in the field of semantic analysis, more and comprehensive semantic features are provided, so that the foundation of sentence meaning structure model analysis to the sentence by a computer is laid. Meanwhile, a definite possibility is provided for automatic labeling of the BFS-CTC Chinese labeling linguistic database, both study and practical application of the linguistic data are important, and the method plays a great promoting role in augment of the BFS-CTC Chinese labeling linguistic database.

A semantic data resource of a domain is processed by calculating relevance scores for terms which occur in domain corpora and weighting the semantic data resource depending on the relevance scores calculated for these terms. The semantic data resource may include domain-specific terms and relations, such as a domain ontology, a domain terminology and a domain classification. The domain ontology may include a domain-specific-hierarchy of terms assigned to nodes which are connected by edges and may be encoded in a web ontology language. The relevance scores may be chi-square scores which are calculated depending on a frequency of a term in the domain corpora and an expected frequency of the term.

A computer implemented system and method that combines data from multiple sources to provide a unified view of data using multi-layered conceptual modeling framework. The invention provides means to model conceptual data models at multiple levels of abstraction and a means to map them with each other and with physical models. A query rewriting approach adopted in the present invention is capable of translating queries down multiple levels of conceptual models into physical models and Conceptual modeling framework provide a means to capture and process semantic intent of data more accurately thereby reducing data integration errors significantly.

Various systems may benefit from computer learning. For example, robotics systems may benefit from learning actions, such as manipulation actions, from unconstrained videos. A method can include processing a set of video images to obtain a collection of semantic entities. The method can also include processing the semantic entities to obtain at least one visual sentence from the set of video images. The method can further include deriving an action plan for a robot from the at least one visual sentence. The method can additionally include implementing the action plan by the robot. The processing the set of video images, the processing semantic entities, and the deriving the action plan can be computer implemented.

The invention discloses a method, a system and a server for the push-delivery of the advertisement theme packet to the game client; the method includes that: the tag of the current advertisement information is sent to the server; the server processes semantic matching towards the tag of the current advertisement information and the tag of the advertisement theme packet; the advertisement theme packet with the highest matching degree is sent to the game client; the game client operates the advertisement theme packet with the highest matching degree in the advertisement column. The system comprises a server and a game client; the analyzing unit of the server is used for receiving the tag of the current advertisement information, processing the semantic matching towards the tag of the current advertisement information and the tag of the advertisement theme packet, marking the advertisement theme packet with the highest matching degree and informing the advertisement push-delivery unit to send the advertisement theme packet to the game client. The method, the system and the server realize the delivery of a plurality of advertisement information, extend the profit space and maintain good game experience for the game players.

The invention relates to a domain term obtaining method and system. The domain term obtaining method comprises the following steps of: obtaining network data, and carrying out domain classification on the network data; extracting a first candidate domain term from the network data subjected to domain classification; processing semantic integrity of the first candidate domain term to obtain a second candidate domain term; and calculating a domain correlation of the second candidate domain term, and comparing the domain correlation of the second candidate domain term with a correlation threshold to obtain a domain term. According to the domain term obtaining method and system, the first candidate domain term is extracted from the network data, and is subjected to the semantic integrity to obtain the second candidate domain term, the correlation of the second candidate domain term is calculated and is compared with the correlation threshold to obtain the domain term, thus a large quantity of domain terms can be more accurately obtained.

The invention discloses a method for verifying the semantic consistency of land-air conversation based on improved LSTM-RNN. The method comprises the steps: making a corpus: making a special vocabulary according to the conversation standard and the corpus of the civil aviation and obtaining a one-hot vector of the words: generating a semantic vector of two sentences in each sentence pair; processing the semantic vectors through an average pooling method and inputting the processed semantic vector into an MLP model; learning the degree of correlation between two semantic vectors through the MLP; determining the consistency of semantics through KNN. An original method for processing a sequence based on LSTM-RNN is lower in verification precision even if preventing the gradient of an independent RNN algorithm from disappearing. The method provided by the invention employs the average pooling method for preventing the overfitting phenomenon, innovatively employs the MLP for the computing,and enables the model to be more complete in the learning degree of the correlation. Because a model in deep learning is used for automatically learning the sample features, the method does not need the statistical analysis of mass data.

A method and system are described for validating process models. A process semantic model including a semantic directed graph including nodes associated with tasks and edges associated with a direction of flow of execution of the tasks may be obtained. Edges entering nodes include annotations including precondition semantic indicators associated with the entering edges and edges exiting nodes include annotations including postcondition semantic indicators associated with the exiting edges. The process semantic model may be traversed to determine a flow of execution of activities associated with the tasks based on visiting the nodes based on a depth-first traversal. A validity of execution associated with a flow of execution of the activities associated with the tasks may be determined based on checking a validity of execution status based on a semantic processing of one or more semantic annotation indicators associated with the precondition semantic indicators and the postcondition semantic indicators.

A method, apparatus and program storage device for delimiting words is provided and requires adding an extensible vocabulary markup to a plurality of markup languages, wherein the plurality of markup languages comprises a knowledge representation, and processing a provided term from one or more vocabularies for use in the knowledge representation.

A method and apparatus for processing semantic vectors are disclosed. The method includes receiving a user request; Obtaining a word vector by performing a preset process on the user request; And inputting the word vector into a preset sentence vector encoding model and outputting a semantic vector of a sentence in the user request. The present application solves the technical problem that the processing effect is poor when the text semantics are similar. The semantic vector of the sentence can be accurately obtained by the present application, and can be used for similarity calculation when two texts are semantically similar.

The invention provides a navigation environment visualization representation method for ship visual navigation. The method includes the following steps that: a geometric layer is constructed, wherein the geometric layer includes a geometric model of navigation environment and thematic elements, initial positions and statuses, ship dynamic data and movement navigation point position information; a physical layer is constructed, wherein the physical layer is used for representing the physical characteristics of the thematic elements in the navigation environment and laws of physics which the thematic elements should follow in the navigation environment; a temporal layer is constructed, wherein the temporal layer concludes the temporal characteristics of the navigation environment and thematic elements to states and processes and specifies a researched time domain as a continuous model based on time density; and a semantic layer is constructed, wherein the semantic layer includes topic semantic, space semantic and process semantic and reflects the physical natures of the dynamic change processes of the navigation environment elements and the influences of the navigation environment elements on ship visual navigation. With the method of the invention adopted, the navigation environment and thematic elements can be integrated under a unified time-space frame, and centralized and unified representation of navigable environment information can be realized, and the accuracy and efficiency of visual time-space analysis can be improved.

The present invention discloses a method and system for processing semantic fragments. Some embodiments of the present invention provides a method for processing semantic fragments. The method comprises: obtaining a plurality of groups of semantic fragments, the plurality of groups of semantic fragments at least including a first group of semantic fragments generated from a first data processing flow and a second group of semantic fragments generated from a second data processing flow, the first data processing flow being different from the second data processing flow; and merging the first group of semantic fragment and the second group of semantic fragment based on semantic equivalence. A corresponding system is also disclosed.

The invention discloses an IO sequence-based virtual machine abnormal behavior detection method and system. The detection system comprises an asynchronous acquisition module, a process obtaining module, a communication module and a detection module; a virtual machine IO sequence is asynchronously acquired in a VMM through an asynchronous virtual machine IO sequence acquisition technology to reduce performance loss in a virtual machine running process to the maximum extent; virtual machine process semantics is dynamically obtained by utilizing a VMM layer through a virtual machine IO sequence process semantics dynamic obtaining technology to realize reasonable mapping of the virtual machine IO sequence and a virtual machine process so as to facilitate rule analysis of the IO sequence and confirmation of a detection result; and an abnormal detection model is built in combination with a virtual machine IO short sequence and a Markov chain through a Markov chain-based virtual machine malicious behavior detection technology to finish the detection of malicious behaviors in a virtual machine. According to the method and the system, virtual machine IO-based abnormal attack behaviors can be accurately discovered, so that the security of a cloud computing platform is protected.

The invention discloses a dispute focus-based class case retrieval method and system, a dispute focus-based class case retrieval device and a medium, and relates to the field of natural language processing semantic similarity. The method comprises a dispute focus extraction processing step, a word segmentation processing step, a word vector acquisition processing step, a sentence vector acquisition processing step, a final dispute focus sentence vector acquisition processing step, an index database storage step, a final dispute focus sentence vector P acquisition step corresponding to a dispute focus to be retrieved and a corresponding retrieval step, a case corresponding to the topK similar dispute focus is obtained. According to the method, a sentence vector of a dispute focus is constructed by using a related model used for generating a word vector in the field of natural language processing, the construction method adopts an unsupervised mode, the complexity of manually labeling similar sentences is removed, the method is simple, and meanwhile, the method is based on semantic retrieval and is high in retrieval speed and accurate in retrieval result; and the recall rate and theaccuracy rate of class case retrieval based on dispute focuses are improved.

The present invention is directed to a method and computer readable medium for automatically generating a Uniform Resource Locator (URL) from schema and semantics of an organizational structure. The semantically intelligent URL conveys knowledge about content it accompanies by defining patterns of possible URL parameters within an organizational structure of related content stored in a database, each pattern being represented by entities and semantic information formed of attributes and relationships. The patterns of the URL organize categories of information and form relationships among the entities. The patterns are used for automatically generating a URL along with semantic information about the content in a resource or web page, wherein said semantic information explains the content, such as with attributes of entities, such as in a name value pair that correlates to the content of a page.

A unit property for defining the manufacturing industry software unit is described with the use of the action of an access to a domain concept data model element of an activity that is realized by the software unit described in an implementation specification format. As a result, the manufacturing industry software unit can be formally described as the property that defines the action of the description when it is described by using the software unit profiling template described in the implementation specification format defined by ISO 16100-2. Also, a request in conducting retrieval can be described in the same format as the above, and the process semantics of the existing manufacturing industry software unit can be efficiently registered in the directly expressed and in a common form, and effectively retrieved.

A unit property for defining the manufacturing industry software unit is described with the use of the action of an access to a domain concept data model element of an activity that is realized by the software unit described in an implementation specification format. As a result, the manufacturing industry software unit can be formally described as the property that defines the action of the description when it is described by using the software unit profiling template described in the implementation specification format defined by ISO 16100-2. Also, a request in conducting retrieval can be described in the same format as the above, and the process semantics of the existing manufacturing industry software unit can be efficiently registered in the directly expressed and in a common form, and effectively retrieved.

The invention discloses a method for optimizing an enterprise business process and support resources based on directed hyper-graph. The method specifically comprises the following steps: modeling an enterprise, establishing a dynamic mapping between models; establishing a business initiator model, recommending a corresponding business process according to a target of the business initiator; establishing a business process model based on directed hyper-graph for optimization according to the actual resource background of the business process initiator and the resource situation of an environment enterprise; selecting a possible cooperative enterprise according to the resource background of the enterprise and the resource conditions in the environment enterprise; and performing the optimizing transformation of process structure and the optimization of process structure and support resource by utilizing properties of the directed hyper-graph and procedural semantics attached on the model according to the specific resource environment situation inside and outside the enterprise to obtain the process with optimized structure and support resource. According to the method disclosed by the invention, the enterprise is instructed to select an appropriate partner enterprise according to the specific targets and the resource conditions of small and medium-sized enterprises, and the information systems are integrated.

The invention relates to a domain term obtaining method and system. The domain term obtaining method comprises the following steps of: obtaining network data, and carrying out domain classification on the network data; extracting a first candidate domain term from the network data subjected to domain classification; processing semantic integrity of the first candidate domain term to obtain a second candidate domain term; and calculating a domain correlation of the second candidate domain term, and comparing the domain correlation of the second candidate domain term with a correlation threshold to obtain a domain term. According to the domain term obtaining method and system, the first candidate domain term is extracted from the network data, and is subjected to the semantic integrity to obtain the second candidate domain term, the correlation of the second candidate domain term is calculated and is compared with the correlation threshold to obtain the domain term, thus a large quantity of domain terms can be more accurately obtained.

The invention provides a method and device for extracting keywords from text. Wherein, the method for extracting keywords from the text includes: performing word segmentation processing on the text; labeling the semantic class numbers for the words obtained by the word segmentation processing; using the semantic class numbers as nodes to form a synonym network; selecting nodes from the synonym network as keywords. By adopting the method and device for extracting keywords from text provided by the present invention, the efficiency of extracting keywords can be improved.

The present invention is directed to a method and computer readable medium for automatically generating a Uniform Resource Locator (URL) from schema and semantics of an organizational structure. The semantically intelligent URL conveys knowledge about content it accompanies by defining patterns of possible URL parameters within an organizational structure of related content stored in a database, each pattern being represented by entities and semantic information formed of attributes and relationships. The patterns of the URL organize categories of information and form relationships among the entities. The patterns are used for automatically generating a URL along with semantic information about the content in a resource or web page, wherein said semantic information explains the content, such as with attributes of entities, such as in a name value pair that correlates to the content of a page.