33 results about "Generic knowledge" patented technology

An electronic assistant which dispatches tasks on the user's behalf and according to his or her preferences is disclosed. The assistant has an enactor for processing data received from a sensor and for changing its environment via an actuator. The enactor receives instruction from a predictor / goal generator, which in turn is connected to a general knowledge warehouse. Additionally, the warehouse and the predictor / goal generator are connected to a plurality of specialist knowledge modules, including a scheduler, an information locator, a communicator, a form filler, a trainer, a legal expert, a medical expert and other experts. The electronic assistant provides an interface which frees the user from learning complex search languages and allows some functions to be automatically performed. A variety of machine learning processes allow the assistant to learn the user's styles, techniques, preferences and interests. After learning about the user's interests in particular types of information, the assistant guides the user through the process of on-line information source selection, utilization, and interaction management via the information locator. The information locator generates a query conforming to the user characteristics for retrieving data of interest. The information locator next submits the query to one or more information sources. Upon receipt of results of the submitted query, the information locator communicates the results to the user, and updates the knowledge warehouse with responses from the user to the results. The assistant supports the ability to refine the query and to manage the costs associated with the search. Further, the assistant automatically incorporates data relating to changes in the query interface and other relevant characteristics of the information sources so that search command sequences can be altered without user interaction. The search configuration of each search carried out by the user is saved in a database. The data maintained in the database includes keywords and concepts for search, interval between subsequent searches, deadline for the search, the number of documents to acquire from each engine, and domain over which to do the search, including the preferred set of search engines or the preferred set of news groups.

Systems and methods disclosed herein relates to building an intelligent assistant that can take in human requests / commands in simple text form, especially in natural language format, and perform tasks for users. Systems and methods are disclosed herein in which knowledge of how to interpret users' requests and carry out tasks, including how to find and manipulate information on the Internet, can be learned from users by the designed assistant, and the knowledge can be subsequently used by the assistant to perform tasks for users. Using the disclosed methods, the designed assistant enables a user to teach the assistant, by actually performing the task manually through the provided user interface, and / or by referring to some knowledge that the assistant already knows; the designed assistant may generate more generic knowledge based on what it learns, and can apply the more generic knowledge to serve requests that it has never seen and never directly learned, and can revise / improve the knowledge according to execution result / feedback. The methods and systems being disclosed here are useful for building an intelligent assistant, especially a universal personal assistant and an intelligent search assistant.

An intellectual asset protocol for defining data exchange rules and formats for universal intellectual asset data objects, and systems, methods, and computer program products related to same. The system includes an intellectual asset protocol system that acts as an engine in the definition of data exchange rules and formats for universal intellectual asset documents. Also included is a front end system that preferably provides a graphical user interface to enable users to access the intellectual asset protocol system. In addition, an intellectual asset database is included that stores collections of intellectual asset objects (and information related to same), one or more embodiments of an intellectual asset protocol, and so forth. The intellectual asset protocol system interacts with an Intellectual Property Asset Manger (IPAM) server.

In an automated Question Answer (QA) system architecture for automatic open-domain Question Answering, a system, method and computer program product for predicting the Lexical Answer Type (LAT) of a question. The approach is completely unsupervised and is based on a large-scale lexical knowledge base automatically extracted from a Web corpus. This approach for predicting the LAT can be implemented as a specific subtask of a QA process, and / or used for general purpose knowledge acquisition tasks such as frame induction from text.

The invention discloses a method and a device for constructing a patent data knowledge map. The method comprises the following steps: obtaining patent data of an existing patent database, preprocessing the patent data to unify a patent data format, and segmenting the patent data after being merged with the same type to obtain the segmentation data of each type of patent data; performing knowledgeextraction of preprocessed patent data, performing data cleaning of word segmentation data of each type of patent data to obtain the corresponding subject original file, extracting keywords to obtainsubject words, and constructing a patent subject database for each type of patent data; defining the entity of patent data, determining the subject of patent data, identifying the entity and subject of patent according to the general knowledge map, mining the semantic relationship between the entity and the subject, and constructing the patent data knowledge map.

A general-purpose knowledge finding method for efficient knowledge finding by selectively sampling only data in large information amounts from a database. Learning means 104 causes a lower-order learning algorithm, inputted via an input unit 107, to perform learning on plural partial samples generated by sampling from data stored in a high-speed main memory 120, to obtain plural hypotheses. Data selection means 105 uses the hypotheses to estimate information amounts of respective candidate data read from a large-capacity data storage device 130, and additionally stores only data in large information amounts into the high-speed main memory 120. A control unit 106 repeats the processing a predetermined number of times, and stores obtained final hypotheses. A prediction unit 102 predicts a label value of unknown-labeled data inputted into the input unit 107 by the final hypotheses, and an output unit 101 outputs the predicted value.

The invention provides a method and device for generation of a Chinese general knowledge atlas with entity relation attributes. The method includes the steps that entities in multiple external data sources and databases are subjected to named entity identification, and the multiple named entities obtained through identification are connected, so that multiple entity relations are obtained; attribute identifiers are added into the multiple entity relations, and the multiple entity relation attributes are obtained; according to the multiple entities, the multiple entity relations and the multiple entity relation attributes, the Chinese general knowledge atlas with the entity relation attributes is generated. According to the method and device, through addition of the identifiers into the entity relations, the entity relation attributes are built, and when answers are extracted through the knowledge atlas, the efficiency and the integrity of answer extraction are improved due to increase of the entity relation attributes, so that a user is provided with service by the knowledge atlas more efficiently.

The invention belongs to the technical field of software engineering and intelligent software development, and particularly relates to an API knowledge graph construction method based on a reference document. The method comprises the steps that a basic skeleton structure of an API element is obtained through document structure analysis, and function description and use mode description are recognized by automatically classifying sentences in descriptive content of the API element; performing common concept identification and linking among the description information of different API elements to realize internal knowledge fusion; and performing concept linkage between the common concept in the API element description information and the related technical concepts in the general knowledge graph to realize external knowledge fusion. The constructed API knowledge graph comprises API packets, classes, interfaces, methods, attributes, abnormalities, method parameters, return values and relations among the elements. According to the constructed API knowledge graph, intelligent applications such as API knowledge semantic query, automatic question answering, auxiliary code understanding andcode recommendation are supported through structured knowledge representation.

The invention discloses a word semantic tendency prediction method based on a universal knowledge network. The method comprises the following steps of: (1) judging whether an unknown word exists in a sentiment word dictionary, if so, returning the polarity of the unknown word, and otherwise, executing the step (2); (2) selecting a positive reference word set and a negative reference word set; (3) calculating the tightness degree of the unknown word and the positive reference word set; (4) calculating the tightness degree of the unknown word and the negative reference word set; (5) calculating difference between the tightness degree of the unknown word and the positive reference word set and the tightness degree of the unknown word and the negative reference word set; and (6) according to the difference in the step (5), selecting a threshold space and determining the polarity of the unknown word. The word semantic tendency prediction method based on the universal knowledge network has the advantages that the semantic similarity of words is taken into consideration, the association of the words is combined, area threshold judgment is performed, the words are prevented from being endowed with wrong sentiment tendency, and the accuracy of semantic tendency judgment is improved.

The present invention relates to a method for answering a query (60) directed to a knowledge base. The basic idea under-lying the present invention is to define a generic knowledge base with generic representatives (30) and generic rules (140). Generic representatives (30) represent the structure of objects (20) of the knowledge base. This enables generic rules (140), which represent the structure of the rules (50) of the knowledge base, to be applied to the generic representatives (30) without evaluating the corresponding objects (20) in a first step. In a second step, the inferred generic answers (90) (which are generic representatives (30)) can be used to restrict the original query (60) to objects (20) which have an appropriate structure. In a similar way, the generic representatives (30) applied for inference can be used to restrict the rules (50) of the knowledge base to objects (20) which have an appropriate structure.

The invention belongs to the technical field of software engineering, and particularly relates to an API element comparison result automatic generation method based on a knowledge graph. The method comprises the basic steps of extracting API structure knowledge from an API reference document to serve as a skeleton of a knowledge graph; extracting eight different types of knowledge from the API structure knowledge and descriptive sentences corresponding to the API elements; extension of concepts and relations is achieved by introducing a general dictionary and general knowledge graph external knowledge, and API knowledge which is extracted from different sources and has different expression modes is fused; and generating a comparison result in a table form for a given pair of API elements in an alignment manner. According to the invention, a mode of comparing the API elements from different dimensions of API functions, characteristics, concepts and the like is provided for software developers, and the developers can quickly know the commonality and difference of the API elements, so that the developers can quickly and accurately select the API elements meeting the current development requirements and learn API related knowledge.

A voice browser for use in a customer interaction management system for interacting with a customer, based on data contained in one or more generic knowledge trees. The system can traverse the knowledge tree based on customer responses to presented questions until a leaf node of the knowledge tree is reached. The information contained in the leaf node is then presented to the customer.

The invention discloses a recommendation method based on a knowledge graph and long and short term interests of a user, which comprises the following steps: acquiring an item set and mapping the item set to a general knowledge graph, and acquiring user-item interaction information; aggregating neighborhood entities by adopting an entity neighborhood aggregation mode based on the knowledge graph convolutional network to obtain project feature vector representation of the to-be-recommended project; learning user long-term preference vector representation through a preference propagation method; inputting interested items in the historical preference set of the user into a gating loop unit for training according to a time sequence to obtain short-term preference vector representation of the user; and overlapping and fusing the user long-term and short-term preference vector representations according to columns, performing full connection layer processing to obtain a final user preference vector representation, performing inner product calculation on the similarity of corresponding feature dimensions with the item feature vector representation of the to-be-recommended item, inputting the similarity into a multi-layer perceptron, and predicting the probability that the user is interested in the to-be-recommended item. The method improves the recommendation effect, and has the characteristics of high interpretability, strong adaptability and high precision.

A method for managing a universal knowledge repository comprising: receiving streams of data from a plurality of data sources; processing the streams of data from the plurality of data sources, the processing the streams of data from the plurality of data sources performing data enriching to provide data enriched data streams; and, storing the data streams and the data enriched data streams within the universal knowledge repository as a collection of knowledge elements.

The invention discloses a knowledge graph application database generation method and device. The method comprises the following steps of obtaining a theme of an application database to be generated, and obtaining an entity of a general knowledge graph; and according to the theme of the to-be-generated application database, screening out entities related to the theme from the entities of the general knowledge graph as core entities, and generating a personalized entity of the to-be-generated application database, and generating an application database according to the personalized entity and the core entity, so that when the application database is generated, the application database is gradually executed according to the preset steps, the logic dependence in the data processing process isreduced, the maintenance cost of the product is reduced, and the technical problems that in the prior art, the logic dependence of the data processing process is complex, and the maintenance cost is too high are solved. In addition, the data information in the general knowledge graph is fully utilized, so that the universality of the whole scheme is ensured.

A computer-implemented system with a processor provides a reversible transfer of an atomic token from one side of an imperfect link to the other, such that if the protocol (or process) on either side fails at a critical moment, the atomic token will be found on both sides to be verifiably incomplete, unless the protocol has completed successfully past its ‘irreversible threshold’ on both sides.

The invention discloses a hybrid federated learning method based on knowledge transfer. The method comprises the following steps: introducing an incentive mechanism in federated learning, firstly uploading, by each piece of equipment, data distribution conditions, and making, by a server, a decision according to demand information of current data, so that a data selection algorithm based on mutual information is performed, and corresponding rewards are given to the equipment uploading the data; then performing training in a deep learning model which is the same as each piece of local equipment by utilizing collected shared data to obtain an auxiliary model; and transferring general knowledge of the auxiliary model to an aggregation model. According to the invention, in different federated training rounds, different transfer methods are used to transfer the general knowledge of the auxiliary model to the aggregation model according to the equipment aggregation model and the auxiliary model, so that an optimized global model is obtained; and the ability to distinguish general features can be provided for the aggregation model in a few rounds, so that local rounds of an equipment model are reduced, and rapid convergence and high accuracy of the global model are realized.

The invention discloses a meta-knowledge fine adjustment method and platform based on domain invariant features, and the method comprises the steps: learning highly-transferable common knowledge, i.e., domain invariant features, on different data sets of similar tasks, and learning common domain features on different domains corresponding to different data sets of similar tasks in a fine adjustment network set; and any different domains can be quickly adapted. According to the method, the parameter initialization capability and generalization capability of the similar task universal language model are improved, and finally, the universal compression architecture of the similar downstream task language model is obtained through fine adjustment. In the meta-knowledge fine adjustment network,a domain invariant feature loss function is designed, and domain-independent general knowledge is learned, that is, a learning target of one domain invariant feature is minimized to drive a languagemodel to have domain invariant feature coding capability.

A system of constructing, referring and manipulating new types of uniform, strictly defined and generically connective objects in computing environments, allowing identification, connection, insertion and re-definition of fully as well as partly defined objects and providing a new self-organizing, open computing space for referring these objects in several independent tree Ostructures, so that the system, its constructed systems, scopes of connections and objects all share the same space lifetime. The invention provides a new approach and architecture for building structures of objects, useful for generic knowledge systems, interactive and automatic language processing and other applications requiring automatic next order processes and global referring in deterministic and non-deterministic environments. The invention also provides a method to develop tools to adapt computer systems to the logic of specific users.

The invention discloses a repeated material entity recognition method based on mutually different feature vectors. The method comprises the following steps: S1, inputting a material data set; S2, preprocessing material data; S3, constructing the mutually different feature vectors and category vectors; s4, training and testing a probabilistic neural network classifier; S5, obtaining mutually different feature vectors recorded by the to-be-tested material; s6, inputting the mutually different feature vectors of the material records obtained in S5 into a trained probabilistic neural network classifier, if the output result of the probabilistic neural network classifier is 1, representing that the two material records have difference in semantic expression, and if the output result of the probabilistic neural network classifier is 0, representing that the two material records have difference in semantic expression; if yes, the two material records are the same in semantic expression. The difference between entity feature descriptions is considered, feature information of entities is fully utilized, and the limitation of a universal knowledge base in measuring semantic similarity between entities in different fields is solved.

One embodiment of the invention relates to the technical field of information processing and discloses a method of making teaching material and terminal equipment. The method comprises following steps: obtaining a plurality of source knowledge point texts, wherein the source knowledge point texts comprise a plurality of test question content; merging the source knowledge point texts according to a preset format to obtain intermediate knowledge point texts; performing de-duplication on the intermediate knowledge point texts to obtain general knowledge point texts; generating a general teaching material from the general knowledge point texts; obtaining a teaching material directory to be processed; extracting test content matched with the teaching material directory to be processed from the general teaching material and preparing teaching material according to the extracted test content and based on the sequence of the teaching material directory to be processed to obtain target teaching material corresponding to the teaching material directory to be processed; by means of the embodiment of the invention, resource can be reused; the preparing flow of teaching material is simplified; the preparing time is shortened and the manpower and finance cost is reduced; the teaching material is general so as to be helpful for later maintenance and update.

A learning apparatus in a digital environment is advantageous to interaction and communication among users who use a knowledge point structure for learning. The learning apparatus in the digital environment constructs a structurized knowledge library by editing knowledge points, tags of the knowledge points and a relationship among the knowledge points; then records a user's mastery degree for the knowledge points on the basis of the tags; and records knowledge learning information and social attribute information thereof for each user, so as to construct a general-class learning record, a single-class learning record and a general knowledge learning system of the user, which can also be used for the user to view the popularity of each tag in the knowledge library.

The invention discloses a human-object-space interaction model construction method based on a knowledge graph, and belongs to the technical field of knowledge graph construction and smart communities. The invention discloses a human-object-space interaction model construction method based on a knowledge graph, and the method comprises the following steps: S1, obtaining information from a large number of active and passive sensing devices, and constructing a knowledge base; s2, fusing perceptual information in the knowledge base constructed in the S1 to form entity-relation-entity structural data, and constructing a general knowledge graph conceptual model with entity-relation attributes; the invention provides a human-object-space interaction model construction method based on a knowledge graph, which can effectively solve the technical problems of difficulty in multi-source information extraction, incapability of fusing heterogeneous data, incapability of human-object-space interaction and the like caused by poor perceptual technology universality and complex entity relationship in an intelligent community environment. And a method support is provided for dangerous event monitoring and early warning and community environment situation awareness in the smart community.

The invention discloses a safety and stability analysis method and system based on a man-machine hybrid power system. The method comprises the steps that in a man-machine mode, according to a general knowledge base, a training execution agent and an interaction agent, and according to the execution agent and the interaction agent, a special knowledge base is determined, and the man-machine mode is a human dominant-machine following mode; and in a robot mode, according to the special knowledge base, executing a power system task, the robot mode being a mode of robot independence-human supervision. Therefore, a man-machine hybrid power system task execution method is formed, the AI interpretability level and the AI credibility are improved, multi-dimensional cognition of complex power grid operation characteristics is achieved, the safety and stability level of a power system is improved, and the workload of people in actual production is reduced.

The invention discloses a motor optimization design method and system, and the method comprises the steps of analyzing the core technology of the motor design, splitting a design function, and obtaining the design generic knowledge and rules; and storing the design generic knowledge and rules to a cloud; enabling a design end to obtain the design common knowledge and rules from the cloud, optimizethe motor design parameters within the range of the design common knowledge and rules, and upload the optimized data to the cloud. According to the motor optimization design method and system, by acquiring the design generic knowledge and rules and storing in the cloud, the reference is provided for the motor design process, when the motor design is carried out, optimization is carried out withinthe design generic knowledge and rules range to acquire the motor meeting the use requirement, so that the design difficulty is reduced, the motor design process is more scientific and efficient, andthe manpower and material resource consumption is reduced. Moreover, the optimized design data is uploaded to the cloud, and a cloud database is enriched.