49 results about "Semantic technology" patented technology

The embodiments of the present invention provide a computer-implemented method and system for translating user keywords into semantic queries based on domain vocabulary. The system receives the user keywords and search for the concepts. The concepts are transformed into a connected graph. The user keywords are translated into precise access paths based on the information relationship described in conceptual entity relationship models and then converts these paths into logic based queries. It bridges the semantic gap between user keywords and logic based structured queries. It enables users to interact with the semantic system by articulating the information in a structured query language. It improves the relevance of search results by incorporating semantic technology to drive the mechanics of the search solution.

A system is provided for managing data sets in a cloud processing and / or federated environment. In an embodiment, the system described herein may be used in connection with cloud processing of big data sets. The term “big data,” as used herein, may be generally defined to describe data sets so large and complex that they become difficult to work with using on-hand database management tools. The system described herein enables the persistent storage of semantic technology statements for big data sets for processing in a cloud processing and / or federated environment.

The invention relates to the technical field of voice semantics, and discloses a statement intention identification method. The statement intention identification method comprises the following steps:preprocessing an original statement sample to obtain a preprocessed sample; extracting a statement feature vector from the preprocessed sample; based on the statement feature vector, training a statement intention identification model by using a cross entropy cost function method to obtain a trained statement intention identification model; obtaining a to-be-identified target statement; based onthe target statement, outputting a statement intention corresponding to the target statement by using a trained statement intention identification model; and obtaining an answer matching the statementintention corresponding to the target statement from an answer database, and displaying the answer to a user. The invention further provides a statement intention identification device and a computerreadable storage medium. According to the statement intention identification method, the abstract features of the voice can be extracted by using the deep neural network, and the statement can be identified more accurately.

The invention discloses a knowledge graph-based cross-department early-diagnosis and decision support system for chronic kidney diseases. The system comprises a patient information model establishmentmodule, a patient information model library storage module, a knowledge graph association module, a knowledge graph reasoning module and a decision support feedback module. According to the invention, a patient information model is constructed, and an OMOP CDM standard term system is utilized to construct electronic medical record data of patients into the patient information model with unified concept codes and unified semantic structures; the advantages of a semantic technology in data interaction and expansion are given to play, so the system has better adaptability and expandability to heterogeneous data of different hospitals; and meanwhile, clinical suggestions obtained through knowledge reasoning based on the knowledge graph are originated from clinical guidelines and doctor experience conforming to evidence-based medicine, and reasoning process and suggestion reasons can be traced and acquired by constructing reasoning instances, so the reasoning process and the suggestion reason can be given while clinical suggestions are given, and the credibility of doctors to decision support suggestions is improved.

This invention discloses an excavation method of generalized association rule used to analysis the compound rule of Chinese medicine; the method is essentially an excavation of association rule combined with excavation method of generalized association rule expressed by field knowledge, using semantic Web technology as the principal means expressed by field knowledge. The method relates to a knowledge discovery device, the device use nomenclature system and field rule provided by the knowledge database to complete the data excavation process and submit the excavation results to knowledge database in the form of knowledge proposal, then certified and evaluated by the field experts. The data excavation process comprises: extract required data from Chinese medicine data resource first, then excavate senseful frequency model in the data and make semantic labels, extract and reason the generalized rules according to the labeled model. The use of semantic Web technology fields to construct knowledgebase, and the attainment of information and knowledge transfer between the components the methods referred to.

The invention relates to an electronic commerce market product / service information management and monitoring platform. The platform comprises a product / service information issuing system, a product / service information management system and a product / service information monitoring system; the product / service information issuing system issues standardized product / service information to the webpage of the electronic commerce platform by utilizing the preset electronic commerce market product / service information standard; the product / service information management system backs up and manages the standardized product / service information issued by the product / service information issuing system; and the product / service information monitoring system draws and analyzes standardized product / service information issued on the webpage of the electronic commerce platform by using semantic technology to realize on-line monitoring to product / service information. The technical scheme in the invention utilizes electronic commerce market product information standard to support producing enterprises to design and issue standardized product information, monitors irregular and counterfeit and shoddy product information of electronic commerce market and publishers of the information, and provides related department law enforcement to purify electronic commerce market information.

The present invention relates to a method and apparatus for extracting the semantic relations of context and a recording medium storing the program source thereof. The present invention is intended to detect technical terms by parsing full text of documents and recommend research topics using the relations therebetween using a method of searching for a first concept connected to a specific vocabulary word through a casual relationship and a second concept connected to the first concept through a causal relationship, determining the relation therebetween and performing recommendation as research topics. The present invention has the industrial utilization advantage of improving the overall search efficiency and utility of massive databases.

The invention relates to an intelligent classroom knowledge management platform and an intelligent classroom knowledge management method based on semantic techniques and gamification. The platform comprises a knowledge management platform for results, wherein the knowledge management platform for results is used for integrating and analyzing individual data collected by an individual data collection platform, used for establishing knowledge banks for each student and each teacher, used for forming association of entity data of the knowledge banks, used for establishing knowledge spectrums, and used for establishing a management platform with intelligent data retrieval and analysis service by using data analysis techniques. By using in-class and off-class gamification intelligent equipment, diverse learning modes are provided for students, and thus the learning enthusiasm of the students can be maximized; and student behaviors are subjected to data analysis, individual recommendation and feedback are provided after analysis, and thus individual education is achieved.

The invention provides a semantic analysis technology-based sensitive public sentiment content recognition method and a pre-warning system. The method comprises the following steps of: importing a Chinese word segmentation standard lexicon, increasing a sensitive word vector library, classifying sensitive words, setting a core word as a standard word for each type of sensitive words, and setting distance vectors from other same-type words to the standard word by taking the standard word as a source node; carrying out voice recognition on an audio file, and recognizing sensitive words and related sentiment words; carrying out clustering analysis, recognizing sensitive words and sentiment words related to semantic meaning, and carrying out semantic analysis judgement; and recognizing and outputting sensitive contents. According to the method, the sensitive word vector library is established on the basis of a traditional natural language lexicon, and vector distance relationship among same-type sensitive words is established. In practical application field, sensitive content recognition and pre-warning can be automatically carried out on real-time communication contents or classroom teaching contents. By utilizing the method, sensitive sematic judgement can be realized and the correctness is greatly enhanced so that sensitive public sentiment content monitoring is realized.

The present invention relates to a method for an online distribution information system. The online distribution information system comprises a plurality of distribution clients, a proxy network, a distribution / subscription module, a plurality of subscription clients, and a management cluster. The distribution / subscription module comprises a message memory, a message sectionalizer, a user subscription manager, and a message dispatcher. The method comprises the key process related to the system. According to the method disclosed by the present invention, in order to solve the problem of information resource understanding of the computer, the semantic technology is combined with the distribution and subscription system organically, and the intention of the distribution client and the subscription client can be grasped from the perceptive of the semantic layer, so that the distributed event can be more accurately matched with the requirements of subscribers, the matching quality of subscription and distribution can be improved, and the performance of the entire system can be improved.

The invention relates to the technical field of voice semantics, and discloses an intelligent semantic document recommendation method, which comprises the following steps of obtaining all document information in a document database as a training set; acquiring historical query information of a user and document information selected by the user in history, and performing preprocessing operation onthe historical query information and the document information selected by the user in history to obtain feature points serving as a tag set; using the training set and the label set to train the constructed intelligent semantic document recommendation model; and receiving query information input by the user, performing document retrieval from the document database by utilizing the intelligent semantic document recommendation model, sorting document retrieval results, and outputting the sorted document retrieval results to the user. The invention further provides an intelligent semantic document recommendation device and a computer readable storage medium.

The invention relates to the technical field of phonetics and semantics, and discloses a music separation method. The music separation method comprises the steps that initial sample data is acquired,the initial sample data comprises initial sample data of various musical instruments, and the initial sample data of one of the musical instruments comprises initial sample voice of the musical instrument and initial sample frames of the musical instrument; the initial sample data of the different musical instruments are selected from the initial sample data and mixed to generate mixed audio dataas training data, and based on the training data, training is carried out to obtain a separation musical instrument acoustic model; then input to-be-separated mixed audios and videos are acquired, thetrained separation musical instrument acoustic model is utilized, and voice of the same musical instruments and images corresponding to the voice of the same musical instruments are separated out from the to-be-separated mixed audios and videos. The invention further provides a music separation device and a computer readable storage medium. The music separation method and device and the computerreadable storage medium have the advantages that convenience can be provided for adjusting the videos and audios for musicians, the separation is accurate, and the consumption of manpower resources isless.

The invention relates to the technical field of speech semantics, and discloses a wrong sentence detection method. The detection method includes the steps that a target sentence is obtained; i words included by the target sentence are identified; the i words are successively input into a pre-trained language model according to the order in the target sentence, and the confusion degree and / or logarithmic likelihood probability of the target sentence are / is calculated by the language model; and when the confusion degree of the target sentence is greater than the preset confusion degree and / or the logarithmic likelihood probability of the target sentence is less than the preset logarithmic likelihood probability, it is judged that the target sentence is a wrong sentence. The invention furtherprovides a wrong sentence detection device and a computer readable storage medium. According to the wrong sentence detection method and device, and the computer readable storage medium, whether a sentence is the wrong sentence or not can be identified.

The invention relates to the technical field of speech semantics, and provides a speech synthesis method, an electronic device and a computer storage medium, the method comprising: acquiring a presetverbal skill in a response scene, the verbal skill comprising a fixed text and a variable text; screening out timbre features corresponding to the response scene from a preset voice library accordingto the response scene to record a fixed text in the verbal skill to obtain fixed voice; performing voice synthesis on a variable text in the verbal technique according to the screened tone characteristics to obtain variable voice with the same tone characteristics as the fixed voice; and finally, splicing the fixed voice and the variable voice to generate a synthesized voice with the timbre characteristics. According to the invention, the voice with unified timbre is synthesized according to the response scene, so that the voice in human-computer interaction is natural and coherent, and the user experience is further improved.

The invention relates to the field of artificial voice and semantic technologies, and discloses a baby cry detecting method. The method comprises the following steps: by utilizing a baby voice frequency monitor placed within the range having a present distance with a baby and adopting the form of PCM data stream, collecting real-time voice frequency data signals of the baby; carrying out framing processing on the real-time voice frequency data signals of the baby, thus obtaining framing voice frequency data signals, calculating the fundamental tone frequency of the framing voice frequency datasignals, and judging that whether the baby is in the cry state or not according to the fundamental tone frequency; and when the judging shows that the baby is in the cry state, through terminal pushservice, pushing the cry information to a preset user terminal, and storing the cry information to a cloud. The invention further provides a baby cry detecting device and a computer readable storage medium. With the technical scheme, the automatic detection and judgment for the baby cry are realized.

The invention discloses a traffic event detection system based on the semantic technology and a method for the same. The traffic event detection system consists of a collection module, a wireless transmission module and a background server module which are connected. The collection module consists of a plurality of detectors for collecting necessary traffic data information including time, the spot, the traffic flow, vehicle occupancy rate and average vehicle speed; the wireless transmission module transmits the collected information to the background server module; and the background server module comprises a support vector machine algorithm processing module and a semantic technology reasoning module for event detection. The detection system of the invention is simple in principle and targeted in functions, improves event detection accuracy and has important significance to the road traffic management efficiency improvement and traffic management effect improvement. Besides, the invention also discloses a traffic event detection system based on semantic technology and a method for the same.

The invention relates to a semantics based oil well indicator diagram measuring method. The semantics based oil well indicator diagram measuring method comprises the steps of firstly semantically packaging data information of an oil well, and then performing indicator diagram oil gauging based on the semantically packaged data. Due to introduction of a semantic technology, an indicator diagram oil gauging system provides a standard noumenal interface to a user, so that more users with limited computer knowledge can visit the indicator diagram oil gauging system, configuration becomes simpler and more convenient and operation becomes easier when the indicator diagram oil gauging system is used; on the other hand, as the mode of analyzing the indicator diagram is adopted to measure the yield of the oil well, the original oil gauging mode of performing manual routing inspection for the oil well and measuring with a need of stopping the well is avoided. Therefore, the well does not have to be stopped while the manpower resource is saved, and the oil production efficiency of the oil well is increased indirectly.

The invention relates to the technical field of voice semantics, and discloses a song training data processing method. The song training data processing method comprises the steps that initial sampledata are acquired, the initial sample data comprise the music score of each song and the singing record corresponding to the music score of each song, the initial sample data are marked, training dataare obtained, and an acoustic feature output model is obtained through training based on the training data. Afterwards, the method processes the target training data, takes the linguistic feature andthe music feature of each song in the target training data as input data of the trained acoustic feature output model, outputs the acoustic feature of each song, and performs pitch transfer on each song according to the acoustic feature of each song and the music feature of each song. The invention further provides a song training data processing device and a computer readable storage medium. According to the method and the device, the number of training samples is increased under the condition that extra corpora is not recorded.

The invention discloses an efficient data screening method based on semantic analysis and relates to the field of Internet retrieval technology. According to the method, weights in a semantic space are processed from the perspective of probability in combination with a semantic analysis technology so that the weights can better reflect the semantic relation between documents and between documents and vocabularies; and by the adoption of a secondary search strategy, a search result is improved so that the search result can better meet a user requirement, and the accuracy rate and the precision rate of retrieval are increased.

The invention provides an universal game-equilibrium semantic cache algorithm. Based on a Nash equilibrium mechanism, parameter expressions of three factors, namely semantics, time and location, are designed, analyzed and calculated, a game equilibrium point is found, a game-equilibrium equation is built, and the covariance coefficient of the relation role is calculated so as to obtain the universal game-equilibrium semantic cache algorithm; the semantic cache algorithm can be extended to the multifactor proportional change, also can be extended to any finite dimension, and can be widely applied to various mobile sensor database systems. How do limited mobile device resources deal with a large number of mobile user query services from a distributed database is an important issue. When a basic semantic cache technology is applied to query optimization of a sensor database, the existing method does not consider numerous reality multidimensional constraints, and therefore such the important issue is still a key and difficult issue in the semantic caching technology.

The invention relates to a voice semantic technology, and discloses a similar text matching method, which comprises the following steps of: performing feature word extraction on an obtained standard text, and constructing standard semantic representation according to an extraction result; generating a standard key value pair table according to standard feature words and the standard semantic representation; carrying out feature word extraction on the obtained target text and constructing target semantic representation; calculating the similarity between the target feature words and the standard feature words, and screening to-be-matched semantic representations according to the similarity; performing representation matching on the semantic representation to be matched and the standard semantic representation to obtain a matching probability; and determining the standard text corresponding to the standard semantic representation of which the matching probability is greater than a presetprobability threshold as a similar text of the target text. In addition, the invention also relates to a blockchain technology, and the standard text can be stored in a node of a blockchain. The invention further provides a similar text matching device, electronic equipment and a computer readable storage medium. According to the invention, the problem of low accuracy of similar text matching canbe solved.

Disclosed are methods and systems for a data discovery solution which harnesses the power of crowdsourcing to improve automated data curation. This is done in two complimentary ways: (a) large scale collective curation through anonymized multi-tenancy, and (b) and through internet scale matching and validation gaming platform using mobile application game. The result is the most extensive library of semantic-technical mappings of the enterprise data, which are immediately at hand to provide a fast, easy and a good understanding of the enterprise data. The data discovery solution forms a gateway for governing and unlocking value from big data.

The present invention relates to a road surface spectrum and GIS vector data fusion method and system based on a semantic technology. According to the method, firstly, a great quantity of ordered interpolating points are calculated by a Cardinal cubic spline curve interpolation algorithm; then points, of which the number is the same with sampling times of a road surface spectrum, are selected at an equal interval from an interpolated point sequence; altitude data of the road surface spectrum is processed by adopting a triangulation network generating algorithm to construct an irregular triangulation network and then GIS two-dimensional vector data is fused into the constructed triangulation network by an embedding algorithm; and next, semantic ontologies are established, according to extracted semantic rules and data fused on the bottom layer, by protege, inference rules are made by the logical relationship between the ontologies and data according with semantics is retrieved from metadata to be output on a query interface. The system consists of a data fusion module on the bottom layer, a fusion module based on the semantic technology, and a query interface module. According to the present invention, a user can know the real condition of a pavement and high-quality data services are provided for realizing intelligent transportation.

The embodiments of the present invention provide a computer-implemented method and system for translating user keywords into semantic queries based on domain vocabulary. The system receives the user keywords and search for the concepts. The concepts are transformed into a connected graph. The user keywords are translated into precise access paths based on the information relationship described in conceptual entity relationship models and then converts these paths into logic based queries. It bridges the semantic gap between user keywords and logic based structured queries. It enables users to interact with the semantic system by articulating the information in a structured query language. It improves the relevance of search results by incorporating semantic technology to drive the mechanics of the search solution.

The invention discloses a spoken language understanding method based on a knowledge graph and a semantic graph technology. The method comprises the following steps: 1, training a sequence to an actionsequence neural network model, and training a spoken language understanding neural network based on sentences and sentence logic expressions; 2, reading a natural language sentence needing to be analyzed; 3, scanning and replacing the part, matched with the semantic resources in the knowledge graph, in the sentence by using a text mapping algorithm; 4, reading the sentences with the semantic resources replaced through a sequence-to-action sequence neural network model, then executing corresponding action sequences to generate semantic graphs conforming to sentence semantics, and then using adepth-first algorithm for traversing the semantic graphs to obtain logical expressions of the sentences; and 5, reading sentences and logic expressions by using a spoken language understanding neuralnetwork, and generating intention information and slot position information. The invention provides a method for parsing spoken language by combining a knowledge graph and understanding sentence semantics.

The invention relates to a speech semantic technology, and discloses a text error correction method, which comprises the following steps of: carrying out character splitting on an acquired original text to obtain a character set; calculating character representation of each character in the character set by utilizing a representation algorithm; constructing a character vector of each character inthe character set according to the character set and the character representation; obtaining a standard character vector set, and respectively calculating the similarity between the character vector of each character in the character set and each standard character vector in the standard character vector set; performing screening out from the standard character vector set according to the similarity; and calculating a prior dependency probability of the similarity vector, and selecting the similarity vector according to the prior dependency probability to perform text error correction on the original text to obtain an error correction text. In addition, the invention also relates to a blockchain technology, and the standard character vector set can be stored in a node of a blockchain. Theinvention further provides a text error correction device and equipment and a medium. According to the invention, the problem of low accuracy of text error correction can be solved.

The invention relates to a voice semantic technology, and provides an intelligent dialogue generation method, an electronic device and a storage medium. The method comprises the following steps of collecting the dialogue contents of the service scene; classifying them, combining the conversation contents of at least more than two classification types to obtain a plurality of conversation behaviorchains, and training the plurality of dialogue behavior chains by utilizing a predetermined model algorithm to obtain a probability distribution value of each dialogue behavior chain, calculating cross entropies of the plurality of dialogue behavior chains according to the probability distribution values, screening out dialogue behavior chains of which the cross entropies are smaller than a threshold value, and storing the dialogue behavior chains in a database. According to the classification type of the user dialogue content, a set composed of all dialogue behavior chains matched with the classification type of the user dialogue content is retrieved from dialogue behavior chains of a database, the dialogue behavior chain with the highest similarity with the real-time dialogue content isscreened out from the set to serve as an answer, and the answer is output and fed back to the user. According to the invention, the input signal of the user can be responded quickly and accurately.

The invention discloses a heart failure detection method based on combined semantic technology and medical image segmentation. The method comprises the following steps: segmenting the inner membrane and the outer membrane of a left ventricle magnetic resonance image; calculating a segmentation result so as to convert image information into triple information; storing the triple information into a heart failure body established by protete; and then, automatically detecting the state of a heart by the powerful reasoning function of a semantic technology. According to the heart failure detection method based on the combined semantic technology and medical image segmentation, an input heart magnetic resonance image sequence is subjected to myocardial mass calculation and left ventricular ejection fraction calculation almost without manual intervention, the extraction precision of the inner membrane and the outer membrane of a left ventricle can be greatly improved, wrong segmentation brought by a fact that the extraction of an inner membrane profile and an outer membrane profile is affected by papillary muscles and an artifact can be reduced.

The invention relates to a voice semantic technology, and discloses an audio data emotion detection method which comprises the following steps of: obtaining original audio data, performing sound channel separation and data cutting processing on the original audio data to obtain standard audio data, performing mute detection on the standard audio data, when the standard audio data is detected to benon-mute data, performing frequency domain conversion on the standard audio data to obtain frequency domain data, converting the frequency domain data into Mel frequency domain data, detecting the Mel frequency domain data by using a pre-trained audio detection model to obtain an emotion initial score, and calculating the emotion initial score by using a preset weighted calculation method to obtain a final emotion score. In addition, the invention also relates to a blockchain technology, and the final emotion score can be stored in a node of a blockchain. The invention further provides an audio data emotion detection device, electronic equipment and a computer-readable storage medium. According to the invention, the problem of low audio data emotion detection accuracy can be solved.

A legal artificial intelligence interactive platform based on text analysis and semantic grabbing technology is compose of a text information receiving and analyzing module for recognizing problem input by user to extract keyword. A keyword accurate matching module, configured to establish a corresponding relationship between a keyword and a legal vocabulary, determine a legal vocabulary corresponding to the keyword, and generate and display an associative sentence containing the legal vocabulary according to the legal vocabulary; A multi-dimensional database matching module for analyzing a solution of generating a case; An intelligent interaction and push module is used for calculating data results and generating legal solutions according to user information and data variables input by the user. Compared with the prior art, the invention can realize intelligent and automatic solution finally by using word part-of-speech word meaning classification and combining with sentence pattern to analyze the central idea of the article. Achieving multiple aggregation of data to form potential new patterns and new data, using machine learning technology effectively; Implementation of multi-domain cross-analysis.