41 results about "Semantic integrity" patented technology

Methods and systems are provided which enable users to map and transform data between different data schemas while preserving the semantic integrity of the data. In accordance with the present invention, computer implemented methods and systems are provided for separation and mapping of representational and linguistic semantics. By representing the semantic portion of the schema using a formal ontology, these methods and systems allow data to be contextually transferred between different data schemas.

The invention discloses a translation text integrity evaluation method based on bilingual text structure information. The method includes the following steps of firstly, extracting training corpus used for establishing a target end text unit integrity model from a target end texture structure tree; secondly, establishing the target end text unit integrity model through the training corpus generated in the first step; thirdly, integrating the target end text unit integrity model in a logarithm linear translation model, and conducting translation generation through an adaptability decoding method. According to the method, with the help of the bilingual text structure information, a decoder can further improve the translation quality of the current statistical machine translation by fully and approximately utilizing the semantic integrity information based on bilingual text language knowledge.

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 remote sensing image semantic segmentation method based on deep learning, and belongs to the technical field of machine vision. Aiming at the problems of difficulty in obtaining features of small objects and insufficient segmentation precision of a semantic segmentation method of a mainstream deep convolutional neural network, the method comprises the following steps: improving a Deeplabv3 algorithm, improving a single up-sampling layer, and performing multi-layer up-sampling by utilizing residual errors obtained in a backbone network to ensure semantic integrity of an image in resolution; and meanwhile, modifying the expansion rate of four expansion convolution layers in the ASPP layer, so that the network has a better effect on small object segmentation. The result shows that the mIou and pixel accuracy of the improved Deeplabv3 semantic segmentation algorithm on a self-made data set reaches 94.92% and 98.01% respectively, which are improved by 3.77% and 2.40% respectively compared with the original algorithm, so that the improved Deeplabv3 semantic segmentation algorithm not only has higher accuracy, but also has better robustness for segmentation of various terrains; the method is suitable for a complex urban remote sensing image environment, and can be well applied to the fields of urban planning, agricultural planning, military war and the like.

The invention discloses a document theme partitioning method based on a domain knowledge map community structure, and the partitioning problem of document resources related to subject knowledge or document knowledge is mainly solved, so that documents related to a theme can be stored in a close logical place, and learning efficiency is improved. The document theme partitioning method is characterized in that a level community discovery algorithm based on the Fast Geedy algorithm and the GN algorithm is proposed, and a theme structure tree is built; in the process of feature extraction, knowledge units directly serve as feature vectors, and due to the fact that the knowledge units have semantic integrality, compared with a traditional method based on participles, the document theme partitioning method can reflect theme characteristics of the feature vectors better; in the process of calculating feature vector values, the method of combination of degree centrality and knowledge unit file frequency is proposed, wherein the concept of the degree centrality reflects the status of the knowledge units in a knowledge map whole situation. Through the method, accuracy of document theme partitioning is effectively improved, and the method is suitable for the document theme partitioning based on the domain knowledge map community structure in general scenes.

A method and apparatus are provided a method and apparatus for reducing a program that preserves branching time properties, including existential and universal aspects. An alternating transition system (ATS) is abstracted, formed by a product of a program, M, with an alternating tree automaton, A, for a property, f. The disclosed program abstraction method generates the abstract program and an altered version of the branching time property, f. An automated program check, such as a model check, is performed on the abstract program for the altered branching time property. The invention provides semantic completeness: i.e., whenever a program satisfies a property, this can be shown using a finite-state abstract ATS produced by the method. Choice predicates can be employed to help resolve nondeterminism at OR states, and rank functions can be employed to help preserve progress properties.

The invention provides an automatic quotation extraction method and device. Characters or character strings, serving as reading focuses, in a text can be used in the method and device to serve as centers to automatically extract contexts, the length of an extracted quotation is within the preserved length scope, it is kept that the extracted quotation has the semantic integrity, a section of semantic scene which is appropriate in length and integral in meaning and enables the selected characters or the selected character strings to serve as the reading focuses can be extracted from the text and formed, and a user can conveniently restore the correct meaning of the reading focuses in the contexts.

The invention discloses a voice processing method and device and equipment. The method comprises the steps that voice data of a to-be-processed voice are received, and the to-be-processed voice is thevoice transmitted by a first object in a several-round voice interaction process of the first object and a second object; according to the voice data, a confidence coefficient that the to-be-processed voice is a complete voice is determined by utilizing a semantic integrity model; and according to the confidence coefficient, mute waiting time for responding of the second object to the to-be-processed voice is determined. According to the method, based on the received voice data of the to-be-processed voice, the confidence coefficient that the to-be-processed voice is the complete voice is determined by utilizing the semantic integrity model, semantic incomplete voice information is effectively recognized, so that the mute waiting time for responding of the second object to the to-be-processed voice is dynamically adjusted according to the confidence coefficient, the situation that a user is interrupted when not finishing voice expression is avoided, and under the situation that semanteme of the voice is determined to be complete, mute duration is shortened, the interaction efficiency is improved, and the user experience is greatly improved.

The invention relates to the technical field of artificial intelligence, in particular to an automatic question answering method and a device thereof, comprising the following steps of: constructing asemantic vector of an input question; Judging the semantic integrity of the input problem according to the semantic vector; If the semantics of the input problem is incomplete, predicting an omittedphrase in the input problem from the above corpus corresponding to the input problem by a semantic supplementary model; Obtaining reply information of the input question from a question answering database according to the semantically incomplete input question and the obtained omitted phrase. The method of depth learning is used to predict the omitted phrases according to the incomplete semanticsof the input question, and the response information of the input question can be obtained according to the predicted omitted phrases combined with the input question, so that the flexibility of automatic question answering can be improved and the user experience can be improved.

The invention belongs to the technical field of computer software, and discloses a method and system for automatically extracting keywords from multiple documents and a computer program. The method comprises the steps of extracting keyword seeds; utilizing statistical information of the mutual positions of words in text to measure whether or not a combined word is semantically integrated; if yes,determining the combined word to be a key combined word; if not, conducting left and right expansion. By means of the extraction method of discovering semantic integrity and cutting off reasonable keyphrases, the readability of a result obtained by automatically extracting the keywords from multiple documents is greatly improved; the extracted keywords have a larger average length, and the themesof multiple documents can be better described through more abundant and integrated semantics; the keywords extracted by a Tf-Idf algorithm are more detailed, the semantics is broader and not specific, and the themes of the documents cannot be represented; the practicality is ensured, labor cost for annotation is saved, and the automatic keyword extraction can be conducted without any annotated corpora.

The invention discloses a multi-modal semantic integrity recognition method and device and electronic equipment, and the method comprises the steps: building a multi-modal semantic integrity model, respectively inputting the text data, audio data and dialogue data of a current user into the multi-modal semantic integrity model, and obtaining a semantic integrity recognition result; and adjusting the user semantics of the current dialogue according to the semantic integrity identification result. The multi-modal semantic integrity model comprises a first model used for respectively extracting audio vector features and text vector features in the audio data and the text data, and a second model used for identifying semantic integrity in combination with the audio vector features, the text vector features and the dialogue features. The integrity of semantics can be recognized according to three modes of the text data, the audio data and the dialogue data, so that the accuracy of model intention recognition is effectively improved, and the voice interaction effect of the voice robot and a user is improved.

The invention discloses a surveying and mapping result directory automatic checking and warehousing release method based on semantic integrity. The method comprises the following steps of creating a metadata semantic integrity rule model which comprises an achievement type, an achievement field and a field specification value and conforms to a standard specification; inputting the surveying and mapping result catalog data to be checked, and sequentially carrying out field integrity checking, field normative value checking and metadata repetitive checking to realize the automatic checking of the catalog data, and then outputting a checking report, finally storing the checked surveying and mapping result catalog data into a database, and uniformly updating the database and releasing the result to the outside. The method has the remarkable effects that by establishing the metadata semantic integrity rule model, the metadata of a to-be-detected surveying and mapping result directory is automatically checked and warehoused, and the checking report can be outputted without manual intervention, so that the workload can be greatly reduced, the user cost is saved, the efficiency is improved, the expandability is strong, and the application range is wide.

The embodiment of the invention provides a voice breakpoint detection method, device and equipment based on artificial intelligence, and the method comprises the steps: the semantic integrity detection of candidate results with the probability being higher than a preset threshold value is carried out through a pre-trained semantic integrity model, after it is determined that the candidate resultswith the probabilities higher than the preset threshold value are complete in semantics, natural language understanding is conducted on the candidate results with the probabilities higher than the preset threshold value, and intentions corresponding to the candidate results with the probabilities higher than the preset threshold value are obtained. And finally, a response corresponding to the query statement is obtained according to the candidate result of which the probability is higher than a predetermined threshold and the corresponding intention. Therefore, whether the user speaks or not can be dynamically judged according to the semantic integrity, the real intention of the user can be recognized more accurately, whether the user speaks or not can be accurately judged under the scenesof repeated speaking, sound dragging and the like of the user, and then the user experience can be improved.

The invention discloses a speech recognition method and device, electronic equipment and a storage medium, and relates to the technical field of deep learning and the technical field of speech in thetechnical field of artificial intelligence; and the method comprises the steps: obtaining state information and context information of an application corresponding to target speech information in response to the obtained target speech information; calculating semantic integrity of the target voice information according to the state information and the context information; determining a monitoringduration corresponding to the semantic integrity, and monitoring the speech information in the monitoring duration; and if the speech information is not monitored within the monitoring duration, performing speech recognition according to the target speech information. Therefore, the semantic integrity of the acquired speech information is determined according to the multi-dimensional parameters, the duration of detecting the speech information is flexibly adjusted according to the semantic integrity, the speech information is prevented from being cut off, and the speech recognition accuracy isimproved.

The invention provides a VAD tail point detection method and intelligent equipment. The method comprises the following steps: determining a probability pa that a speech packet contains a VAD tail point, decoding the speech packet, determining the probability pn of semantic integrity of a text instruction corresponding to the speech packet, and determining whether the speech packet contains the VADtail point or not according to pa, pn and a preset threshold value, and combining acoustic VAD detection and semantic VAD detection to realize VAD tail point detection. The VAD tail point detection is more accurate, false detection caused by speech pause can be avoided, the problem of slow response caused by overlong VAD tail point in the speech packet is solved, and the user experience in the speech man-machine interaction process is improved. The invention further provides a server and a computer readable medium.

The invention provides a speech VAD tail point determination method. The method comprises the following steps: receiving speech information of a user, dividing the speech information into data packets, and uploading the data packets to a server according to a time sequence; when it is judged that the current data packet is a mute packet, calculating the current mute duration t; and triggering theserver to detect the semantic integrity of the speech information according to the current mute duration t and a preset first threshold T1 to make the server determine a tail point of the speech information according to a semantic integrity detection result. The tail point of the speech VAD is not cut off by the intelligent equipment any more, the intelligent equipment uploads the data packet divided by the speech information to the server according to the time sequence, and the server is triggered to detect the semantic integrity of the speech information, so that the server determines the tail point of the speech information according to the semantic integrity detection result. The mute duration of the tail point of the speech information is changed into a dynamically adjustable value from the original fixed duration.

The invention discloses a Chinese word segmentation method and system based on word vector representation learning. The method comprises the steps of performing preliminary word segmentation processing on a text to be subjected to word segmentation to obtain a preliminary word segmentation text; inputting the preliminary word segmentation text into a BERT model for training to obtain corpus word vectors; inputting the corpus word vectors into a Bi-GRU model for training to obtain a plurality of feature word vectors; calculating cosine similarity of two adjacent feature word vectors to obtain acosine similarity value; judging whether the cosine similarity value is greater than or equal to a preset threshold value or not; and if the cosine similarity value is greater than or equal to the preset threshold, combining the preliminary word segmentation results of the words corresponding to the two adjacent feature word vectors. By adopting the method and the system provided by the invention, the problem of word segmentation of the unregistered words in the specific field is solved, the word segmentation performance is improved, the semantic integrity of the words is not damaged by the word segmentation result, and a large amount of manpower tagging corpora can be avoided.

The invention discloses a viewpoint tracking method based on geometrical reconstruction and semantic integration. The viewpoint tracking method is characterized by comprising the steps of 1, establishing the visual characteristic set of subgraphs, 2, integrating semantic characteristics, 3, active learning algorithm, 4, optimization updating, and 5, ranking and orderly linking the subgraphs to obtain a viewpoint tracking path. The viewpoint tracking method is capable of quickly and accurately detecting various remarkable regions of images, improving the accuracy of viewpoint tracking and then improving the prediction capacity of a viewpoint transfer route.

The invention relates to a semantic integrity determination method and device, electronic equipment and a storage medium. The method comprises the following steps: predicting the first to-be-processedtext through the text prediction model, determining a target probability table corresponding to the target element position, wherein the target probability table is used for representing the prediction probability that each text element in a plurality of text elements contained in a preset element set appears at the position of the target element; and determining a prediction probability corresponding to each target text element in the element set according to the target probability table so as to determine whether semantics of the first to-be-processed text are complete or not according to the prediction probability corresponding to each target text element, the target text elements being used for representing semantic integrity of the first to-be-processed text. According to the methodand the device, whether the semantics of the first to-be-processed text is complete or not can be accurately determined, and the situation that the man-machine conversation system acquires the text with incomplete semantics due to the fact that input of the text is judged by mistake is avoided, so that the accuracy and efficiency of man-machine conversation are improved.

The invention discloses an XML keyword query method, belongs to the field of information retrieval, and discloses BLCEA query semantics and a solving algorithm thereof and a method for semantic return result ranking. According to the BLCEA query semantics, after entity offspring containing all query keywords are omitted on the basis of classification of nodes in an XML document, an entity node still meeting query conditions is defined as a meaningful BLCEA semantic entity, and the recall ratio and semantic integrity of a query result are guaranteed. The method for return result ranking is designed by combining the matching degree and compactness of the keywords in a sub-tree with the result node as the root, time complexity of XML keyword query is lowered, and the precision ratio of the keywords is guaranteed when fuzziness exists.

The invention discloses a voice processing method and related equipment thereof, and the method comprises the steps: determining the semantic integrity representation information of a user according to a current voice segment after the current voice segment is obtained; according to the semantic integrity representation information of the user, the waiting duration of the voice to be used is determined, when it is determined that the next voice segment is not obtained within the waiting duration of the voice to be used after the voice ending moment of the current voice segment, it is determined that the user ends speaking, and a user voice text used for representing the speaking content of the user is obtained; therefore, corresponding response operation can be performed on the user based on the voice text of the user subsequently, so that the reception waiting time length can be dynamically adjusted based on the complete semantics of the content spoken by the user, the defects existing in the man-machine interaction process of performing reception control based on the fixed reception waiting time length can be overcome, and the user experience is improved. And the man-machine interaction effect can be effectively improved.

The invention relates to a model training method and device and a semantic integrity recognition method and device. The method comprises the steps of obtaining a to-be-identified statement; obtaininga semantic complete sample; obtaining a semantic incomplete sample according to the semantic complete sample; obtaining a model training sample of an integrity recognition model according to the semantic complete sample and the semantic incomplete sample; training the integrity recognition model according to the model training sample to obtain a trained integrity recognition model; wherein the trained integrity recognition model is used for performing semantic integrity recognition on a to-be-recognized statement. The training sample contains both the semantic complete sample and the semanticincomplete sample, and the model obtained by training can recognize semantic integrity by combining semantic complete features and semantic incomplete features, so that the semantic integrity recognition accuracy can be improved.

The invention provides an event name generation method and device, an electronic device and a storage medium. The method comprises: obtaining the event type and at least one event attribute of the to-be-processed text by carrying out event classification on the to-be-processed text; extracting an event attribute value of each event attribute from the to-be-processed text based on an event map technology; filling each event attribute value into an event name template corresponding to the event type, to obtain the event names of the to-be-processed text according to the event names of the to-be-processed text, so that the event names are automatically generated. As different event attributes and event name templates are adopted for the to-be-processed text of different event types, the language fluency and semantic integrity of the generated event names of the to-be-processed text are higher, and the actual use requirements are met.

The invention is applicable to the field of computer input methods and provides a method for constructing an input method word bank according to a Chinese language model. The method is composed of a Chinese language model module and a word creation module, wherein the Chinese language model module provides word formation information and word bank management information for the word creation module, the word creation module generates input method terms in batch by aid of database management software according to the word formation information provided by the Chinese language model, and addition / deletion, retrieval, ordering and other operation can be conveniently performed on the constructed word bank by use of the word bank management information provided by the Chinese language model module. The Chinese language model is customized according to the requirements of phonetic pause and semantic integrity in language communication and fully reflects characteristics of the Chinese language. Through the method, the key problems that traditional input method word banks are short of scientific word collection standards and low in word collection efficiency, word bank content is not comprehensive and not systematic and cannot be effectively managed, a user has no way of grasping the word bank content in an input process, the input speed is low, and language communication experience is lacked are effectively solved.

The invention designs a distributed keyword query method based on an RDF graph, and belongs to the field of information retrieval. The method comprises the steps of firstly, converting the RDF data graph into an RDF sentence graph; secondly, by utilizing a conditional depth-first algorithm and a simulated annealing algorithm, segmenting the RDF sentence graph according to two most basic principlesincluding minimum of an edge cut set and data balance between sub-graphs after segmentation; and finally, refining the segmented RDF sentence graph into the RDF data graph, obtaining a vertex cut setof the RDF graph, and by utilizing a reverse search algorithm and a Hadoop distributed computing framework, realizing efficient and fast query of keywords. Under the condition of guaranteeing the atomicity and semantic integrity of RDF data, the limitation of a traditional algorithm on the segmentation efficiency of a large-scale data set is effectively eliminated, and the query efficiency of thekeywords is greatly improved.

The invention discloses a value mapping and row right method for digital rights and interests. According to the invention, an operation pool data structure T and two operations acting on an operationpool are designed: G operation and E operation; the G operation comprises that the elements of T are increased or decreased in the same direction and certificates s are increased or decreased correspondingly; the E operation comprises that the elements of T are increased or decreased reversely and certificates corresponding to the elements are locked or released correspondingly; increase and decrease of the certificates s can be used for mapping asset attribute changes of rights and interests, and increase and decrease of the elements in T can be used for mapping market value changes of the rights and interests; the G operation and the E operation are independent and semantic self-consistent, and the operation pool T maintains semantic self-consistency before and after each operation, so that the semantic integrity can be maintained when the method runs under a decentralization condition, and the row right efficiency is improved.

The invention discloses an image restoration method and system based on a progressive learning strategy, a storage medium and terminal device, and the method comprises the steps: inputting a to-be-restored image into a rough generation model, so as to obtain a rough generation result; extracting a feature map of the rough generation result to obtain a foreground feature map; and matching the foreground feature map with a background feature map, and reconstructing the foreground feature map by using the obtained background sample block with the highest matching degree with the foreground feature map. According to the method, when the large-area damaged area in the image is generated, the semantic integrity of the image is kept, clear details are generated, a good restoration effect is achieved on the image with the high structure, and therefore the image restoration experience of a user is improved.

A precise image semantic segmentation and optimization method using an interaction means comprises the steps that interaction is achieved through marking and delimiting points input by a user, then semantic features are extracted through a deep convolutional network to achieve preliminary segmentation of a to-be-segmented image, and then a conditional random field model of the to-be-segmented image is constructed, the preliminary segmentation result is optimized by using the underlying features of the to-be-segmented image, and finally the accurate segmentation result is further optimized through an interaction means. The delimiting point is used as interactive input, the operation is simple and convenient, and the segmentation effect is improved; image semantics are extracted by using a full convolutional neural network, and a segmentation result has semantic integrity; the user is allowed to guide modification through manual intervention, and then image bottom layer features are used to assist in result optimization.

The invention discloses a document theme partitioning method based on a domain knowledge map community structure, and the partitioning problem of document resources related to subject knowledge or document knowledge is mainly solved, so that documents related to a theme can be stored in a close logical place, and learning efficiency is improved. The document theme partitioning method is characterized in that a level community discovery algorithm based on the Fast Geedy algorithm and the GN algorithm is proposed, and a theme structure tree is built; in the process of feature extraction, knowledge units directly serve as feature vectors, and due to the fact that the knowledge units have semantic integrality, compared with a traditional method based on participles, the document theme partitioning method can reflect theme characteristics of the feature vectors better; in the process of calculating feature vector values, the method of combination of degree centrality and knowledge unit file frequency is proposed, wherein the concept of the degree centrality reflects the status of the knowledge units in a knowledge map whole situation. Through the method, accuracy of document theme partitioning is effectively improved, and the method is suitable for the document theme partitioning based on the domain knowledge map community structure in general scenes.

The invention relates to a method for ensuring semantic integrity for the isolation and control of TCP connection and a system thereof, which pertains to the field of information safety. After TCP processing step after the modification of the method of the invention ensures that an SYN message is received, the information of the characteristics of the message is notified to another network processing unit; the SYN message is regenerated and sent to a network by another network processing unit; after another network processing unit receives an ACK, a protocol stack generates a corresponding message. The system of the invention comprises a TCP handshake processing module, a TCP connecting and checking module, an application-layer data checking module, an application-layer protocol processing module and a safe data switching channel module. The method and system of the invention can not only isolate the TCP connection between an internal network and an external network and carry out application-layer data switching, but also can ensure the semantic integrity of the TCP and the situation that a server is closed, but a client still can establish the TCP connection cannot occur.