70 results about "Semantic dictionary" patented technology

Whenever a document is going to be included into the textual database, a semantic binder is used to associate the document with one or more semantic nodes which are defined in a semantic taxonomy. When a search is performed, a search application looks through a semantic dictionary (which contains a table mapping queries to nodes on the semantic taxonomy) to see whether any corresponding semantic node can be found for the searchers query. If a match is found, the search application transforms the user's query into [“original query” OR “semantic node”] so that relevant documents, even they do not contain any user's keyword, can also be found in the database. The system binds semantic nodes arranged in a hierarchical structure of the taxonomy using a Log Analyzer which periodically looks through the system log for new queries and through textual indices for documents added to the database to generate the semantic dictionary and to bind the semantic nodes to the queries in the textual indices of the documents.

The invention relates to a multilanguage question and answer interface fast constituting method based on domain ontology and template logics. The method comprises the following steps of (1) multilanguage domain ontology structure building; (2) first-order template logic system building based on the domain ontology; (3) multilanguage question template structure design based on domain ontology and template logics; (4) domain-oriented question template base building; (5) user question preprocessing and question template matching; (6) user question semanteme obtaining and multilanguage intertranslation method. The multilanguage question and answer interface fast constituting method has the advantages that the question template semanteme can be more precisely expressed through the domain ontology and the template logics, the question template expression ability is improved by combining template operation characters, higher representativeness is realized, the template base scale can be reduced, in addition, a multilanguage body is similar to a multilanguage semantic dictionary, and the cross-language information inquiry can be easily realized. According to the method, various multilanguage man-machine interactive interfaces of field-oriented intelligent information retrieval and automatic question and answer systems can be fast constituted.

The invention discloses an intelligent service application platform and method for multi-source heterogeneous data fusion and relates to the field of data fusion application. According to the method, a visualization technology is utilized to realize collection of multi-source heterogeneous data and definition of data exchange; an automation technology is utilized to realize automatic real-time collection and cleaning of the multi-source heterogeneous data, and meanwhile the data exchange is completed in real time; and a semantic template library capable of being dynamically expanded is adopted to realize semantic mapping between multi-source heterogeneous data sources, a semantic dictionary is generated, semantic cleaning rules can be defined, it is guaranteed that text data can realize semantic automatic switching between the heterogeneous data sources, and intelligent cleaning and exchange of the text data are realized. Through the technical scheme, the blank of semantic automatic switching and exchange of unstructured multi-source heterogeneous data currently at home and abroad can be filled up, and the intelligent service application platform and method can be widely applied to domestic governments and enterprise and public institutions to solve the problems, existing in the governments and the enterprise and public institutions, about multi-source heterogeneous data switching and exchange.

The invention discloses an LDA (latent dirichlet allocation) and VSM (vector space model) based similar Chinese herb literature recommendation method. The method includes: adopting an IKAnalyzer to perform word segmentation on topics and summary information of literature on the basis of a terminological dictionary for Chinese herbs, constructing a vector space, performing dimensionality reduction on the vector space, constructing a semantic dictionary, numbering all lexical items in the dictionary in sequence, performing vectorization through each document on the basis of the semantic dictionary, constructing term vectors of each document, utilizing LDA and a Gibbs sampling algorithm to perform training to obtain probability distribution of each document on themes, then computing a value of similarity between every two documents by the aid of KL divergence, computing cosine similarity of the term vectors of each document on the basis of term frequency, performing joint weighting on the two kinds of similarities prior to performing similarity sorting, and then making recommendation. By the method, the literature, similar both in content and theme, in the Chinese herb literature can be recommended to users, and recommendation results are closer to user requirements.

The invention provides an information searching method for discovering sub-topics of a query statement and clustering the sub-topics. By adoption of the method, words of an original query statement and a historical query statement are divided respectively to obtain a query word sequence, and the similarity between the original query statement and the historical query statement is calculated. Moreover, the original query can be expanded through a semantic dictionary, the similarity between an expanded query statement and the historical query statement is calculated, and the similarity between the historical query statement and the original query statement is corrected; and the similarity between the historical query statement and the original query statement is further corrected according to click information of the historical query statement. Then, final sub-topics are selected according to a preset threshold value of the similarity, and are clustered, and a tree-shaped hierarchical structure is constructed for the sub-topics. A user acquires retrieving results with different classification granularities by selecting different leaf nodes of the tree-shaped hierarchical structure, so the information search method provides convenience for the user to browse the retrieving results according to topic categories.

The invention discloses a neural network and tag library-based statement similarity algorithm in the information retrieval field, which is characterized by comprising the following steps: (1) loading a semantic dictionary and a synonym lexicon with a neural network respectively; (2) inputting a complete statement to be analyzed; (3) analyzing the integral syntactic structure of the statement by using a dependency grammar analyzer, then layering the statement, and acquiring an effective component sequence of the statement; (4) determining a corresponding header field of the statement in an exUCL tag library according to the layering and the effective component sequence thereof; and (5) judging whether the statement has similar word pairs, if so, calculating the similarity of the statement, otherwise, re-inputting a new statement to be analyzed, and performing the similarity calculation again. The algorithm combines the advantages of dependency-based statement similarity algorithm and edit distance algorithm so that the calculation precision is greatly improved.

The invention provides an indoor robot navigation method based on environment characteristic detection, and aims to solve the problem that navigation semantic information is insufficient in positioning and is low in intelligence degree at present. The method comprises the following steps that: training an object detection model on a robot end, and establishing a semantic map which combines an indoor grid map with an object position; enabling a robot to rotate for one circle on the spot, utilizing the object detection model to identify an observed object, and utilizing a maximum likelihood estimation method to solve the global pose of the robot; and subscribing for the voice result topic of a remote control end by the robot end, carrying out mapping matching on an obtained voice result anda semantic dictionary in the semantic map, identifying a destination position to be navigated, and utilizing a center line in a corridor to plan the global path by the robot end. By use of the method,the problem that initial positioning can not be carried out in a traditional method is solved, the method is efficient and accurate in positioning, the planned path can more conform to the perceptionof the robot for an unknown environment so as to be safe, and realized semantic navigation can be more conveniently integrated in a robot product.

The invention relates to an illegal online commodity detection method, comprising the following steps: step (1) of taking information of a page on which to-be-detected commodities exist through a web crawler; step (2) of finding a least-depth node comprising a plurality of information blocks with similar structure in a webpage as a critical node through analyzing a DOM (document object model) tree structure of an electronic commerce website on which to-be-detected commodities exist, forming an associated information point to extract and establishing a template, and extracting commodity attribute information data from webpage information acquired by the crawler; step (3) of establishing a semantic dictionary, performing word segmentation to the extracted commodity attribute information through a method based on character matching; step (4) of manually establishing an illegal semantic library, recognizing and judging words in the illegal semantic library and content words field related to the commodity attribute information processed through word segmentation according to a function Illegal List, and judging an illegal class of the illegal online commodity according to a function return result. The method provided by the invention is characterized by simple calculation and good timeliness, and is suitable for changeful page modes.

The invention discloses an automatic image annotation and translation method based on decision tree learning. A new image is automatically annotated, and a text word list with a visualized content is translated by a machine so as to realize the machine retrieval of image data, comprising a training annotation image set and image automatic annotations, wherein the training annotation image set utilizes an image segmentation algorithm to segment a training image set into sub areas and extract low-level visual features of each sub area; the feature data is discretized, and then the training annotation image set is classified by a clustering algorithm based on a low-level feature discrete value to construct a semantic dictionary; the low-level feature discrete value is used as an input attribute of the decision tree learning; and self training learning is carried out on the constructed dictionary by a decision tree machine learning corresponding to preset semantic concepts so as to generate a decision tree and obtain a corresponding decision rule. The training annotation image set has expandability and robustness and can improve the recall ratio and the precision ratio of the retrieval when the training annotation image set is applied to semantic image retrievals.

The invention discloses a method and device for identifying text emotion types based on a cognitive structure model. The method comprises the steps of automatically constructing an emotion dimensionality dictionary for input massive open-source texts by using a statistical approach based on a general semantic dictionary and a syntax dependence relationship; carrying out refinement on the constructed emotion dimensionality dictionary, wherein the refinement specifically comprises the steps of carrying out inconsistency processing of semantics and emotional tendency and filtering non-emotional words; obtaining corresponding emotion types by combining the corresponding relation of the emotion dimensionality values and the emotion types in the emotion cognitive structure model based on the high-quality emotion dimensionality dictionary obtained after the refinement. According to the technical scheme, the method for identifying the text emotion types based on the cognitive structure model is obviously superior to an existing method on the design concept, the interpretability, the using flexibility and the validity, and can be used for emotion analysis and identification for texts in the fields such as business intelligence, social public sentiment and decision-making evaluation.

The invention provides a terms disambiguation method based on a semantic dictionary. In the method, term concept relevance is calculated to realize pretreatment on automatic text summarization, namely term disambiguation; the correlativity of concepts is used for realizing the term disambiguation method; requirements of the concept, concept paraphrase, synonym of the concept, expanded paraphrase of the concept, synset of expanded concept and other factors as well as sentence coherence are comprehensively considered; and a correlation degree calculation formula of the concept and a backtracking method are used for selecting the optimum meaning of a word, thus realizing the semantic disambiguation based on the context. An experiment proves that recall and accuracy of the semantic disambiguation can be improved and the method can be better used in acquisition of text summarization.

The invention provides a novel semantic ontology matching method, and in the method, the structure features and semantic features of ontologies are taken into overall consideration, and an ontology to be matched is measured from three aspects. The method comprises the following steps: firstly, carrying out preprocessing including operations such as word segmentation, classification or deletion and the like on an ontology 1 and an ontology 2; then, respectively calculating the element similarity, attribute based structure similarity and relation based structure similarity of the two ontologies by virtue of a semantic dictionary based semantic similarity algorithm; and finally, carrying out weighted averaging on the three similarities, and giving a matching result according to a preset threshold value. The method has stronger semantic properties, because in the method, the semantic features of concepts in the ontology and the semantic features in the structure of the ontology are taken into consideration, and a semantic dictionary based semantic similarity measuring method is used in the processes of calculating the three similarities, the semantic features of the ontology to be matched can be exploited well, therefore, the results of ontology matching implemented by using the method are extremely good.

The invention relates to text classification of natural language processing, and specifically relates to a class center vector text classification method based on dependency, word class and a semanticdictionary. To overcome the semantic defect of a feature selection algorithm based on statistics, the invention introduces the dependency, the semantic dictionary and the word class to optimize and cluster text features, provides an improved weight calculation formula, and further provides an improved class center vector text classification method. The text classification method of the inventionhas advantages of both high classification efficiency of a traditional class center vector method and high classification precision of a K-nearest neighbor algorithm, and can be widely used in variousclassification systems.

The invention relates to a semantic-computing-based interaction system and a semantic-computing-based interaction method for person with hearing or language disorder. The system is characterized by comprising two parts, namely portable terminals and a cloud-computing-based cloud server, and comprises at least two types of portable terminals, namely a first-type portable terminal and a second-type portable terminal or a portable terminal combining the two types of terminals. The portable terminals are used for performing related computing, data also can be uploaded to the cloud server for accurate computing, then a result is returned to the portable terminals, and different users can customize different semantic dictionaries. The system and the method have the advantages that words spoken by normal person are converted into the animation display of hand signs, so that the normal person can transmit information to the person with the hearing or language disorder without barriers; and the hand signs of the person with the hearing or language disorder can be converted into voices for play or into texts for display, so that interaction between the normal person and the person with the hearing or language disorder is further facilitated.

The invention relates to a semantic dictionary establishing method based on comment data. The method comprises the steps of 1) establishing a seed semantic dictionary according a small amount of comment data; 2) performing word segmentation on the comment data; 3) judging the semantic classes of the comment data word by word and replacing the comment data with semantic class tags; 4) generating templates according to the name of each semantic class and specific words contained in each semantic class; 5) applying the templates to the semantic class tag-replaced comment data to extract semantic words of each semantic class; 6) grading the templates according to the importance, the generalization performance and the accuracy of the templates; 7) selecting the templates with the highest scores, calculating the scores of the extracted semantic words of each template, and selecting the semantic words with the highest scores to expand the semantic dictionary; 8) performing iteration on the steps 3)-7) and acquiring a final semantic dictionary and template library after the step is over. The method enables a large-scale semantic dictionary to be obtained in a short time and enables multiple semantic classes to be extracted at the same time.

The invention provides a medical information search engine system and search method. The medical information search engine system comprises a medical semantic dictionary storage module used for storing synonyms, near-synonyms and associated words in the medical field, a semantic analysis module used for receiving a query instruction and generating a participle combination by analysis on basis of the medical semantic dictionary storage module, an index module used for generating participle indexes based on a database, storing the participle indexes on storage nodes in a distributed mode, and generating or extending data stored by a distributed type index data storage module, the distributed type index data storage module used for storing the participle indexes, and a search module used for searching the distributed type index data storage module for the participle combination. By arrangement of the medical semantic dictionary storage module, searched key words are extended accurately, and the search performance in the medical field is improved remarkably.

Whenever a document is going to be included into the textual database, a semantic binder is used to associate the document with one or more semantic nodes which are defined in a semantic taxonomy. When a search is performed, a search application looks through a semantic dictionary (which contains a table mapping queries to nodes on the semantic taxonomy) to see whether any corresponding semantic node can be found for the searchers query. If a match is found, the search application transforms the user's query into [“original query” OR “semantic node”] so that relevant documents, even they do not contain any user's keyword, can also be found in the database. The system binds semantic nodes arranged in a hierarchical structure of the taxonomy using a Log Analyzer which periodically looks through the system log for new queries and through textual indices for documents added to the database to generate the semantic dictionary and to bind the semantic nodes to the queries in the textual indices of the documents.

The invention provides a thesis recommendation method based on a multilayer drawing, which comprises the following steps: Step I, taking a target thesis and candidate theses as nodes in the drawing, constructing a thesis drawing according to reference relationships among the target thesis and the candidate theses and among the candidate theses; Step II, calculating weights of words in the theses by a word weight calculation method, removing the words with the weights less than a threshold, taking the rest words as the nodes, constructing a keyword feature map by virtue of semantic dictionary information; Step III, formulating a target equation by a similarity learning method based on the multilayer drawing; and Step V, optimizing a target function by a heuristic search algorithm of the map, updating weights of sides in the multilayer drawing, and recommending the thesis according to a final multilayer drawing result. The method allows a computer to be capable of recommending the target thesis according to the interested theses read or marked by a user.

The present invention discloses a video retrieval method, which is based on a noumenon and annotation technology, belonging to the video retrieval technology field. The prior video retrieval method has disadvantages such as nonstandard search words and low retrieval efficiency. According to deficiencies in prior art, the present invention proposes a video retrieval method, which is based on the noumenon and annotation technology. The present invention designs a noumenon-based semantic dictionary and uses standard field words to describe video contents and structures, solving the problem of nonstandard search words. The present method annotates lens and key frames, uses XML to display video data, defines a layered video meta-data structure, combines XML hierarchical features to demonstrate the hierarchical relations of video data and realizes a nonlinear organization of video data, thereby making video retrieval more efficient and convenient. The experiment result shows that compared with prior art, the present invention can obtain a higher retrieval accuracy and faster retrieval speed.

The invention discloses a semi-automatic knowledge map construction method, Most of the existing relationship extraction methods rely on the pre-determined relationship type system, This process is complex and time-consuming, the invention is based on dependency analysis, aiming at several Chinese sentence patterns, combined with a semantic dictionary, While exporting open relationships, semantictagging of words in relation, The semantics of unknown words are inferred based on statistics, and the semantic relation patterns on a large number of corpus are statistically clustered to form a relation type system. In this process, most of the links are carried out automatically, in which the results of semantic annotation and relation clustering of unknown words can be checked manually. Compared with the existing open relation extraction method, the invention optimizes and expands, and the extraction of the open relation and the formation of the semantic relation type complement each other, so as to improve the accuracy rate of the two.

The invention discloses a method for extracting text information through secondary semantic annotation. The method includes the following steps that first, a task semantic dictionary is established according to task requirements in text to be processed; second, semantic annotation is performed on the text to be processed for the first time through a prestored general semantic dictionary; third, semantic annotation is performed on the text which is subjected to the first semantic annotation for the second time through the task semantic dictionary; fourth, the text which is subjected to the secondary semantic annotation is analyzed and information of the text is extracted through sublanguage grammar based on semantics, and the information comprises concepts required by tasks and relations corresponding to the concepts. The method solves the problems that an existing large-scale semantic dictionary is low in coverage and meanwhile semantic marks cannot serve specific sublanguage grammar, and a better solution is provided for establishing information extraction with task adaptability in the Chinese environment.

The invention discloses a semantic dictionary construction method and device, a storage medium and an electronic device. The method comprises the following steps of: obtaining a first target entry, the grade of which is labelled, from a plurality of entries, and adding the first target entry and first-grade data corresponding to the first target entry into a first target semantic dictionary; training the first target entry and the first-grade data in the first target semantic dictionary to obtain a target classification model; classifying ambiguity degree of a second target entry, the grade ofwhich is not labelled, from the plurality of entries through the target classification model so as to obtain second-grade data; and adding the second target entry and second-grade data correspondingto the second target entry into the first target semantic dictionary so as to obtain a second target semantic dictionary. According to the method, the technical problem that the semantic dictionary construction efficiency in correlation technique is low is solved.

The invention discloses a voice interaction system and a voice interaction method. The voice interaction system comprises a preprocessing module, an acoustic model library, a language model library, an identification module and an interaction center, wherein the preprocessing module comprises a voice preprocessing module and an endpoint detection module; the acoustic model library comprises an HMMmodel matching unit, a TDNN model matching unit, an HMM model database and an ANN model database; the language model library comprises a N-Gram model database and a Rule-based model database; the identification module comprises an MFCC parameter feature extraction unit and an identification control unit; and the interaction center comprises a semantic understanding module, an interaction processing module, a response information library and a semantic dictionary database. The voice interaction system and the voice interaction method utilize a feedback module to monitor identification information and a feedback instruction of a client, demonstrate the identification information to the client by means of the identification control unit, control and change a voice matching model and the language model library, and optimize the interactive identification accuracy rate of the voice interaction system.

The invention relates to the field of natural language processing, in particular to a Chinese semantic relation recognition method and device. The problem that the Chinese semantic relation recognition accuracy is low is solved. In order to solve the problem, the Chinese semantic relation recognition method comprises the following steps of 1, judging whether or not a preset semantic dictionary contains Chinese word pairs to be detected, wherein if yes, the semantic relation of the Chinese word pairs to be detected is determined according to the preset sematic dictionary, and if not, the step 2is executed; 2, utilizing a word vector of the Chinese word pair to be detected to obtain the first semantic relation; 3, utilizing word structure features of the Chinese word pair to be detected, obtaining a second semantic relation, and according to the second semantic relation, adjusting the first semantic relation and obtaining the final semantic relation. According to the Chinese semantic relation recognition method and device, the words can be considered from multiple dimensions, and the Chinese word semantic relation can be efficiently, rapidly and accurately recognized.

The invention discloses a judicial case searching method based on multi-feature fusion. The specific steps are as follows: a user inputs a query request; the user query request is preprocessed and segmented, and stopped words are removed to obtain a set of query keywords; the query word set is traversed in turn, each query word in the query word set is subject to query semantic extension by a semantic dictionary to obtain an expanded query semantic keyword list; information points are used to filter documents, a feature inverted index is searched, different feature vectors of the keyword listare obtained, and then multi-feature fusion is carried out; the fusion similarity value between documents and query sentences is obtained, and the final similarity score is obtained; and the final search result is sorted and output. The invention has the advantage of high accuracy.

The invention discloses a dialogue type intelligent teaching system. The system comprises a knowledge base unit and a functional unit which are mutually connected, the functional unit comprises an input preprocessing module, an answer reasoning unit, an evaluation module and a dialogue management module, and the input preprocessing module, the answer reasoning unit, the evaluation module and the dialogue management module are mutually connected. The knowledge base unit comprises a filed body, an interactive module, a semantic dictionary and a student module. According to the dialogue type intelligent teaching system, the field body which is good in semantic relation and hierarchical structure is utilized to model field knowledge of the system, and knowledge based on reasoning can be provided for the system through the conceptual hierarchical structure and semantic relation. Meanwhile, a body-driven dialogue management mechanism is provided, and a dialogue content and a dialogue sequence can be quickly formed.

The invention discloses a verb meaning disambiguation method and device based on dependency constraint and knowledge. The method comprises the steps that dependency parsing is carried out on large-scale linguistic data, obtained dependency tuples are obtained, the frequency of the dependency tuples is calculated, and a dependency knowledge base is built; dependency parsing is carried out on a sentence with an ambiguous verb, and seventeen kinds of dependency tuples meeting set conditions are provided to be used as a dependency constraint set for ambiguous verbs; according to a semantic dictionary, a synonym set, an antonym set and a hypernym set are sequentially extracted as verb meaning representative verb sets of corresponding verb meanings of the ambiguous verb; according to the dependency knowledge base and the verb meaning representative verb sets, the posterior probabilities of the verb meanings of the ambiguous verb in the dependency constraint set are sequentially calculated; the correct verb meaning of the ambiguous verb is selected according to the posterior probabilities. By means of the verb meaning disambiguation method and device, the role of dependency parsing brought into full play, and the verb meanings of the ambiguous verb is more accurately judged.

The invention discloses a method and a system for carrying out semantic description on audio and video contents. The method comprises the following steps of: firstly, splitting the audio and video contents into multiple fragments, endowing a structure attribute used for marking sequence and nest relation of each fragment for each fragment, and generating an XML (Extensive Makeup Language) file provided with the sequence and the nest relation; secondly, respectively carrying out semantic description on each fragment in the XML file according to a structural dictionary and a semantics dictionary to form a new XML file; and at last, adding an XML Schema announcement and a copyright announcement to the new XML file, adding into original audio and video files to generat an audio and video file comprising the XML file, or simultaneously adding a position of the fragment of the corresponding audio and video file into the new XML file to generatthe XML file corresponding to the original audio and video files. According to the method and system provided by the invention, the precise search of audio frequency and video frequency based on the audio and video files or the XML file corresponding to the original audio and video files can be realized.

The invention provides a method and a system for carrying out semantic description on content of an electronic-book (e-book). The method comprises the following steps: firstly generating an XML (Extensible Markup Language) Schema file comprising a structure dictionary and a semantic dictionary according to structural and semantic description rules and the assignment range in the e-book by adopting XML; dividing the content of the e-book into a plurality of fragments according to the structure dictionary; endowing each fragment with structural attributes for marking out sequence and embedded relationship of the fragments, and generating an XML file containing the sequence and the embedded relationship; respectively carrying out semantic description on each fragment in the XML file according to the semantic dictionary so as to form a new XML file; and finally adding XML Schema statement and copyright statement to the new XML file so as to form an e-book file.

The invention discloses an automatic professional knowledge graph construction method. The method comprises the following steps of obtaining a professional knowledge text; performing word segmentationon the professional knowledge text and removing stop words in the word segmentation text; converting each text into a plurality of word sets; obtaining a part-of-speech label through part-of-speech labeling, and obtaining a dependency label and a dependency tree through dependency analysis; performing noun phrase detection, verb phrase detection and candidate relationship detection based on the part-of-speech labels and the dependency labels; performing semantic annotation on words in the candidate relationship to obtain a candidate semantic relationship mode; clustering the obtained candidate semantic relationship modes to obtain a group of final semantic relationship modes; obtaining professional knowledge data by utilizing the semantic dictionary and the semantic relationship mode; andwriting the data into an owl file and importing the owl file into a protege. According to the method, optimization and expansion are carried out, the professional knowledge graph construction efficiency is high, the cost is low, time and labor are saved, and the constructed professional knowledge graph is high in accuracy.