51 results about "Knowledge integration" patented technology

A framework for to enable collaborative research includes enabling scientific researchers to dynamically interact with others, collaboratively author, annotate, review, comment on others' data, and discuss their research. The framework allows for providing a dynamic collaboration environment by harnessing collective contributions from a group of researchers and facilitating broader participation of people into research projects. The framework is a Web-based collaborative platform to dynamically integrate information for scientific research. Besides data integration, the system provides essential collaboration capabilities to boost user participation and collaboration. The system will not only enable users as active information contributors to bring significant new values to the data, but also provide an environment for scientific researchers to do collaborative research in a large research community.

The invention relates to a template-based integrated design platform construction method, which belongs to the technical field of integrated design environments. In the construction method, centered on templates, various design rules, design experiences, design algorithms and the like are packaged into the templates to generate the single templates, and then the independent templates are combined into combined templates by a flow parser, thereby realizing knowledge reuse and quick design and provide simple means for the knowledge integration, flow integration, data integration, algorithm integration, specification integration, application integration and the like in an engineering and product design process to fulfill the aims of reducing the design cost and shortening the development cycle. A platform of the invention can be widely applied to the digital integrated design of various industrial products, effectively improve the design efficiency, reduce the design cost, reduce the workload of designers and shorten the design cycle, and has vast market prospect and high application value.

The invention discloses a speech content extracting method and a speech content extracting device based on a cloud platform. The speech content extracting method is characterized in that audio materials and video materials of a speech are acquired, and are cached in a PC for a pretreatment; the pretreated audio materials, the pretreated video materials, related data including speech slides, and related reading materials are transmitted to a server; the server is used for voice segmentation of the received audio materials, and is used for segmenting the audio materials according to speakers; automatic voice identification is used for converting the segmented audio materials into words, and adopts acoustic self-adaption and language model self-adaption; key words are extracted from texts after the voice identification, and content notes are generated. By adopting the speech content extracting method, the audio materials are converted into the text form, which can be read repeatedly, by the audio identification, and the language model self-adaption and the acoustic model self-adaption are used to improve the identification accuracy. Because of knowledge integration, time for reading redundant information is saved. The invention also discloses the speech content extracting device based on the cloud platform. The s speech content extracting device comprises a speech recording module, a material transmitting module, a voice segmenting module, a voice identifying module, and a key word and a content note extracting module.

The invention discloses a satellite plate and shell structure-based finite element automatic modeling method. For characteristics of a satellite plate and shell structure, a finite element automatic modeling system is developed by adopting PCL language programming through taking mainstream finite element modeling software MSC.Patran in the domestic aerospace field as a platform. The method comprises the following steps of (1) model importing; (2) model editing; (3) grid division; (4) material/unit attribute setting; (5) load/boundary/initial condition setting; and (6) submitting analysis and calculation. According to the method, knowledge integration of product structure design can be improved; the product structure analysis/optimization process can be standardized; the model research and development cycle can be shortened; and the model research and development cost can be reduced.

The invention relates to a template-based integrated design platform and a construction method thereof, and belongs to the technical field of integrated design environment. The integrated design platform of the invention comprises a database generation module, an asynchronous data conversion module, a specification integration module, a knowledge integration module, a flow analysis module, an application integration module, an algorithm integration module, a template generation module and a combined template generation module, wherein the databases generated by the database generation module comprise a product database, a specification base, a knowledge base and a flow base. When used by a user, the platform directly faces the service environment of the combined template; and only the input and output of the combined template need to be operated and various integration and operation behaviors of the intermediate modules do not need to be considered. The platform solves the problems of non-uniform design environment, obvious data switching barrier, nonstandard flow and specification, low knowledge reutilization efficiency and the like in the engineering and product design process, and provides a uniform integrated design environment for the digitalization design of the engineering and products.

The invention provides a biotechnology information knowledge graph construction device and method based on text mining. The device comprises: a data acquisition module which is used for collecting biotechnology information data from the Internet and taking the biotechnology information data as an input source of the device; a data extraction module which is used for extracting the collected biological semi-structured and non-structured information data to generate structured data; a data cleaning module which is used for carrying out preprocessing operation on the biotechnology information data to generate standard data; a knowledge fusion module which is used for carrying out knowledge integration on the extracted fragmented information units to form complete biotechnology information knowledge base data; and a knowledge graph maintenance module which is used for carrying out maintenance operation on the stored knowledge graph data.The beneficial effects of automatically and efficiently analyzing, mining and processing massive information and the like are achieved.

The invention discloses an intelligentized sewage treatment expert system, and belongs to the technical field of environmental engineering .The expert system comprises a data storage device and a display device; input data and a data analysis result can be displayed in the display device; the display device can display two first-stage modules including the operating management problem query module and the effluent quality standard exceeding reason reasoning module; the effluent quality standard exceeding reason reasoning module is provided with five second-stage modules including the effluent TN quality standard exceeding reason judging module, the effluent NH3-N quality standard exceeding reason judging module, the effluent COD quality standard exceeding reason judging module, the effluent SS quality standard exceeding reason judging module and the effluent TP quality standard exceeding reason judging module .According to the expert system, sewage treatment field expert-level experience and knowledge are integrated into the same system platform, a more convenient way capable of effectively solving the practical problems is supplied to technicians in the sewage treatment field, checking can be conducted in time, an effective solution can be presented in time, and the operating management level of a sewage treatment plant is increased.

The invention relates to a knowledge graph construction method fusing an inference engine, and belongs to the field of knowledge graphs. The method comprises the following steps: manually constructing a knowledge graph schema according to the actual condition of a service, accelerating the definition of entity attributes and relationships by utilizing an inference engine, importing semi-structured data to perform knowledge extraction, and cleaning and normalizing and disambiguating knowledge graph data; and then integrating rule design to carry out deep reasoning processing on knowledge to form a final rule knowledge base. According to the method, mass information is integrated and managed based on the structure data of the knowledge graph, rule reasoning is combined, the knowledge reasoning effect can be improved, and a more accurate reasoning result and visual display are obtained. According to the knowledge graph, multi-source heterogeneous data can be fused, the organization and relevance of the data are enhanced, and mass data can be conveniently stored and managed; wherein, an inference engine is introduced in the construction process of the knowledge graph to carry out rule design, so that deep inference and knowledge integration are effectively carried out on knowledge, the construction process of the knowledge graph is accelerated, and the data accuracy is improved.

The invention discloses an ultrasonic image sequence segmentation method based on sparse characteristics and shape correlation. The method comprises steps that firstly, a sparse expression theory is utilized to construct an over-complete characteristic dictionary based on a target and the background, reconstruction errors of the target relative to a target and background characteristic dictionary are utilized to construct an active contour search strategy based on sparse characteristic competition; secondly, for inputted to-be-segmented ultrasonic image sequences, non-monitoring learning of change of target shapes between the ultrasonic image sequences is carried out to acquire a matrix forming an image sequence shape, the matrix is consistent with low rank attributes, the low rank attributes are taken as prior knowledge of target shape change to constraint active contour evolution; and lastly, the active contour search strategy based on sparse characteristic competition and the target shape prior knowledge are integrated into an active contour segmentation framework. The method is advantaged in that the segmentation result is more accurate and robust, and computer auxiliary treatment efficiency and the effect are improved.

The invention discloses a speech abstract generation method and apparatus. The method comprises the steps of analyzing a content note and a speech slide to generate a speech outline; generating an extended speech outline by utilizing the speech outline and related reading materials; and analyzing the extended speech outline to obtain a speech keyword related to speech knowledge content, finding a corresponding interpretation entry of the speech keyword in a knowledge base, and generating a final abstract. According to the method, multi-aspect content such as the content note, the speech slide and extended reading are subjected to knowledge integration and extension to obtain the speech abstract, and accurate interpretation of the speech keyword is obtained by utilizing the external knowledge base; and the accurate and structured abstract of the speech content is obtained. The invention furthermore discloses the speech abstract generation apparatus based on the speech slide, the content note and the extended reading. The apparatus comprises a speech outline generation module, an extended speech outline generation module and a speech keyword entity linking module.

The invention discloses a tag recommendation method of an online question and answer platform based on knowledge graph and tag association, and relates to the technical field of artificial intelligence, natural language processing and recommendation systems. The invention aims to solve the problem that the existing tag recommendation method is not suitable for the scene of a question and answer platform, so that the recommendation effect is poor. The method comprises the following specific steps: inputting a question text and an external knowledge graph of an online question and answer platform into a trained KOCIN model to obtain a recommended tag; the KOCIN model comprises a knowledge integration layer, a sequence coding layer and an association capture layer; the knowledge integration layer is used for extracting a knowledge triple from the question text qi and an external knowledge graph, and integrating the knowledge triple into the question text qi to generate a Qtree; the sequence coding layer is used for converting the Qtree into dense vectorization representation of the Qtree so as to obtain a predicted problem text original label; and the association capture layer is used for obtaining a recommendation tag of the problem text according to the predicted original tag of the problem text. The method is used for obtaining the recommendation tag of the question-answering platform.

The embodiment of the invention discloses a neural network training method and device, equipment and a computer storage medium. The neural network training method comprises the steps of executing a loop process until soft labels meeting a preset number of anchoring samples are obtained; training a neural network at least based on the soft labels of the preset number of anchoring samples and the preset number of anchoring samples, wherein a current anchoring sample and at least one knowledge transfer sample are determined from the current training sample set in each period of executing the loop process; based on the neural network, determining the similarity between the current anchoring sample and each knowledge transfer sample, and the prediction probability of the current anchoring sample and the prediction probability of each knowledge transfer sample; and determining a soft label of the current anchoring sample based on the similarity between the current anchoring sample and each knowledge transfer sample, the prediction probability of the current anchoring sample and the prediction probability of each knowledge transfer sample. Knowledge integration under a self-distillation algorithm is realized.

The invention discloses a citrus knowledge modeling and large scale ontology generation method based on an agriculture eight-word policy. The method comprises the following steps: S1, based on the agriculture eight-word policy, knowledge in the citrus planting field is classified, wherein each word is corresponding to one type, and eight types are formed; S2, as for each type, concepts contained in the type are recognized, and the relationship among the concepts is built; S3, the eight types are used to form a citrus planting knowledge framework KF with the relationship among the concepts, wherein KF=(T, S), T is a citrus planting knowledge type, and S is a semantic relational layer; S4, the citrus planting knowledge framework is converted to eight sub ontologies; and S5, the eight sub ontologies are merged to form a large scale ontology. According to the method of integrating the citrus planting field knowledge based on the agriculture eight-word policy, the difficult problems of agriculture field knowledge integration and ontology collaborative development can be solved, and a complete scheme is provided for building and managing the large scale agriculture ontology.

The invention belongs to the technical field of software development, and particularly relates to an intelligent knowledge integration and retrieval system and method of a Call Center operation and maintenance service platform based on an ontology. The system comprises a document processor, a query conversion module, a proxy retrieval module, a customized processing module, a user interface, a body library and an information library. The user inputs a query statement through a user interface; The query conversion module searches a corresponding concept in the ontology library, performs query conversion on the query according to the obtained information according to the matched concept in the domain ontology, and then queries in an information base by the retrieval agent module; And the query result is processed by the customized processing module and is displayed by the user interface. According to the invention, more humanized services are provided for users; Operation and maintenanceare enabled to adapt to user requirements; Accurate integrated matching retrieval is realized, and the information of the knowledge base follows the trend day-to-day difference.

The embodiment of the invention provides a method for maintenance of field equipment in a factory, and the method comprises the steps: obtaining equipment data of the field equipment, wherein the equipment data represents a current event of the field equipment; generating recommended actions for events based on the equipment data by using a knowledge model of the field equipment, the knowledge model including possible event types for the field equipment and corresponding actions for each event type; and providing a recommended action to maintain the equipment device. Through the method, the knowledge integration level and transparency of the field equipment are improved, and accurate maintenance suggestions can be recommended to field personnel. On-site personnel do not need to spend energy and time in learning maintenance manual of different field equipment, and can quickly and effectively maintain the field equipment, so that the accuracy and high efficiency of maintenance are ensured, and the production efficiency of a factory is improved. In addition, along with continuous knowledge expansion and accumulation of a knowledge model, more intelligent and more accurate recommendation can be realized.

The invention relates to the technical field of knowledge retrieval, and discloses a digital library knowledge retrieval method, which comprises the following steps: processing data collected in a digital library sentence by sentence to obtain key sentences for revealing knowledge contents; converting the key statement into a knowledge unit according to a defined knowledge unit structure model; splitting each knowledge unit into a plurality of RDF triples, and creating an instance index according to each RDF triple; creating a class index according to the association relationship among all theknowledge units; obtaining user input information, retrieving a selected RDF triad related to the user input information in the instance index, and retrieving an associated RDF triad related to the selected RDF triad in the class index; and returning a retrieval result according to the selected RDF triad and the associated RDF triad. According to the invention, fine-grained retrieval can be realized, and the knowledge integration level is high.

The invention belongs to the technical field of target recognition, and particularly relates to a domain adaptive target recognition method based on deep knowledge integration. According to the method, feature-level and decision-level deep knowledge integration is realized. A public mapping matrix and a specific mapping matrix are designed at a feature level to realize knowledge integration, and the robustness of target recognition performance is improved; and the public mapping matrix fully excavates public knowledge of heterogeneous features, and the specific mapping matrix retains specific knowledge of different features. Feature weights are designed at the decision level to quantify the importance degrees of different features, meanwhile, the feature weights are updated through online learning by means of target domain samples, the data distribution difference of different fields is overcome, and field self-adaptive target recognition is achieved. Therefore, the domain adaptive target recognition method based on deep knowledge integration provided by the invention is an intelligent domain adaptive target recognition method.

The invention relates to the technical field of manufacturing scheduling, in particular to an integrated optimization system for ascheduling problem of a distributed zero-latency flow shop, which mainly comprises a workpiece initialization module, an inter-factory sequence optimization module and an in-factory sequence optimization module. In the workpiece initialization module, model knowledge and problem information are collected, a series of initial solutions is constructed as initial feasible solution groups of the distributed zero-latency flow shop scheduling optimization integrated system with the delivery window; the inter-factory sequence optimization module estimates the inter-factory sequence distribution by collecting the distribution rule of excellent individuals in an initial solution, and determines a rough potential search range through effective knowledge integration; and in the in-factory sequence optimization module, a variable neighborhood descent algorithm is adopted to carry out local adjustment on sorting in all factories, in the algorithm, knowledge in the problem is combined, three domain structures are provided to guide local adjustment, and adjustment blindness is reduced.

The invention discloses the technical field of target detection, and particularly relates to a data self-distillation method based on batch processing knowledge ensemble (BAKE), which comprises the following steps: S1, giving a small batch of samples and a network in training, and storing similar samples; s2, avoiding self-cognition enhancement, and normalizing the affinity matrix A; s3, predicting samples in one batch, and then obtaining and forming a soft learning target; s4, realizing complete integration of knowledge. Compared with an existing knowledge integration method, the problem that traditional knowledge integration needs to be processed through an additional network or branch is solved, the needed calculation amount and memory overhead are saved, BAKE is applied to a network architecture, the classification performance of the network architecture can be improved, and wrong supervision caused in the processing process is avoided.

The invention provides an internet important infrastructure knowledge graph construction method and device, and the method comprises the steps of constructing an internet important infrastructure ontology model, and obtaining each piece of structured and semi-structured data of an internet infrastructure based on the internet important infrastructure ontology model; based on a preset extraction rule, extracting knowledge from the structured and semi-structured data, and constructing a knowledge graph; using a Neo4j graph database for storing the knowledge graph, providing two import modes of full-database import and incremental import, and using the corresponding import modes according to different import scenes to form the Internet important infrastructure knowledge graph. According to the scheme, multi-source heterogeneous data and knowledge are integrated into a unified model for analysis and utilization; the network security situation is inquired and displayed based on a knowledge graph display mode.