58 results about "Infosphere" patented technology

Method and system are provided for generating a natural language processing model for an information domain. The method includes: deriving a skeleton of a natural language lexicon from a source model of the information domain; applying a set of syntactical rules defining concepts and relationships; and expanding the skeleton of the natural language lexicon based on reference documents from the information domain to provide a natural language processing model for the information domain, wherein expanding the skeleton includes clustering and scoring terms for concepts and relationships.

The present invention is a system, method for accessing domains of information to identify heretofore unknown relationships between disparate sources of data to seek and obtain knowledge, the invention includes a source of data with one or more domains of information, an Object-Relationship Database for integrating objects from one or more domains of information and a knowledge discovery engine where relationships between two or more objects are identified, retrieved, grouped, ranked, filtered and numerically evaluated.

A clustering method yields a searchable hierarchy to speed retrieval, and can function dynamically with a changing document population. Nodes of the hierarchy climb up and down the emerging hierarchy based on locally sensed information. Like previous ant clustering algorithms, the inventive process is dynamic, decentralized, and anytime. Unlike them, it yields a hierarchical structure. For simplicity, and reflecting our initial application in the domain of textual information, the items being clustered are documents, but the principles may be applied to any collection of data items.

The invention is applicable to the technical field of data processing, and provides an information pushing method and equipment, and the method comprises the steps: obtaining an information query record of a target object; calculating the similarity between the candidate object and the target object according to the query record and the candidate query record of the candidate object; selecting anassociated object of the target object from the candidate objects based on the similarity corresponding to each candidate object; based on the association information base of the association object, generating a recommendation information list matched with the target object, wherein the association information base comprises historical query information of the association object; and combining therecommendation information lists of all the associated objects to generate a to-be-pushed information list of the target object. According to the query record of the target object, the associated object with high object attribute similarity with the target object is determined, the recommendation information list is generated, the information field of the push information can be automatically expanded, the range of the information field is widened while accurate push is ensured, and the information push efficiency of the target object is improved.

The invention discloses a public opinion information extraction and knowledge base generation method based on natural language processing. The public opinion information extraction and knowledge basegeneration method comprises the following steps of 1, text preprocessing; 2, named entity recognition: identifying company institution names and names, and finishing named entity recognition by adopting a neural network-based method; 3, relationship extraction: extracting six types of relationships in the financial field by adopting a feature layer + GRU + Attention; 4, entity linking; a Jaro winker distance method is adopted, and the distance between a link entity and a target entity is calculated to judge whether the link entity and the target entity are the same entity or not, so that entity disambiguation is achieved. According to the method, an end-to-end model and a feature extraction input class model are combined, a one-stop process from financial unstructured text to structured data storage is constructed, financial news context information is fully utilized, knowledge is extracted with fewer parameters and faster training prediction speed, and good performance is achieved inthe field of financial public opinion information.

The invention discloses a data migration method, device and system and a storage medium, and relates to the field of computer information, and the method comprises the following steps: obtaining migration parameters which comprise source end information and target end information; migrating the database from the source end device to the target end device according to the migration parameters; obtaining and archiving migration logs in the migration process; checking the row number of the database before and after migration to judge whether the data of the migrated database is matched or not; and integrating the archived migration log into the migrated database. A user only needs to input corresponding source end information and target end information according to a prompt and then automatically runs; according to the method, the remote database can be migrated to the target-end equipment only by inputting the source-end information and the target-end information on the page, a user doesnot need to input a complex database management command at all, the migration operation of the database can be completed in a'one-key 'mode only by inputting the source-end information and the target-end information on the page, thereby being simple and convenient.

The invention relates to the field of information, in particular to an event severity evaluating method and an event severity evaluating system based on network information. The evaluating method comprises the following steps of: (A), acquiring an event from a network; (B), quantifying the event by adopting more than one index to acquire an index quantification value of the event in each index; (C), determining the index weight of each index by adopting a hierarchical analysis method; and (D), determining the event severity level according to the index quantification value and the index weight of each index. The evaluating system using the evaluating method comprises an acquisition module, an index quantification module, an index weight determining module and a level determining module. The method and the system can be used for reasonably evaluating the severity of the event.

The invention discloses a Transformer-based natural language problem generation method in the civil construction information field, which comprises the following steps of: analyzing information distribution learned by each layer in a Bert model, training different corpora of different modules of a Transformer in combination with finite characteristics of training data, and proposing a low-layer network for training syntax grammatical characteristics, the high-level network is used for training a mechanism for obtaining semantic characteristics. Then, a UniLM thought is adopted to finely adjusta downstream task of the Bert, so that the natural language text generation capability of the model in the field of civil construction information is improved; the model provided by the invention hasrelatively high feasibility and effectiveness in problem generation in the field of civil construction information, and reaches a relatively high natural language problem generation level.

Network services or other resources service are bound to advertisement web pages. Before a service is provided user needs to obtain certain permission of service parameter. User must browse advertisement web pages to obtain service parameter, which can found from a place in web pages or generated by clicking a relevant options. Services include redirection of network hyperlink, loading pictures, displaying image, playing MTV etc.

The present invention relates to the field of information, in particular to a method and a system for express information services in the aspect of e-business. By solving the technical problem, the express information generated in the generating process of an e-business purchase order is transmitted to an express company in a certain simple and effective way to essentially improve the express information quality and speed up the express pick-up process. The adopted technical scheme is that the express information service system is provided and is associated with an e-business system and an express system to realize uploading and storage of the express information generated by the e-business system as well as search and pushing of the express information under a call request of the express system. The method and the system are applied to the aspects of e-business and express information services.

The invention discloses a service processing method, device, equipment and a medium. The method comprises the steps of obtaining basic information of a configuration application; determining a domainservice directory corresponding to the application; determining a workflow engine service corresponding to the application; displaying a preset process model stored in a process resource library of apreset process platform, and receiving a selection instruction for the target process model; and after a release command is received, generating a release document according to the basic information,the domain business directory, the workflow engine service and the target process model so as to call the workflow engine service in the release document to execute business processing. According to the method, the workflow engine service is managed by using the unified preset process platform and the reusable process resource library is provided; according to the method and the system, the unified process platform is accessed to the application system used by the user, the business processing can be completed based on the unified process platform, the switching among different application systems is avoided, the user experience is improved, the reuse of process resources is realized, and the working efficiency is improved.

The invention relates to a cross-platform dissemination trend evaluation and grading method based on specific events, and belongs to the field of network information. The cross-platform disseminationtrend evaluation and grading method based on the specific event specifically comprises four functional modules, namely an original network element detection module, a topic propagation calculation module of the specific event, a grade evaluation and threat analysis module of topic dissemination model parameters, and a task scheduling and optimization module. The problem that a dissemination trendevaluation and grading method of events is not considered in the prior art is solved, a quantitative evaluation method for specific event dissemination situations and a grading means with discrimination are provided, an information system framework is established, and cross-platform dissemination trend evaluation and grading of the specific events are realized. The subsequent monitoring work is guided according to the cross-platform dissemination trend evaluation of the specific events, so that the judgment capability on the emergency or influence degree of event dissemination is effectively improved, and the actual work is guided.

An alignment index forecasting method based on the integration of quotient spaces and knowledge sources for a photoetching process belongs to the field of advanced manufacturing, automation and information. The characteristic of the invention lies in that the alignment index forecasting method based on the integration of the quotient spaces and the knowledge sources for the photoetching process is proposed for the problem of product rework caused by alignment inaccuracy of a photoetching device in a micro-electronics production line. The method comprises the following steps of: firstly, using historical data of a plurality of alignment indexes of a wafer as different knowledge sources, and integrating the different knowledge sources to obtain a knowledge source integration model to be optimized, wherein the output of the model is the plurality of alignment indexes; secondly, carrying out polar decomposition on the above model to obtain three low-rank factors, and carrying out symmetrical transformation on each low-rank factor to obtain an equivalent-class factor; and finally, optimizing each equivalent-class factor by using an alternative-direction optimal strategy in the quotient spaces, and achieving synergic modeling on the plurality of alignment indexes by the optimization process of the knowledge source integration model. The method proposed by the invention can be used for combined forecasting of the plurality of alignment indexes.

The embodiment of the invention discloses a network slice cutover alarm shielding method and device, electronic equipment and a storage medium, and relates to the field of electronic information. The method comprises the steps of receiving a network slice cutover request, and obtaining cutover feature information contained in the network slice cutover request; acquiring a slice cutover topological graph corresponding to the cutover feature information, and determining slice cutover topological data corresponding to the slice cutover topological graph; inputting the slice cutover topological data into a preset alarm shielding model to obtain shielding configuration information output by the alarm shielding model; and performing alarm shielding processing according to the shielding configuration information. According to the mode, the alarm item to be shielded can be intelligently determined by means of the slice cutover topological graph and the alarm shielding model, and the accuracy and efficiency of alarm shielding setting are improved.

The embodiment of the invention provides a self-adaptive Chinese word segmentation method and a device for building information model. The method comprises the steps of constructing a BIM model dictionary according to a target BIM model; embedding a pre-constructed building domain term dictionary and the BIM model dictionary into a pre-trained Chinese word segmentation model; and performing word segmentation on to-be-segmented sentences in the building field based on the Chinese word segmentation model embedded with the BIM model dictionary and the building field term dictionary. According tothe method of the invention, when data mining needs to be carried out on the BIM model, the Chinese word segmentation model is optimized by using the feature information actually used in the BIM modelin a targeted manner; the word segmentation result is more suitable for data mining of the BIM, self-adaptive Chinese word segmentation of the BIM is achieved, the efficiency and accuracy of Chineseword segmentation in the field of building information are improved, the universality of existing BIM retrieval, entity matching and other methods can be improved, and the application range of the methods is widened.

The invention discloses a double-lock control method based on regular data processing and a medium, and belongs to the field of software information. The problems that in the prior art, in a request of a large data volume, the load of a system is increased. The consumption of memory resources is increased, and the processing speed of the system is reduced are solved. In a data regular processing scene, a user side applies for regular processing services to a system, and when information data of the regular processing services are generated in the system, a control lock in the current service period and a control lock in the next service period are added to the regular processing services; and if the system initiates a service switching period, or if the system receives a periodic processing request of the user side, performing periodic batch task processing in the service period or processing the request of the user based on the control lock. The method is used for controlling regularprocessing of data.

The invention is applied to the field of entertainment and information and provides a two-dimension code poker and an information processor. The two-dimension code poker comprises a poke basic body of materials and technology of any kind, a two-dimension code image printed on the poke body, a two-dimension word name and selectable word codes. The information processor comprises a sensing unit, a displaying unit, a keyboard inputting unit KEY, a power supply, an MPU information processing unit, a loudspeaker unit, a data storage unit, an externally arranged unit, an external port unit and the like. The sensing unit is connected with an information inputting port of the MPU. The displaying unit displays corresponding information through the MPU. The KEY is used for inputting operational instructions and data. The data storage unit stores work information content and the loudspeaker unit plays audio information. The externally arranged unit and the external port unit can achieve other functions. The two-dimension code poker and the information processor have the advantages of being low in cost and large in information capacity, visual in study, easy to understand and capable of overcoming the defect that a conventional poker is single in function.

The embodiment of the invention discloses an application log analysis method and device, electronic equipment, and a storage medium, and relates to the field of electronic information. The application log analysis method comprises the following steps: conducting similarity calculation on multiple collected application logs, and conducting filtering processing on the similar application logs according to a similarity calculation result; training the filtered application logs through a first algorithm and a second algorithm to obtain a plurality of prediction models; inputting to-be-classified application logs into the plurality of prediction models, and determining classification results of the to-be-classified application logs according to output results of the prediction models; analyzing the to-be-classified application logs according to the classification results of the to-be-classified application logs. According to the method, similar application logs can be filtered out, so that prediction models obtained through training can be more accurate, the log classification condition is accurately determined based on the prediction models, and log analysis is performed according to the log classification condition. According to the mode, the log analysis efficiency and accuracy are improved.

The method is suitable for the information field and provides a natural resource data management and distribution system based on spatio-temporal information cloud. The system comprises a natural resource data processing module, a space-time information cloud server, a data storage module, a data distribution module and a geographic information processing module, the spatio-temporal information cloud server is in communication connection with the natural resource data processing module; the natural resource data processing module comprises a natural resource data counting unit, a natural resource data checking unit and a natural resource data integration unit which are connected in sequence; the natural resource data integration unit is in communication connection with the spatio-temporalinformation cloud server, the data storage module, the data distribution module and the geographic information processing module are all in communication connection with the spatio-temporal information cloud server, and the database distribution server is further connected with the communication module. By integrating the natural resource data and the geographic information, the service and sharing of the geographic space information are realized, and the integrated data can be transmitted outwards through the communication module, so that the effective transmission of the data is realized.

The invention discloses a remote business processing method and device, electronic equipment and a storage medium, and belongs to the field of electronic information.The remote business processing method comprises the steps that: business handling data of remote business are obtained; the business handling data is examined and approved; and the approval state of the approval to a remote business handling system, and the approval state is uploaded to enable node members to achieve consensus and store certificates. According to the method, distributed processing of the business handling data is achieved through the block chain, the problem that an existing vertical management system is huge and complex is avoided, and meanwhile, cross-regional handling of government affair service business is achieved through the method.

The invention discloses a double-lock control method based on regular data processing and a medium, and belongs to the field of software information. The problems that in the prior art, in a request of a large data volume, the load of a system is increased. The consumption of memory resources is increased, and the processing speed of the system is reduced are solved. In a data regular processing scene, a user side applies for regular processing services to a system, and when information data of the regular processing services are generated in the system, a control lock in the current service period and a control lock in the next service period are added to the regular processing services; and if the system initiates a service switching period, or if the system receives a periodic processing request of the user side, performing periodic batch task processing in the service period or processing the request of the user based on the control lock. The method is used for controlling regularprocessing of data.

The invention discloses an information storage device for electronic information equipment, and belongs to the field of electronic information.The information storage device comprises a main body, check blocks are connected to the two ends of the main body in an embedded mode, a bottom plate is fixedly connected to the bottom faces of the check blocks, a first cavity and a second cavity are formed in the top face and the bottom face of the bottom plate respectively, and a fixing mechanism is installed in the first cavity; a positioning mechanism is installed in the second cavity, a protection mechanism is fixedly connected to the top face of the main body, the fixing mechanism comprises a rotating shaft, the rotating shaft is rotationally connected with the bottom face of the first cavity, and the two ends of the rotating shaft are fixedly connected with a connecting block and a rotating rod correspondingly. Meanwhile, the main body is easy to take and operate, the use efficiency of the device is improved, the main body can be protected, the risk of device damage is reduced, the service life of the device is prolonged, and cost consumption is reduced.

The invention discloses an electronic information transmission method, which belongs to the field of electronic information. In this method, three sets of safety paths are set at the lower level of the wireless and wired transmission of digital information, and under the action of the protection mechanism of the cloud translation device, the safety path is switched in real time and the safety path is selected, so as to achieve the safe transmission of information; by controlling the electromagnetic controller and The magnetic adsorption effect of the adsorption piece controls the disengagement and connection of the fastening piece and the limit piece, and under the action of the spring, it realizes the expansion and contraction of the control shaft, so that the fastening and positioning magnetic block and the connecting piece are disengaged and connected. It is convenient and timely to cut off the safety path, and select other paths, which can effectively play the role of safety protection; when the fastening and positioning magnetic block is separated from the connecting piece, the electrical connection between the first connecting block and the second connecting block is realized. , so that the warning light lights up and plays a warning role.

The invention discloses a hot topic participation behavior analysis method and system of users based on an implicit link, and belongs to the field of computer network information. Information such as past behaviors and self attributes the users are obtained; the common past behaviors between the users are extracted, thus obtaining a label similarity coefficient; the concern extents of the users to the common past behaviors are analyzed through a cloud model, thus obtaining the similarities of the users; whether an implicit link exists between two users or not is judged according to a threshold value; finally a drive force model is established according to the hot topic participation drive factors such as an explicit link, the implicit link, personal interests and behavior randomness of the users; and decisive factors of participating in topic discussions by the users are determined. According to the method, the implicit link is imported in the drive force analysis model. The multi-dimensional cloud model is provided for the users to establish the implicit link. An empowerment problem is avoided. The drive force factor of participating in topic discussions by the user is analyzed specific to each user. The method can be widely applied to related fields such as social network data analysis.

The invention discloses a message passing method and device and a readable storage medium, and relates to the field of electronic information.A message sender is controlled to send an original message to an intermediate component, then the intermediate component forwards the original message to message middleware and stores the original message in a database, and whether a message receiver receives the original message or not is judged, so that the original message is transmitted to the intermediate component. If the original message is not received, the original message is sent to the message receiver through the database, and compared with the previous mode that the message is transmitted directly through the message middleware, the original message can be sent to the message receiver through the database even if abnormal conditions such as message middleware faults and network fluctuation occur, so that the message is prevented from being lost or failed to flow, and the service life of the message is prolonged. And the storage capacity of the database is far greater than the data storage capacity of the message-oriented middleware, and when the service flow is large, the original message can be sent to the message receiver through the database, so that the accumulation of the original message in the message-oriented middleware is relieved, and the timely consumption of the message and the reliability of message transmission are ensured.

The embodiment of the present invention discloses a display method, device, electronic device and storage medium of an entity object end, relating to the field of electronic information, including: receiving an object display request triggered by a client, determining the region information corresponding to the object display request; parsing According to the user interaction data of each entity object end, determine the information change trend of each entity object end and its information ranking in the same object group; filter a preset number of entity object ends as candidate object ends according to the information change trend and information ranking, and obtain each candidate The object location information of the object end is used to calculate the push range of each candidate object end; the candidate object end whose push range matches the geographical information is pushed to the client for display by the client. This method can accurately sort the entity object side, and fully considers the matching degree between the user's location and the geographical location of the object side, which significantly improves the accuracy of the push results.

The invention provides a robot dialogue control method. The method comprises the steps: receiving input information; acquiring content information, domain information and intention information according to the input information; detecting whether a first classification rule contains the content information or not; under the condition that the first classification rule contains the content information, matching a corresponding result for the content information and the intention information according to the first classification rule; and under the condition that the first classification rule does not contain the content information, matching corresponding results for the content information, the domain information and the intention information according to a second classification rule.