39results about How to "Realize association analysis" patented technology

The invention relates to a network security situation awareness system and method based on information correlation. The system comprises a data acquisition module, a network security situation evaluation module, a network security situation prediction module and a network situation visualization module, wherein the data acquisition module is used for acquiring network essential information; the network security situation evaluation module is used for quantitatively analyzing threat, frangibility and stability of the network by utilizing the network essential information, thereby analyzing the current network security situation; the network security situation prediction module is used for predicting the network security situation according to historical information and current state of the network security situation; and the network situation visualization module is used for visually displaying network security indexes according to the analysis and prediction result of the network security situation. The invention solves the problem that the existing network situation awareness system lacks data validity verification, data correlation and quantitative analysis, so that the network security situation awareness is more accurate.

The invention discloses a method for quickly positioning a service fault based on a log. The method comprises the following steps: a) adding a tracking ID in a first service request operated by a customer at a front end; b) when a network device, a server and an application program receives the service request having the tracking ID, adding the tracking ID in a generated new request or reply after processing, and recording the tracking ID in the log; and c) searching the log having the tracking ID to position the service fault. According to the method disclosed by the invention, the tracking ID is set to serially connect necessary service logs in the modules to realize association analysis of massive log information of multiple modules and multiple systems, quickly and accurately position the service fault and solve the problems that the existing log information is difficult to be matched and that the association analysis cannot be carried out; and in addition, the specific tracking ID is set and the customer ID, the request time and the service type therein are utilized, so that the counted information can be used for quickly positioning the fault and optimizing the service performance.

The invention discloses an audit and depth analysis system and an audit and depth analysis method of business data. The system comprises a network receiving unit and a processing unit, wherein the network receiving unit is used for receiving a network data packet; the processing unit is used for receiving the network data packet to form metadata including an applicant, request time, a target system and business data, extracting business information according to the business information extraction rule, carrying out correlation analysis on the business information according to correlation analysis knowledge, and maintaining an extraction rule base and a correlation analysis knowledge base. The method comprises the following steps: building a knowledge base including the business information extraction rule and business information; receiving the network data packet, extracting the metadata and storing the metadata into the knowledge base; reading the extraction rule according to the metadata, and extracting the business information according to the extraction rule. The audit and depth analysis system and the audit and depth analysis method have the beneficial effects that depth utilization of the network data packet is achieved by different business information extraction rules and business information, and correlation analysis of the business data and other business data is achieved on the basis of metadata updating of the knowledge base.

The invention discloses a unified log data analysis method. A unified log data set can be generated using unified log scripts in a website system according to the requirements of the unified log scripts, thereby facilitating automatic analysis on log data. First, the unified log data in the unified log data set is grouped according to a user identifier and a log type. For multiple pieces of unified log data in the same group, a complete operation track of the same user can be obtained through the combination of function points to track the service of the user. For multiple pieces of unified log data in the same group, the function points having the durations exceeding a threshold are determined by comparing the duration of each function point to analyze the performance of the function points. For multiple pieces of unified log data in the same group, the use frequency of each function point is counted to analyze hot spots of the function points, and the front and back functions of eachfunction point are counted to analyze the correlation of services.

The invention discloses a visual analysis method and system for multi-modal data based on sketch interaction. The method comprises the steps that (1) an original dataset is divided into multiple visual data structures; (2) matching between the visual data structures and visual forms is performed; (3) according to a mapping relation from the original dataset to the visual data structures to the visual forms, the original dataset is decomposed into multiple associated information sides, each information side is displayed in a view through one visual form, and a multi-viewing-angle association view is generated in combination with layout information selected by a user; and (4) sketch symbols drawn by the user who performs sketch selection operation on the multi-viewing-angle association viewaccording to a target analysis demand are recognized, then the implication of a sketch gesture composed of the sketch symbols is analyzed, and a new view is generated according to the implication of the sketch gesture. According to the visual analysis method and system, the user is assisted in performing visual analysis on the multi-modal data based on sketch interaction, the user can be helped toquickly acquire data knowledge, and the data association law can be mined.

The invention provides a perceptual information integrated access system based on urban important infrastructure. The perceptual information integrated access system comprises a supporting layer, a core business layer and a data application layer;the supporting layer comprises a basic library and a knowledge library; the core business layer comprises a diagnosis center, an emergency dispatching center, a safety center, a data center, a monitoring center, an operation center, a user center and an operation and maintenance center; and the data application layer comprises a statistical analysis center and a data sharing center. The system obtains various real-time monitoring information based on the Internet of Things technology, is in butt joint with existing monitoring system data based on the MQTT technology, establishes and maintains a holographic digital mapping model through a big data and deep learning multi-dimensional perception data association analysis algorithm, and realizes intelligent evaluation, analogue simulation, real-time early warning and quick response of urban important infrastructure disasters and emergencies.

The embodiment of the invention provides a fault detection method and device for an NFV network. The method comprises the steps of obtaining information technology IT information data in a network function virtual NFV network and communication technology CT flow data of an application service, wherein the IT information data comprises IT information data of a virtual network function VNF layer, avirtual machine VM layer and a physical host HOST layer; associating the IT information data with the CT flow data to generate a user XDR call bill; and detecting the fault of the NFV network according to the XDR call ticket. According to the embodiment of the invention, the fault detection efficiency and accuracy of the NFV network are improved.

The invention belongs to the technical field of three-dimensional modeling, and particularly relates to an epidemic prevention management and control method and system based on a community information model and a storage medium, and the method comprises the following steps: S1, constructing a three-dimensional community information model by fusing a GIS technology and a BIM technology; S2, collecting multi-source heterogeneous data generated by community residents and summarizing the multi-source heterogeneous data according to categories; S3, performing data cleaning processing on the multi-source heterogeneous data obtained in the S2, and extracting key data information from the multi-source heterogeneous data; S4, performing semantic extension on the community information model according to the key data information; S5, constructing a community space-time knowledge graph according to the community information model after semantic extension; S6, building a community epidemic situation management and control visualization system based on the community space-time knowledge graph. The community epidemic situation management and control visualization system can solve the problems that the current community epidemic prevention management and control work needs to spend a large amount of manpower and is low in efficiency, and achieves the purpose of more conveniently and more efficiently realizing the community epidemic prevention management and control work.

The invention discloses a distribution transformer multi-state monitoring method and system. The method comprises the steps that distribution transformer data in each system device is acquired, wherein the distribution transformer data comprises archive data of distribution transformers and terminals, electric energy and load data, voltage monitoring data and distribution transformer running statedata; according to preset data identifiers, the distribution transformer data is allocated to corresponding monitoring modules; and the monitoring modules perform multi-state calculation and displayon the corresponding distribution transformers according to logic rules corresponding to the data identifiers. According to the distribution transformer multi-state monitoring method and system, by collecting the distribution transformer running data and fusing multi-system metering data for analytic application, multi-aspect quasi-real-time monitoring of distribution transformer metering and running states and unified collection, calculation and display of the distribution transformer data are realized, convenience is brought to unified supervision by service personnel, correlation analysis among the data is benefited, a power grid company is helped to grasp the states of the distribution transformers in time, and therefore the power supply reliability and the customer power supply guarantee ability are really improved.

The invention discloses a brain wave recognition method based on a quantum neural network algorithm. The method comprises the following steps that step 1, a brain wave multichannel collection module collects brain wave signals; step 2, a brain wave multichannel signal amplification module amplifies the brain wave signals; step 3, a brain wave multichannel filtering module calls a quantum state multichannel preprocessing module and a quantum neural network module to filter the brain wave signals; step 4, a brain wave multichannel recognition module calls the quantum state multichannel preprocessing module and the quantum neural network module to carry out mode recognition and feature extraction on the brain wave signals; and step 5, the brain wave multichannel recognition module outputs a brain wave feature recognition result. According to the method, brain waves can be described by using a quantum state wave function, characteristics of coherence, superposition, wave particle bijection and the like among different brain wave signals can be described, and self-learning and training are further completed according to different scenes, so that the recognition effect is improved.

The invention discloses an intelligent preventive test operation management and control system, which comprises a standard test database module, an annual production plan intelligent supervision module, a test report intelligent diagnosis and analysis module and a test data intelligent analysis module. The standard test database module is used to establish a standard data structure model, form a data standard system, and construct a new standard test database; the annual production plan intelligent supervision module is used for carrying out intelligent analysis and accurate matching on arrangement data of the production plan; the test report intelligent diagnosis and analysis module is used for examining the normalization of the text content of the test report; and the test data intelligent analysis module is used for carrying out mutual comparison and clustering analysis on test report data of the test equipment, globally analyzing problems of plans and test reports, and researching test data generality laws according to cities, units, equipment and test types. According to the test data intelligent operation management and control system integrating planning, management, supervision and analysis, a traditional working mode is changed, existing scientific and technological means are fully used, the management efficiency and quality of test operation are greatly improved, essential changes are brought to overhaul test work, and the informatization degree of the overhaul test work is comprehensively improved.

The present invention provides a pulse sequence signal correlation analysis and display system. A data processing unit comprises two basic data processing nodes and a plurality of extended data processing nodes, wherein the two basic data processing nodes are a time-domain envelope detection node and a pulse time-domain parameter calculation node respectively, and the extended data processing nodes establish processes of parameter data processing according to needs and perform different data processing according to the difference of association operation; a correlation processing unit comprises a data correlation processing node and a display correlation processing node; the correlation processing unit, on the basis of pulse time-domain parameter information, performs different interaction operation on a pulse parameter table to establish an automatic correlation index table, and performs correlation analysis of data information and correlation processing of display information; and a display processing unit is used for performing display processing on various kinds of information of pulse sequence signals. With the pulse sequence signal correlation analysis and display system of the invention adopted, an effective means is provided for the comprehensive analysis and display problem of pulse sequence signals.

The invention discloses a cloud resource fine management traceability system, which comprises a cloud platform, cloud resources, hardware equipment and a bottom information management component; the bottom information management component is used for collecting hardware equipment information; the underlying information management component is used for analyzing the collected hardware equipment information and storing an association relationship between the hardware equipment information; the underlying information management component is used for providing an interface for managing hardware equipment information, an interface for managing hardware equipment to be collected and an interface for querying hardware equipment information related to virtual resources to the outside; and the cloud platform is used for calling the underlying information management component interface to inquire and display the associated hardware equipment information when the cloud resources are displayed. The invention further provides a corresponding cloud resource fine management traceability method. On the basis of a traditional cloud computing platform architecture, the relation between upper-layer cloud resources and lower-layer hardware equipment is broken through, and rapid tracing of the cloud resources is achieved.

The invention discloses a land attribute classification method based on rental and return data of a shared object, and the method comprises the steps: carrying out the clustering analysis of the rental and return position information of the shared object, so as to form a virtual rental point of the shared object; taking the virtual rental point location as a unit, collecting the rental and returndata of the shared object to construct a time rental and return curve corresponding to the virtual rental point location; normalizing the rental and return curves at each time; based on Fourier transform, deconstructing each normalized rental and return curve is deconstructed into a rental and return curve function expression formed by superposing a plurality of sine wave functions; and constructing a rental and return curve parameter set of the virtual rental points, and performing clustering analysis based on parameters in the rental and return curve parameter set to form classification of each virtual rental point. According to the method provided by the invention, the land attribute can be identified by means of the data, especially the rental and return data, such as shared bicycles in the sharing economy, and the land attribute can be analyzed and classified by utilizing the existing data.

The invention discloses a privileged account threat analysis system, which comprises an intelligent threat auditing unit, a real-time threat monitoring unit and an overall configuration management unit which are connected with one another, and is characterized in that the intelligent threat auditing unit is used for auditing and monitoring account threat abnormal behaviors, and an administrator performs auditing analysis from different security dimensions and performs visual display. The privileged account threat analysis system can be compatible with structured and unstructured data of various operating systems, applications, databases, network equipment, security equipment and the like, can solve the data island problem existing in privilege threat analysis of a traditional system or tool, effectively carries out unified centralized management on external and internal threat behavior audits of the privileged account, realizes real-time omnibearing monitoring and association analysisof privileged account behaviors, can actively send the detected suspected privileged account threat behaviors to related personnel for real-time alarm, can improve the real-time protection capability,and can reduce further loss caused by the privileged account threat behaviors in an enterprise to the greatest extent.

The invention relates to the technical field of network security, in particular to an FTP behavior data processing method and an abnormal FTP behavior recognition method and device. The method comprises steps of obtaining FTP behavior data; performing feature extraction on the FTP behavior data based on a plurality of user-related features to obtain feature data, each user-related feature comprising more than one user-related sub-feature; based on the feature data, performing kernel density estimation on each user-related sub-feature to obtain a probability density value of each user-related sub-feature; determining a probability density value of each user-related feature according to the probability density values of all user-related sub-features contained in the user-related features; and determining a multi-dimensional probability density value of the FTP behavior data according to the probability density values of all the user-related features, so as to identify an abnormal FTP behavior according to the multi-dimensional probability density value of the FTP behavior data. According to the invention, the accuracy of identifying the abnormal FTP behavior is improved.

The invention discloses an industrial economy development condition monitoring method based on energy power big data. The method comprises the following steps: step 1, constructing a power index evaluation system; step 2, carrying out industrial economy data extraction according to a power index evaluation system of industrial economy, including data extraction code writing, system data extraction and mass data merging, so that basic data required by model calculation is formed; step 3, giving a reasonable decision index weighting method according to an internal statistical rule and an authority value between the basic data; step 4, according to the economic model, carrying out running calculation on the basic data to obtain an index weight and a bottom-layer index score, and according to the index weight and the bottom-layer index score, carrying out index calculation by using an index calculation formula; and step 5, carrying out data reorganization again according to the index score, and mining generality and personality laws shown in the data. According to the method, the industrial development condition can be accurately analyzed in the electric power field.

The invention discloses a multi-time scale data fusion quality analysis and optimization method and system, and the method comprises the following steps: carrying out the data association of a product production process, and constructing a quality database; the product data are associated through part numbers, and the part numbers comprise material part numbers and material batch numbers; process parameters and detection parameters are collected based on the part numbers and are associated with the part numbers, and a quality database is constructed. Constructing a quality evaluation model; constructing a quality evaluation model based on artificial experience and a quality database; and according to the current machining condition, the control process parameters are issued to the parameter control strategy in a single-piece or batch mode. Performing online control; and for the processing equipment, performing online control on the processing process through the parameter control strategy so as to ensure that the processing equipment performs processing according to the control process parameters issued by the quality evaluation model. According to the method, the problem of quality loss caused by local management and control and post management and control in the existing quality analysis is solved.

The invention relates to a correlation analysis mining method for tax-related data. According to a unified social credit code of a tax-related initial enterprise, a tax-related initial enterprise related person, a tax-related initial enterprise invoicing MAC code result set, a tax-related initial enterprise invoicing IP address result set and a tax-related enterprise mainboard serial number result set corresponding to the initial tax-related enterprise can be searched. Furthermore, a related person tax-related associated enterprise information set, an invoicing MAC code tax-related associated enterprise information set, an invoicing IP address tax-related associated enterprise information set and a mainboard serial number tax-related associated enterprise information set can be respectively obtained; and the tax-related associated enterprises contained in the related person tax-related associated enterprise information set, the invoicing MAC code tax-related associated enterprise information set, the invoicing IP address tax-related associated enterprise information set and the mainboard serial number tax-related associated enterprise information set are used to confirm the tax-related associated enterprises as the tax-related suspected enterprises, so that the association analysis and mining of the tax-related data can be effectively realized, and the method is safe and reliable.

The invention relates to the field of word net processing, and particularly provides a method for calculating entity relevance based on a word net. The method comprises the following steps: S1, a keyword acquisition component; S2, an entity space positioning assembly; and S3, an entity relationship calculation component. Compared with the prior art, the method has the advantages that the scatteredunits are marked in a unified mode on the basis, an intelligent discrete unit analysis system is constructed, the mining analysis and processing functions of discrete units are effectively achieved,correlation analysis of low-correlation data is achieved, and the method has good popularization values.