74results about How to "Fully portrayed" patented technology

The invention relates to a method and system for establishing a multilayer oil pool inter-well connectivity model. The multilayer oil pool inter-well connectivity model capable of simulating oil and water dynamic states is established oriented to the limitation that by means of a current connectivity model, only the dynamic change of production liquid can be predicted, the dynamic states of the oil phase and the water phase cannot be calculated respectively, and connectivity analysis cannot be carried out layer by layer. According to the model, an oil pool system is layered and scattered into a series of inter-well communication units, matter balance equation calculation is carried out with the inter-well communication units as simulating objects, pressure solving and inter-well flow calculation are achieved, an inter-well saturation degree tracking calculating method is established with the combination of the lead edge boosting theory, and the output dynamic states of oil and water of all layers at well points are obtained; a model parameter inversion method is established in a fitting mode through dynamic states based on a random disturbance approximation method, a projection gradient method and the like. Layered inter-well flow distribution coefficients, single well longitudinal production liquid and production oil splitting coefficients and other information can be obtained in real time, the oil pool plane, the longitudinal oil and water flow relation and the production measure change can be more accurately reflected to guide the practical production of an oil field.

The invention discloses a multidimensional field key knowledge-based recruitment information matching method. Keywords are extracted, direct matching between resume keywords and a position is conducted and correlation matching between a resume and position keywords is conducted to realize multidimensional precise matching between the resume and the position. The method comprises the steps of: extracting the keywords in position information and resume information, and expressing the position information and the resume information with a vector; carrying out the direct matching to obtain a direct-matching-obtained text similarity score; carrying out the correlation matching to obtain a correlation-matching-obtained text similarity score; acquiring a measurement index for predicting difference; and performing weighting calculation to obtain a final resume and position text similarity score which is used as an information matching result to complete recruitment information matching. The method can comprehensively show the information and knowledge between the keywords and the information and knowledge which are expressed by the keyword combination, effectively improves the matching precision of the position and the resume and reduces the human cost of HR (Human Resource) in recruitment.

The invention discloses an industrial control abnormal behavior detection method based on multiple machine learning algorithms. The method comprises the following steps of firstly, collecting and establishing flow data of a power generation distributed control system; respectively constructing a training sample set and a test sample set by adopting the calibrated flow data; carrying out multi-dimensional feature extraction and vectorization processing on the sample; then establishing a classification model based on multiple machine classification learning methods, finally collecting real-timeflow data of the power generation distributed control system and inputting the real-time flow data into the classification model, if an output classification result is malicious flow, judging that anindustrial control abnormal behavior occurs, and otherwise, judging that the industrial control abnormal behavior does not occur. According to the method, samples can be effectively classified and detected, the abnormal behavior problem existing in the industrial control system can be rapidly detected, and potential malicious and abnormal behaviors can be accurately identified.

The invention provides a financial information processing method. The financial information processing method comprises steps that the financial data is acquired and stored, and the acquired financial data is extracted; the acquired financial data is pre-processed, including cleansing, association and classification; the pre-processed financial data is processed, including desensitization protection, grading processing and packaging storage, risk analysis on the financial data respectively after pre-processing and after processing is carried out; a risk analysis result is outputted and displayed. The invention further provides a financial information processing system. Through the method and the system, repayment capability, repayment willingness and post-loan risk of a loan user can be accurately determined.

The invention relates to a cone beam CT (Captive Test) iterative reconstruction algorithm and especially relates to a creating method of cone beam CT iterative reconstruction algorithm projection matrix. The invention provides a projection matrix characterization method based on a finite element model and a Radon operator for the high-precision characterization problem of the cone beam CT iterative reconstruction algorithm projection matrix. The invention provides a new projection matrix characterization method by combining the respective characteristics of a ray coverage model and a basis function model and fully considering the mathematic characterization of various physical processes of the projection according to the ray projection rule beginning from the mathematic characterization of a continuous three-dimensional natural image, so that the projection process is fully characterized. As shown in a test result, the method disclosed by the invention effectively improves the characterization precision and reconstruction quality of the model.

The invention discloses a remote sensing image variation detecting method on the basis of weighted Gabor wavelet characteristics and two-stage clusters. The processed objects include optical remote sensing images and SAR (synthetic aperture radar) images and the remote sensing image variation detecting method includes (1) generating difference images according to remote sensing image types; (2) subjecting the difference images to Gabor wavelet transform; (3) extracting multiscale and multidirectional characteristics of the Gabor wavelet transform of the difference images; (4) designing the weighting coefficient and acquiring the weighted Gabor wavelet characteristics; (5) clustering the weight Gabor wavelet characteristics by means of the two-stage cluster strategy; (6) acquiring variation detecting results. By the remote sensing image variation detecting method, loss of marginal information is reduced, stronger, weak and slight varying areas can be detected at the same time, the total mistake pixel number is decreased, more detail information is reserved and variation results can be effectively extracted.

The invention discloses a remote sensing image recommendation method based on multi-attribute fusion, and belongs to the technical field of intelligent recommendation and distribution of remote sensing images. The method comprises the following steps: firstly, carrying out multi-dimensional knowledge graph modeling on responsibility information and ordering information of a remote sensing user andattribute information of a remote sensing image, and respectively carrying out embedded representation on the knowledge graph to obtain a user responsibility embedding vector, a user ordering embedding vector, an image order embedding vector and an image attribute embedding vector; performing semantic information extraction such as target detection and surface feature classification on the remotesensing image to obtain an image surface feature proportion vector to solve a problem of an information representation method of a user and an image in a recommendation process in the remote sensingfield effectively; and then, with a collaborative filtering recommendation network based on multi-attribute fusion, realizing fusion of multi-dimensional input by means of high-dimensional space modeling capability of a neural network, and embedding various input vectors in different spaces into a unified vector space to realize a better pairing recommendation effect.

A polarized SAR image classification method based on superpixels and a topic model comprises steps of: inputting a polarized SAR image to be classified, and subjecting the same to refined Lee filtering; acquiring the sample points of the image processed the step 1 by using uniform sampling to obtain a sample point set; extracting three types of features from sample point set and normalizing the three types of features separately to obtain feature sets F1, F2 and F3; clustering the feature sets F1, F2 and F3 separately to form visual dictionaries V1 and V2 , V3, and merging the visual dictionaries V1 and V2 , V3 to form a multi-feature visual dictionary V; performing over-segmentation on the image processed in the step 1 to obtain a plurality of superpixels, and performing sparse coding on each of the superpixels according to the dictionary V; performing feature learning on the sparse coding of the superpixels by using the topic model, and performing classification by using a SVM classification method to obtain a final classification result. The polarized SAR image classification method can represent the heterogeneous region of the polarized SAR image.

The invention provides a user portrait constructing method and device as well as a recommending method and device. The user portrait constructing method comprises the following steps: acquiring a user's characteristic data, and determining the type of the characteristic data; constructing a data portrait according to the characteristic data and the type of the characteristic data; and constructing a user portrait according to the data portrait. The technical scheme of the invention realizes that the user owned personalized information is described by using sentence vectors and article vectors, the insufficiency in portraying the user by text labels is compensated, and implicated information at the semantic level can be expressed, so the portrait has semantic information, and the user is more comprehensively and accurately portrayed.

The invention provides a method and a system for predicting a function label of an App (Application). The method comprises the steps of acquiring a target description text for describing a target App; carrying out word segmentation on the target description text to acquire prediction data; and generating probability distribution of the target App in a plurality of function labels according to the prediction data and a target theme model, wherein the plurality of function labels are a plurality of themes of the target theme model. According to the method for predicting the function label of the App, word segmentation is carried out on the target description text of the App to acquire the prediction data, the prediction data are processed through the theme model, the probability distribution of the App in the plurality of function labels can be obtained, thereby classifying the multifunctional App objectively and comprehensively describing the function of the App. Furthermore, because the App is automatically labeled by use of a computer in the method, the complicated process for manually labeling the App is avoided.

The invention provides a multi-dimensional feature construction method and device, an electronic device and a storage medium. The multi-dimensional feature construction method comprises the followingsteps: when a feature construction instruction is received, acquiring historical access data of the target application program, detecting current access data of the target application program, calculating the similarity between the historical access data and the current access data by adopting a K-L divergence algorithm; obtaining first feature data, obtaining current time and determining a targettime slice to which the current time belongs, calling a target activity degree corresponding to the target time slice from pre-stored activity degree data, processing the target activity degree; acquiring second feature data, further acquiring third feature data acquired by a configuration sensor, performing normalization processing on the first feature data, the second feature data and the thirdfeature data, and performing data analysis to obtain multi-dimensional features, so that the description of user features is more comprehensive, and the flexibility is higher.

The invention provides a river course three-dimensional modeling method based on multi-attribute supervoxel and graph cutting, belonging to the field of geological body three-dimensional model construction. Aiming at the shortcomings of single seismic attribute describing river course, the invention provides a multi-attribute fusion method based on improved local linear embedding, the preferred attributes are fused into new attributes by ISOLLE algorithm, Considering the non-linear relationship between seismic attribute data, a non-linear fusion method is adopted, the fusion attribute is better than the previous attribute, and the accuracy of river edge and region description is improved, which lays a good foundation for the next segmentation and reconstruction. The invention is based on the river course segmentation method of supervoxel and graph cutting, and generates three-dimensional supervoxel through simple linear iterative algorithm, and the generated supervoxel is well close tothe edge of the river course, and has good homogeneity, and then obtains the final segmentation result by combining the graph cutting frame, and obtains the three-dimensional model of the river course surface in the way of extracting the isosurface.

The embodiment of the invention discloses a data source quality evaluation method in a sensing cloud environment. The method comprises the following steps: acquiring current and historical monitoringdata of a sensing cloud storage data source, the sensing cloud being a combination of cloud computing and a wireless sensor network and used for collecting and processing monitoring data from a plurality of sensor nodes or sensor sub-networks; integrating the monitoring data of the data source based on the spatial correlation and the time correlation, and determining a data true value; generatingan initial quality evaluation vector of a data source based on the data true value, and adjusting the initial quality evaluation vector of the data source according to a quality rule; and calculatinga final quality evaluation result of the data source according to the adjusted initial quality evaluation vector of the data source. By adopting the method and the device, the data source quality canbe described from multiple angles, and the description of the data source quality is more comprehensive.

The invention discloses a credit score value determination method and device, electronic equipment and a storage medium, and relates to the field of machine learning, and the determination method comprises the steps: obtaining a plurality of account features of a target account based on an account identifier of the target account, obtaining a feature set, inputting the feature set to a pre-trained federal learning model, and obtaining a credit score value of the target account; the method comprises the steps of obtaining a parameter value of each account feature, performing weight coding on each account feature to obtain an evidence weight value of a feature box to which the account feature belongs, and determining a credit score value of a target account in a current scoring mechanism based on the parameter value of each account feature and the evidence weight value of the feature box to which the account feature belongs. According to the method and the device, the technical problem that the security of customer data privacy cannot be ensured in a mode of improving scoring accuracy by acquiring data of a plurality of institutions to jointly train a customer credit scoring model in related technologies is solved.

A medium and long-term load curve generation method for satisfying an electricity quantity constraint comprises the steps of reading an annual maximum load, an annual total electricity quantity, typical daily load curves in summer and winter, monthly load characteristics and 8,760 hours of historical load data into a program, and organizing the 8,760 hours of historical load data by the month; calculating a daily maximum load in a prediction year with the obtained 8,760 hours of historical load data organized by the month as original data, and then obtaining a time-sequence load curve of the prediction year according to a conservation of charge as a principle; and finally, calculating to form an accurate continuous load curve and a sectional continuous load curve on a 168-hour cycle. According to the medium and long-term load curve generation method for satisfying the electricity quantity constraint, typical daily load curve characteristics and monthly load characteristics are fully considered in the switching of the time sequence load curve and the continuous load curve, shape characteristics of the load curves are guaranteed to satisfy the requirement on the premise of the electricity quantity constraint; and the load tendency and the change characteristics can be described and depicted more comprehensively, and a foundation is laid for establishment of an optimized planningmodel of an electric power system.

The embodiment of the invention provides a sound processing method, device and equipment. The method comprises the following steps that at least two feature vectors of an audio frame are obtained, wherein the audio frame comprises voice signals and noise signals, and acoustic characteristic type of the at least two feature vectors are different; fusion processing is conducted to the at least two feature vectors to obtain a fusion feature vector; the label value of the audio frame is obtained based on the fusion feature vector, wherein the label value is used for indicating the power ratio of the voice signals to the noise signals in the audio frame; and processing results of the audio frame are determined based on the label value of the audio frame and the noise signal power of the audio frame. The sound processing method, device and equipment are used for improving the effect of noise suppression.

The invention provides a day-ahead low-carbon scheduling decision-making method with the consideration to wind power integration based on the theory of ordinal utility. The method employs a multi-scene method for describing the uncertainty of wind power output, and enables a power scheduling decision-making problem to be constructed into a two-stage coordinated optimization model, wherein the model can achieve the depicting the day-ahead clearing and real-time balance process of scheduling decision making, and enables the scheduling decision making result to be higher in adaptability under theuncertain wind power integration. Meanwhile, the method employs the theory of ordinal utility for the consideration of the economical performance and carbon emission target, and achieves the more complete depicting of the preference information of a decision maker for each target.

The invention provides a rapid flood forecasting method and device based on rainfall similarity and model parameter intelligent adaptation, and the method comprises the steps: 1, building a rainfall flood conversion model, and recognizing the rainfall flood features; step 2, calculating the similarity of flood-causing factors: starting from the magnitude of flood-causing rainstorm, the rainfall pattern and the size of each component influencing rainfall in the early stage, calculating a similarity index vector based on a similarity theory, and quantifying the similarity among the flood-causing factors; step 3, transferring parameters, and constructing adaptation mapping among rainfall and flood samples: mutually transferring the parameters among the rainfall and flood samples, creating a model precision matrix after parameter transfer, establishing mapping between similarity index vectors and the precision matrix, and determining the importance degree among similarity indexes; and step 4, intelligent parameter adaptation: based on similarity calculation, clustering analysis and collaborative filtering theories, the reference model uses each set of parameters to forecast the precision of historical similar flood, the parameter with the highest precision is adapted to the model, and flood forecasting is carried out.

The embodiment of the invention discloses a method, a device, an electronic device and a storage medium for acquiring a user portrait. The method comprises the following steps of: acquiring the facialfeature information of the current user in each application system; Acquiring a current user portrait of the current user in each application system according to the face feature information of the current user in each application system; A target us portrait of that current user is determine based on the current user portrait of the current user in each application system. The user portrait canbe obtained more comprehensively, and the user characteristics can be described more accurately.