107 results about "Conditional probability table" patented technology

The invention discloses a method for performing fault diagnosis by using model conversion. According to the method, processed related information of a fault mode effect analysis model is converted into a corresponding Bayesian network model by using a self-defined data structure while ensuring complete data, an elementary event, a logic gate and an intermediate event of a fault tree in a fault tree analysis model are converted into nodes in a Bayesian network respectively, and a corresponding conditional probability table in the Bayesian network is set. The fault diagnosis is performed through the converted Bayesian network model. The method of the invention expands the use of the Bayesian network model in the fault diagnosis, improves the diagnosis accuracy of a fault diagnosis model in practical application, ensures the universality of model conversion, and can realize cross-tool conversion among different fault mode effect analysis, fault tree analysis result and the generated Bayesian network.

A method, system and computer program product for distinguishing between a sensor fault and a process fault in a physical system and use the results obtained to update the model. A Bayesian network is designed to probabilistically relate sensor data in the physical system which includes multiple sensors. The sensor data from the sensors in the physical system is collected. A conditional probability table is derived based on the collected sensor data and the design of the Bayesian network. Upon identifying anomalous behavior in the physical system, it is determined whether a sensor fault or a process fault caused the anomalous behavior using belief values for the sensors and processes in the physical system, where the belief values indicate a level of trust regarding the status of its associated sensors and processes not being faulty.

The invention discloses a wireless sensor network abnormal event detecting method based on multi-attribute correlation and belongs to an anomaly detection technology in data mining.The wireless sensor network abnormal event detecting method comprises the specific steps that non-space-time attribute dependency model is established based on a Bayesian network, the attribute correlation confidence is calculated according to a conditional probability table, the similarity of points to be detected and abnormal points in a non-space-time attribute correlation mode is reflected; time correlation detection is performed based on a sliding window model, readings simultaneously meeting the time correlation mode and abnormal event attribute correlation mode are marked as temporal abnormal points to detect whether abnormal events occur or not by cooperating with neighbor node information.In addition, the wireless sensor network abnormal event detecting method calculates recent abnormal nodes closest to an abnormal event area center by adopting a Ritter's smallest enclosing circle algorithm, uploads the abnormal information of the abnormal nodes and can effectively reduce the data transmission amount and reduce the energy consumption of sensor nodes.The wireless sensor network abnormal event detecting method can be applied to wireless sensor network event detection of multiple sending components.

The invention discloses an overhead power transmission line running state assessment method based on a bidirectional Bayesian network. The method can be used for conducting a real-time assessment on the running state of an overhead power transmission line. According to the method, a Bayesian network structure for the assessment of the running state of the power transmission line is constructed with various factors which influence the running state of the power transmission line serving as a condition attribute set and the running state of the line serving as a decision attribute, a conditional probability table is obtained according to sample training, and by utilizing the bidirectional reasoning technology dedicated to the Bayesian network, the running state of the line can be judged by means of causal reasoning, and the hidden danger of the state can also be recognized by means of diagnostic reasoning; when an assessment error exists, a self-feedback system can be used for conducting early warning and correction, an assessment database, the network structure and parameters can be modified dynamically in real time so as to be adapted to an update, and therefore healthy running of the power transmission line is truly guaranteed.

The invention relates to a Bayesian network model based public transit environment dynamic change forecasting method. The Bayesian network model based public transit environment dynamic change forecasting method comprises the following steps of screening out various factors affecting public transit passenger flow fluctuation or travel time change; abstracting random jamming conditions of exterior environments and passenger flow or travelling time decision variables into nodes of a Bayesian network, determining a station set and the value range of the station set, and performing discretization on the historical information data of the station set and the value range of the station set; analyzing the influence relation between exterior environment jamming input nodes and passenger flow or travelling time decision nodes and establishing a Bayesian network structural diagram for public transit dynamic environment forecasting; determining a conditional probability table between determinant conditions and the decision nodes; computing the posterior probability when certain public transit passenger flow or travelling time occurs, and accordingly, achieving forecasting of public transit environment dynamic change. Combined with public transit incident detection under the environment of an Internet of vehicles, the Bayesian network model based public transit environment dynamic change forecasting method achieves a dynamic passenger flow time and space change forecasting function and provides data support for daily public transit operation and management.

The invention provides an anomaly detection device and method for security information interaction. The method includes the following steps of completing preprocessing operation on the basis of security information interaction records in a database to build a classifier model and conditional probability table and a group table for sequence alignment, analyzing security information interaction data, and judging the types of security information interaction related to the security information interaction data on the basis of the classifier model and conditional probability table and the group table for sequence alignment. The anomaly detection device and method for security information interaction have high security, the low false alarm rate and the low missing report rate.

Various embodiments provide systems, computer program products and computer implemented methods. In some embodiments, a system includes a method of providing a confidence-estimation-based inference, the method includes receiving a query concerning a patient from a user, accessing an electronic health record (EHR) for the patient, the EHR including a first component regarding the patient, querying the user, using a conversational interface, for a second component regarding the patient, the second component being in a natural language information form, receiving the second component regarding the patient in response to the query, calculating a first probability density function using the first component, and a second probability density function using the second component, combining the first and second probability density functions using a Gaussian mixture model, calculating at least one conditional probability table using the Gaussian mixture model and providing the confidence-estimation-based inference based on the at least one conditional probability table.

The present invention provides a reliable consumable risk assessment method based on Bayes reinforcement learning which solves the key information of consumable safety and provides data basis for design and recalling of consumables so as to guarantee the customer safety and market stabilization. The method performs historical data processing and screening, defines and establish a 'customer-product-environment' multi-type and multi-layer risk factor feature model, a damaging factor relation model and a Bayesian network topology structure, and establishes a condition probability table to calculate a particular commodity damaging event generation probability based on the election EM algorithm to realize prediction of damaging event generation, obtain the posterior probability when some types of damaging event are generated and finally five out multi-dimensional information output comprising the maximum risk grade and a task factor association relation so as to provide rich data basis for risk elimination of a consumable design phase.

The invention discloses a hydrologic process data and historical prior data combined flood prediction method. A weighted clustering algorithm is adopted to perform cluster analysis on rainfall information to approximate water content information; cleaning is conducted on data, missing data is supplemented, and wrong data is corrected; a general hydrologic model is selected according to the basin characteristics to obtain causal reasoning relationships between variables in the hydrologic process; a directed graph manner is adopted to represent each set of reasoning relationship, wherein nodes represent physical variables, the direction of edges indicates the causal reasoning relationships, and a bayesian network is constructed by using a topological structure; historical data is utilized incombination with the general hydrologic model for simulation to generate intermediate result data to form a sample set together; historical prior knowledge is obtained according to the sample set, and a conditional probability table is calculated to achieve training on the network and complete the comprehensive modeling of hydrologic process knowledge and the historical prior knowledge; test datais input to achieve flood prediction.

The invention discloses a wind power climbing event probability prediction method and system based on a Bayesian network, and the method comprises the steps: mining the dependency relationship betweena wind power climbing event and related meteorological influence factors such as wind speed, wind direction, temperature, air pressure, humidity, and the like, and building a Bayesian network topological structure with the highest fitting degree with sample data; quantitatively describing a conditional dependency relationship between the climbing event and each meteorological factor, estimating the value of each conditional probability in a conditional probability table at each node of the Bayesian network, and forming a Bayesian network model for predicting the wind power climbing event together with a Bayesian network topological structure; deducing the probability of occurrence of each state of the climbing event according to the numerical weather forecast information of the mastered prediction time; the value of the corresponding conditional probability at each node is adaptively adjusted, so that the inferred conditional probability result of each state of the climbing event is optimized, and the compromise between the reliability and the sensitivity of the prediction result is realized.

The invention discloses a method for diagnosing a fault of an oil-immersed transformer on the basis of a rough set and a bayesian network. The method comprises the following steps that (a) the type of the fault is determined, as much as possible input fault characteristic vectors are selected in an original sample set, and an input attribute set is determined; (b) discretization processing is carried out on a fault data set through a data discretization method in the rough set theory, and a discretization decision table is established; (c) establishment of the bayesian network is carried out through Matlab; (d) a conditional probability table is initialized, wherein all the possible conditional probabilities of each node relative to the father node of the node and the quantitative description of the corresponding problem domain are listed in the conditional probability table; (e) parameter learning is carried out, and a deduction engine is established to carry out deduction after the bayesian network is established; (f) a test sample set is input, the posterior probability is solved, and the type of the fault is judged. The method for the oil-immersed transformer on the basis of the rough set and the bayesian network can simplify the scale of a diagnosis network, enhance the anti-interference performance of the network, diagnose various faults of the transformer rapidly, and reduce the outage rate of the transformer greatly.

The invention provides a method and a device for diagnosing fault reason of a charging pile. The method comprises the following steps: the fault type of a target charging pile is acquired when the target charging pile breaks down; the fault type of the target charging pile is used as a prediction sample to be input to a preset bayesian network model, and the output of the bayesian network model isused as a fault reason diagnosis result corresponding to the fault type of the target charging pile, wherein the bayesian network model comprises a topological structure of the bayesian network and acorresponding condition probability table, and the topological structure of the bayesian network is used for expressing the corresponding relation between each fault type of the charging pile and each fault reason diagnosis result. According to the method and the device, automatic diagnosis of the fault reason of the charging pile can be realized, the diagnosis process is efficient, the diagnosisresult is accurate, further, the cause for fault occurrence can be quickly confirmed when the charging pile breaks down, and the charging pile can be maintained in time and in a targeted mode.

The invention provides a household air conditioner fault detection and diagnosis method based on the Bayesian network. The structure of the Bayesian network qualitatively describes the complicated causal relationship between most typical faults of household air conditioners and dependence occurrence conditions and fault symptoms thereof. The condition probability table represented by a directed edge of the Bayesian network qualitatively describes the probability value in the causal relationship. By means of the method, qualitative diagnosis information and quantitative data can be integrated,the knowledge and experience of industry experts and the additive information of diagnosis objects are sufficiently utilized, the fault diagnosis efficiency and accuracy are improved, and accurate fault diagnosis is achieved under the condition that diagnosis information is incomplete and inaccurate. The Bayesian network provided by the invention can effectively detect and diagnose most faults ofthe household air conditioners.

A method for predicting the success of a new chemical entity, including the steps of providing a signal related to the new chemical entity, providing a therapeutic index, providing a conditional probability table, providing a prior probability distribution, providing a prior N, and calculating a posterior probability distribution for the new chemical entity. An apparatus for predicting the success of a new chemical entity including an input for a signal related to the new chemical entity, a therapeutic index, a conditional probability table, a prior probability distribution, a prior N, and a processor calculating a posterior probability distribution for the new chemical entity.

The invention relates to a BNM-based (Bayesian network model-based) fishery forecasting method. The method includes the steps of discretizing historical ocean information datasets of fishery environment; establishing a table of conditional probability between Bayesian network structure charts and Bayesian network nodes; calculating a posterior probability distribution formula of fishery by the Bayesian network structure chart obtained by optimal learning algorithm; and forecasting the fishery by the obtained posterior probability distribution formula. The function of rapidly forecasting for the fishery can be realized by the BNM-based fishery forecasting method.

The invention provides a traffic violation severity level prediction method based on a Bayesian network. According to the method, directed diagram description based on the network structure is performed, the association relationship and the degree of influence between all the elements are described by using a directed acyclic graph in the network structure, all the elements are expressed by node variables, the association relationship between all the elements is expressed by the directed edges between the nodes, and the degree of influence between all the elements is described by the conditional probability. According to the method, the severity level of the traffic violation can be better predicted by the established Bayesian network model, and the prediction result can be applied to traffic violation management and traffic accident prevention.

The invention provides a question-answering method based on a Bayesian network. The method includes: receiving and parsing a question sentence, which is input by a user through a client, to identify atarget parameter, which represents user intent, and attribute parameters, which are associated with the target parameter, in the question sentence; inputting the target parameter and the attribute parameters into a pre-trained Bayesian network model, and utilizing a directed acyclic graph (DAG) and a conditional probability table set of the Bayesian network model to infer and obtain a value of the target parameter; and feeding back the value, which is of the target parameter and inferred and obtained by the Bayesian network model, to the user. According to the method, causal reasoning is carried out on the question sentence, and a question provided by the user is answered on the basis of a reasoning result. The invention also provides a question-answering device based on the Bayesian network and a computer-readable storage medium.

This invention patent application describes mathematical methods to evaluate and validate the numbers in the conditional probability tables of a Bayesian network. Using the methods described here, the nodes of interest in the network could be evaluated for validity of the information they contain and errors could be detected by domain experts or knowledge engineers very easily. If there is a disagreement between knowledge engineers or domain experts belief of what the interaction should be and what is detected in the behavior of nodes selected for validation as shown in the reports, then, those errors could be easily located in the structure of the Bayesian network by pin pointing the table, column and row of the problematic cell. Then, the knowledge engineer or domain expert could modify the numbers to reflect the correct behavior. These methods also provide significant insight into the structure and efficiency of the structural design of the Bayesian network as well as value of information in the network. Using this information, hypothesis oriented application of Bayesian network is possible and evidence most relevant to the hypothesis of interest could be instantiated first. Additionally, the shortest path to rule-out or rule-in of a hypothesis could be known before the network is used. Applications of these methods in computer software could allow for streamlined and semi-automated design and validation process and construction of Bayesian networks. Furthermore, by using an almost reverse process, information about a domain can be captured and sorted lists prepared which in turn will be used to prepare a preliminary Bayesian network. Data elicitation using the network created in this fashion will complete the structure and probability tables of the Bayesian network.

The invention discloses a machine tool thermal error compensation method based on an augmented naive Bayes network, and aims to effectively improve the machining precision of a machine tool. The method comprises the following steps: first, establishing a naive Bayes network classification model and a BAN network structure according to actually measured sample data; then, obtaining a conditional probability table through calculation and completing classification; finally, conducting corresponding compensation regulation on the machine tool through the real-time predication of thermal errors of the machine tool. According to the method, the thermal errors of the machine tools under specific work conditions can be predicted, the thermal error prediction can be more precise, the precision of thermal error compensation for the machine tool can be ensured, and the machining precision of the machine tool can be effectively improved.

Provided are improved systems, methods, and computer programs to facilitate predictive accuracy for strategic decision support using a query specific Bayesian network. An unconstrained domain model is defined by domain concepts and causal relationships between the domain concepts. Each causal relationship includes a value for the weight of causal belief for the causal relationship. In response to a query, the unconstrained domain model is transformed into a query specific Bayesian network for the domain model by identifying one or more cycles in the unconstrained domain model, eliminating the one or more cycles from the unconstrained domain model; identifying a sub-graph of the unconstrained domain model that is relevant to a query and creating one or more conditional probability tables that comprise the query specific Bayesian network.

The invention discloses a multistate system probabilistic importance analysis method taking epistemic uncertainties into consideration. The multistate system probabilistic importance analysis method includes firstly, quantifying the epistemic uncertainties by an evidence theory, and establishing a Markov model for computing state probability distribution intervals of all components of a multistate system; combining a logical combination relation between system states and unit states to obtain a system reliability interval and a conditional reliability interval of all the component states according to a conditional probability table; on the basis of a multistate system reliability model taking the epistemic uncertainties into consideration, putting forward the multistate system probabilistic importance analysis method based on an evidence theory frame and an importance interval ranking criterion based on an interval possibility degree method. The multistate system probabilistic importance analysis method taking the epistemic uncertainties into consideration has the advantages that the situation that a great number of epistemic uncertainties exist in a system service stage due to the small-batch and customization characteristics of a modern complex system is taken into full consideration, and accordingly the multistate system probabilistic importance analysis method is higher in engineering value as compared with a traditional multistate system importance analysis method based on a great quantity of sample data.

The invention discloses a fault diagnosis method based on a fault-test correlation matrix. The method comprises the steps of 1), determining a system fault-test correlation matrix D through a correlation graphic model of a system; 2), establishing bayesian network nodes; 3), connecting two layers of bayesian network nodes in the step 2) according to each element value in the matrix D; 4), setting a condition probability table for the bayesian network nodes expressing faults; 5), setting the condition probability table for the bayesian network nodes expressing test items; 6), setting an evidence variable according to a test result of each test item; and 7), obtaining diagnosis results. According to the fault diagnosis method based on the fault-test correlation matrix provided by the invention, logic relationships between the faults and tests can be described by employing concise and visual graphs, the occurring probabilities of the faults of the system are effectively integrated, the defects of carrying out the existing fault diagnosis method by applying the correlation matrix are overcome, compared with the correlation matrix, the method has the advantage that the diagnosis analysis results are more quantitative.

The present invention provides an intergenic interaction relation excavation method based on Bayesian network reasoning. The method comprises the following steps of: 1, employing a method of estimation of entropy by employing a Gaussian kernel probability density estimation quantity to calculate interaction information between genes, between genes and phenotypic characters and between phenotypes and the phenotypic characters; 2, employing a three-stage dependence analysis Bayesian network structure learning method to construct a Bayesian network including genes and phenotypic character nodes;3, employing the Bayesian estimation parameter learning method to perform parameter learning to obtain a contingent probability form between nodes; and 4, employing a Gibbs sampling Bayesian network approximate reasoning method to calculate the contingent probabilities of genes with different quantities and the phenotypic characters, and obtaining an intergenic interaction relation influencing thespecial phenotypic characters according to the calculation result. The intergenic interaction relation excavation method based on Bayesian network reasoning can help biology researchers of obtainingof epistasis genetic locuses influencing the special phenotypic characters to assist in gene function excavation and provide reference for hereditary basis analysis of complex quantitative charactersof different species.

The invention discloses a water pump rotor system power frequency vibration abnormity fault diagnosis method comprising the following steps: 1, building a T-S fuzzy fault tree taking the water pump rotor system power frequency vibration abnormity as top events; 2, converting the built T-S fuzzy fault tree into a diagnosis Bayes network according to specific conversion rules; 3, using the diagnosis Bayes network strong inferences to carry out fault diagnosis. The method can well solve the problems that Bayes network structural elements and a condition probability table are hard to determine, and the diagnosis result is better than a conventional method based on the Bayes network.

The invention discloses an evaluation method for the running state of a power transmission line. The method comprises that faults, emergency or severe defects and common and typical defect data over the years of parts in the power transmission line are arranged, parameters of the parts of the power transmission line are quantified, and a multidimensional matrix used for principle component analysis is constructed; key parameters of the parts in the power transmission line are determined; a Bayes network structure of the running state of the parts in the power transmission line is constructed, and the parts in the Bayes network structure include basic and protective facilities, towers, grounding wires, insulator chains, armor clamps, lightning-protection facilities, grounding devices and line protective zones; the state probabilities of the parts are calculated to construct a condition probability table; and defects of the parts of the power transmission line are evaluated. A mathematical model for evaluation parameters of different parts is established on the basis of the principle component analysis method, and the efficiency in obtaining evaluation indexes is improved.

The invention discloses a method for predicting virtual network resource states so as to realize the purpose of predicting resource state data of virtual network elements according to expected targets of service quality elements. For example, the method can comprise the steps that a historical data set composed of historical quality monitoring data and historical resource state monitoring data is obtained; the service quality elements and the virtual network elements are defined as Bayes network nodes, Bayes network learning is conducted by using the historical data set, a Bayes network and a conditional probability table corresponding to all the nodes in the Bayes network are established, by using the directed relationships between the nodes in the Bayes network and the conditional probability table corresponding to the nodes, maximum probability values of the nodes corresponding to the virtual network elements are found out on the condition that the value of the node corresponding to the specific service quality element is within the range of set quality data, and the prediction resource state data of the virtual network elements are obtained. In addition, the embodiment of the invention provides a device for predicting the virtual network resource states.

The invention discloses a latent variable model-based user preference extraction method. The method comprises the steps of firstly selecting N commodity relative attributes to form a commodity property set, building according to historical data to obtain a bayesian network, searching in the bayesian network to obtain a maximal semi-clique, and then inserting a latent variable L, showing user preference, into the maximal semi-clique, so as to obtain a latent variable model, wherein L being equal to 1 shows that a user prefers, and L being equal to 0 shows that the user does not prefer; performing parameter learning on the latent variable model to obtain a conditional probability table of various nodes in the latent variable mode; then according to the conditional probability table of the latent variable L, performing user preference extraction: searching to obtain an attribute combination item corresponding to a conditional probability maximum when L is equal to 1, wherein the attribute combination item corresponds to commodity types most preferred by the user; searching to obtain an attribute combination item corresponding to the conditional probability maximum when L is equal to 0, wherein the attribute combination item corresponds to commodity types least preferred by the user. By aiming at the user preference hidden in commodity evaluation data, the more objective and realistic user preference results are extracted by the structure of the bayesian network.

The invention relates to a system fault positioning technology based on a Bayesian network, and the technology comprises the steps: carrying out training of a Bayesian network model based on a historical fault data set, so as to generate a Bayesian network inference machine, and summarizing the law that a system interface is liable to cause a fault through the Bayesian network inference machine ina conditional probability table form; executing observation of system change, and triggering the Bayesian network inference machine to predict after determining the observed node with the fault; foreach change source which may cause a fault at the observed node, enabling the Bayesian network inference machine to calculate a posterior probability from which the observed node is caused to have afault according to the conditional probability table; and selecting the calculation result with the maximum posterior probability as a fault positioning result and outputting the result.

The invention provides a man-machine interaction system task reliability modeling method considering a path dependence effect, and the method comprises the following steps: (1) analyzing a man-machineinteraction system task profile, and recognizing main factors causing all function faults of a system; (2) proposing a man-machine interaction system task reliability modeling method, and establishing a system task reliability model by adopting a Bayesian network (BN) method; (3) for factors which may cause human errors, locally correcting the system task reliability model, and endowing the modelwith a conditional probability table based on a path dependence condition; and (4) calculating a system task success probability. According to the method, the task reliability characteristics of theman-machine interaction system with the path dependence effect can be accurately described, so that the established task reliability model is more accurate.

The invention discloses a reservoir flood regulation multi-dimensional uncertainty risk analysis method based on a Bayesian network. The reservoir flood regulation multi-dimensional uncertainty risk analysis method comprises the following steps: identifying risk factors of regulation starting water level uncertainty and flood forecasting uncertainty; carrying out bayesian network structure learning based on an expert experience method; carrying out bayesian network parameter learning to acquire a conditional probability table (CPT) of each node; performing Bayesian network probability reasoning; carrying out risk calculation and analysis. In order to couple the influence of regulation starting uncertainty and reservoir flood forecasting errors on reservoir flood control scheduling risks, areservoir flood control risk analysis model based on the Bayesian network is established, and coordinated conversion of reservoir interest benefits and flood control risks in the flood season can beachieved; the bidirectional reasoning of the Bayesian network can establish a reservoir scheduling risk bidirectional analysis and evaluation mode, and has a good application prospect.