30 results about "Causal reasoning" patented technology

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 provides a large aircraft aviation big data fault detection and causal reasoning system and method based on a deep random forest algorithm. The system comprises a fault diagnosis platform, a fault reason reasoning platform, a database storage computer and a client. The method comprises the following steps of comprehensively monitoring and acquiring operation parameters of each systemof an airplane in real time to form massive data sources, acquiring typical characteristics of the signals through calculation of the characteristic parameter spectrum, and extracting and describingfault characteristics in residual signals using the characteristics as parameters and storing the characteristics into a parameter database. The airplane parameter database is established through thefault diagnosis computer and the fault reason reasoning computer, so that fault information of an airplane or possible faults of the airplane is covered, faults and reasons are determined through diagnosis of the fault diagnosis computer and reasoning of the fault reason reasoning computer, a maintenance/isolation scheme is provided, and health monitoring and fault diagnosis of all systems of thewhole airplane are further realized.

The invention discloses an online monitoring method for the abrasion forms and the abrasion state of a drill bit in the high-temperature alloy drilling process. The online monitoring method is used for solving the technical problem that an existing drilling process drill bit abrasion state monitoring method is poor in applicability. According to the technical scheme, signal features are extractedfrom the influence rule of drilling force signals based on the different drill bit abrasion forms of the high-temperature alloy drilling process, and a Bayesian network model of the drill bit abrasionforms and the signal features is set up through the drilling force and drill bit abrasion data based on the Bayesian theory on this basis; and then the drill bit abrasion forms are judged through Bayesian diagnosis and inference according to monitoring signals, and the signal features influencing the abrasion forms are acquired through Bayesian causal reasoning. Meanwhile, according to the tool abrasion curve rule, the signal features are monitored through an accumulation and control chart method, the target of monitoring the drill bit abrasion states in real time is achieved, and high practicality is achieved.

Methods, apparatus, and systems are disclosed for estimating causal relationships between observed variables. According to the method provided by the present disclosure, responding to the received observation data of the mixed observation variable, a mixed causal relationship target formula suitable for the continuous observation variable and the discrete observation variable is determined, whichincludes a causal relationship target formula for the continuous observation variable and a causal relationship target formula for the discrete observation variable, and the fitting inconsistency is adjusted based on a weighting factor of the observation variable. Then optimal solution is performed on the mixed causal relationship target expression by utilizing the mixed observation data and through mixed sparse causal reasoning suitable for continuous observation variables and discrete observation variables under the constraint of a directed acyclic graph so as to estimate the causal relationship among a plurality of observation variables. The embodiment of the invention is suitable for causal relationship estimation of the mixed observation variable, and the causal network structure haslow sensitivity to the observation variable estimation error, so that the accurate causal relationship can be obtained.

The invention discloses a causal reasoning method for correcting popularity deviation of a recommendation system. The causal reasoning method comprises the following steps: acquiring a matching score of a user and an article in the current recommendation system; predicting an article score according to the popularity degree of the article, and predicting a user score according to the preference of the user; and aggregating the matching score of the user and the article, the article score and the user score, predicting the matching score of the user and the article, and removing the influence caused by popularity deviation to obtain the final matching score of the user and the article. The method provided by the invention is a model-independent anti-fact reasoning framework, can be suitable for various recommendation systems, and can provide high-quality and accurate personalized recommendation contents for users by eliminating popularity deviation and improving the recommendation performance of the recommendation systems.

The invention relates to the fields of machine learning, natural language processing, causal reasoning and the like, in particular to a judicial decision reasoning method based on an interpretable causal model. The judicial decision reasoning method comprises the steps of analyzing and representing judicial data, defining a judicial decision inference framework, and constructing a judicial decision inference cause and effect model. Wherein defining of the judicial decision inference framework comprises the steps of extracting fact elements, identifying the fact elements and obtaining a decision result. The model construction comprises the steps of obtaining a reason variable, obtaining a result variable, constructing a correlation function, and constructing a judicial decision reasoning cause and effect graph. According to the judicial decision reasoning method, causal reasoning of judicial judgment is realized, and the problems that an existing judicial judgment reasoning method lacksa reasoning mechanism and is poor in interpretability can be effectively solved, and the accuracy of a case judgment result easy to confuse can be obviously improved.

The invention provides a method for carrying out commonsense causal reasoning on a short text. The method comprises the following steps that 1, a causal knowledge network is constructed; 2, the short text is subjected to causal analysis based on the causal reasoning algorithm, the sorting mechanism of commonsense causal relation strength included in the short text is obtained, the technical method can be applied to an existing question-answering system, and the answering result to a causal question (why-question) is improved. According to the method, the causal relation of the short text is analyzed through a causal mode, the coverage face is guaranteed by setting the size of a sliding window, and compared with a traditional Internet text corpus causal relation extracting method, the method is more accurate; the mode of combining sufficient causes and effects and necessary causes and effects is utilized for achieving the aim of effectively measuring the causal relation strength between lexical items.

The invention discloses an effective physiological feature selection and medical causal reasoning method based on interpretable machine learning. The method comprises the following steps: acquiring medical data from an electronic medical record; decoupling the feature space into a combination of a plurality of effective features through a plurality of feature selection methods; different feature selection methods are compared, and the rationality of adopting the SHAP value in the causal reasoning field is explained; the model features are evaluated based on the SHAP, and the association between the feature space and the prediction result is analyzed; causal information is incorporated into a feature space, and an interpretable machine learning model is constructed; according to different causal models, various Shapley Values are used for providing reasonable explanations; and outputting the importance degree of each feature, the contribution degree to the sample and the causal relationship with the prediction result. According to the method, disease development is explained and disease reasoning is carried out according to effective characteristics, and the effect and interpretability of the model and the accuracy of disease diagnosis are improved.

The invention provides a method, a device and a system for estimating a causal relationship between observation variables. A method for estimating a causal relationship between observation variables according to the present disclosure may include, in response to receiving expert knowledge for at least a portion of a plurality of observation variables, converting the expert knowledge into constraints required to be met by a causal relationship target of the plurality of observation variables; and by using the observation data of the observation variables, through sparse causal relationship reasoning, under the constraint of the directed acyclic graph and the constraint which needs to be met and is obtained by converting expert knowledge, carrying out optimal solution on the causal relationship target expression so as to estimate the causal relationship between the observation variables. The causal reasoning method can be merged into the causal reasoning process in a simple mode, expertknowledge is fully utilized, and a more accurate causal relationship is obtained.

The invention discloses an equipment efficiency evaluation device based on knowledge base rule reasoning. The equipment efficiency evaluation device comprises a sensing interface, a semantic analyzer, a rule reasoning machine, a natural environment influence knowledge base and an execution interface, the natural environment influence knowledge base comprises an entity knowledge base, an influence rule base and an algorithm base; the rule reasoning machine reasons rule information by by using deductive reasoning, inductive reasoning, causal reasoning, condition missing speculation, equipment analogy speculation and decision table judgment reasoning rule reasoning calculation methods; the semantic analyzer reads natural environment, equipment information and the like to obtain a semantic extension set. The invention discloses an equipment efficiency evaluation method based on knowledge base rule reasoning. The accuracy and completeness of the evaluation conclusion depend on the knowledge base and the constraint rules, and by continuously improving the concept elements and the constraint rules of the knowledge base, the method can adapt to various types of combat actions, natural environments and the like, and the overall intelligent level of efficiency evaluation is improved.

The invention provides a method for constructing a disease prognosis risk assessment model based on causal reasoning and a disease prognosis risk assessment system based on causal reasoning. Based oncausal reasoning, the probability of re-admission of a patient with cancer or cardiovascular and cerebrovascular diseases within one year is predicted by utilizing a full-connection neural network, and the effects of different treatment schemes are evaluated, so that doctors are assisted in making reasonable diagnosis and treatment measures, supporting clinical decisions and reducing medical expenditure.

The invention provides an event causeing method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining a target event set, respectively obtaining a cause graph event subset and a result graph event subset of the target event set in an event cause graph and an event result graph, and according to the cause graph event subset and the event cause graph, generating a first subsequent event set; according to the result graph event subset, the event result graph and the event cause graph, generating a second subsequent event set; and generating a subsequent event set according to the first subsequent event set and the second subsequent event set. Therefore, when a certain event or some events are known, subsequent events can be reasoned, and the event cause graph and the event result graph are utilized to perform causal reasoning on the time sequence of the events, so that the reasoning efficiency is improved.

The invention provides a common sense causal reasoning method and system based on weak evidence aggregation. The causal reasoning problem of a user is deduced by using the weak causal evidence and the common sense causal knowledge so as to obtain a causal sentence pair. The method is characterized in that the method comprises the following steps: acquiring a causal reasoning problem, determining a corresponding problem type, a problem premise and a candidate item according to the causal reasoning problem; identifying the type of the problem to obtain an identification result, and preprocessing the premise of the problem and the candidate item according to the identification result to obtain preprocessed data; inputting the preprocessed data into a pre-trained common sense causal reasoning model, and obtaining a causal sentence pair, so that a user makes judgment according to the causal sentence, wherein the common sense causal reasoning model not only considers the coverage rate and the accuracy rate of causal evidence, but also fuses various sparse weak evidence aggregation models containing common sense causal knowledge, so the method has a good causal reasoning effect.

The invention provides an accident causal reasoning method and device, electronic equipment and a readable storage medium. The accident causal reasoning method comprises the following steps: acquiring a plurality of accident related factors; according to the plurality of accident related factors, constructing an uncertain information Bayesian network topological structure used for reflecting the mutual influence relationship among the plurality of accident related factors; performing calculation according to the accident investigation reports and the uncertain information Bayesian network topology structure to obtain first parameter information and second parameter information; calculating a comparison result and an uncertain information Bayesian network topological structure according to the posterior probability of the first parameter information and the second parameter information, and determining an accident causal reasoning model; and inputting the received accident evidence into the accident causal reasoning model, so that the accident causal reasoning model outputs an accident causal reasoning result. Therefore, by implementing the implementation mode, the potential relationship between accident factors during small sample data can be reflected, so that the accuracy of a causal reasoning result is improved.

The invention is suitable for the technical field of computers, and provides an event reasoning method and device. The method comprises the following steps of: 1, acquiring a stimulation keyword; 2, searching a reasoning model matched with the entrance reasoning logic according to the stimulation keyword; 3, performing reasoning according to the matched reasoning model and outputting a reasoning conclusion to a conclusion stack; 4, judging whether a newly added reasoning conclusion exists in the conclusion stack; 5, when it is judged that a newly added reasoning conclusion exists in the conclusion stack, searching the reasoning model matched with the causal reasoning logic according to a conclusion keyword of the reasoning conclusion, and executing the step 3. By means of the method, logicmodeling, reasoning and repeated reasoning are conducted according to influences generated by the stimulation keyword in a news event, and whether the event causes changes of related industries and product index data or not, the specific change form, the change range and the like are analyzed, therefore an analyst is assisted in conducting fundamental analysis.

The invention relates to the field of project risk decomposition and identification, and discloses a project risk decomposition identification method based on a safe operation basic function requirement, wherein the method settles a problem of high difficulty in performing accurate risk identification and logical relation carding. The method comprises the steps of performing system decomposition on a complicated project system from a high grade to a low grade so that mutual independence and completeness of a lower sub-system and an upper parent system are realized; finding out a to-be-satisfied function requirement of each sub-system based on the safe operation basic function requirement; analyzing possible events which affect realization of the function requirement; performing causal reasoning on each possible event, and finding out a reasoning event of the possible event; and analyzing the result of each reasoning event. The project risk decomposition identification method is suitable for complicated systematical projects such as water power engineering and civil engineering. The project risk decomposition identification method has an important application prospect in risk identification and analysis aspects of the complicated systematical projects and can realize a Bayesian risk network for constructing a complicated systematical project.

The invention discloses a hyperbolic space-based causal reasoning method and system for event atlas enhancement, and the method comprises the steps: obtaining an event pair and a event atlas, obtaining an evidence event of the event pair based on the event atlas, and forming a causal evidence graph through the evidence event and the event pair, and the event pair comprises a reason event and a result event; converting the causal evidence graph into an event representation, mapping the event representation into a hyperbolic space from an Euclidean space, and obtaining a hyperbolic event representation; and performing directed causal reasoning on the hyperbolic event representation to obtain the final embedding of the event pair, and calculating the final embedding of the event pair to obtain a causal intensity score. According to the technical scheme, the stability and interpretability of causal reasoning can be improved.

A method, device, and system for estimating a causal relationship between observed variables are disclosed. According to the method disclosed by the invention, In response to receiving observations ofobserved variables, and the causal relationship target expression is determined based on the fitting inconsistency when the observation variables are used for fitting and the sparsity constraint on the causal network structure. Wherein the fitting inconsistency is adjusted based on a weighting factor of an observation variable, and the weighting factor of the observation variable represents the lower limit of the minimum cost required for fitting the target variable by using other observation variables except the observation variable. Then, through observation data and sparse causal reasoning, under the constraint of the directed acyclic graph, the causal relationship target expression is subjected to optimal solution, so that the causal relationship among multiple observation variables is estimated. By utilizing the method and the device, the sensitivity caused by an observation variable estimation error can be reduced, and a more accurate causal relationship can be obtained.