135 results about "Bayesian probability" patented technology

The present invention is uniquely designed to interact with web retailer's customers with real agent reaction times as they give astute answers directly concerning web retailer's products and goals. The entire process is manageable through a third-party website which includes scripting, settings and other parameters selected by the web retailer. An artificial intelligence engine uses the combination of Bayesian probability keyword selection, natural language parsing and regular expression processing. The technology updates its response database with every client interaction-learning always takes place as it simulates a live agent, in real time. Every client interaction is recorded and analyzed, and as a result of the analysis the changes in the answer database are made.

A computer-implemented method for discovering network attack paths is provided. The method includes a computer generating scoring system results based on analysis of vulnerabilities of nodes in a network configuration. The method also includes the computer applying Bayesian probability to the scoring system results and selected qualitative risk attributes wherein output accounts for dependencies between vulnerabilities of the nodes. The method also includes the computer applying a weighted-average algorithm to the output yielding at least one ranking of nodes in order of likelihood of targeting by an external attacker.

The invention provides an intelligent fault diagnosis method of a numerical control machine tool. The intelligent fault diagnosis method uses fault tree diagnosis as a base, faults are diagnosed by combining a Bayes probability and rule reasoning method, fault cause is found, and maintenance advices are proposed. The intelligent fault diagnosis method includes: firstly extracting fault information from historical maintenance records and machine tool operation instructions, and establishing a numerical control machine tool fault information knowledge base; searching fault trees belonging to a fault position in the fault information knowledge base according to fault characteristic information, calculating probability of occurrence of each fault tree under the current condition if a plurality of fault trees fit conditions, and determining sequence for diagnosing the fault trees according to size of the probability; and finally judging whether all sub-events of top events of the fault trees occur, returning to a fault treatment method corresponding to the events and serving as a solution if the sub-evens are bottom events of the fault trees, otherwise searching sub-events of the events continuously until finding all bottom events fit the fault characteristic information. The intelligent fault diagnosis method of the numerical control machine tool is rapid in diagnosis speed and accurate and reasonable in diagnosis result.

The present invention is uniquely designed to interact with web retailer's customers with real agent reaction times as they give astute answers directly concerning web retailer's products and goals. The entire process is manageable through a third-party website which includes scripting, settings and other parameters selected by the web retailer. An artificial intelligence engine uses the combination of Bayesian probability keyword selection, natural language parsing and regular expression processing. The technology updates its response database with every client interaction-learning always takes place as it simulates a live agent, in real time. Every client interaction is recorded and analyzed, and as a result of the analysis the changes in the answer database are made.

The invention relates to a fabric defect detection and classification method based on unsupervised segmentation and ELM (Extreme Learning Machines), and belongs to the fields of computer vision, pattern recognition and image application. Fabric defect images are input, and fabric defect image segmentation is realized to obtain segmented fabric defect images; shape and texture features of the segmented defect images and the original defect images are extracted to realize feature extraction of the fabric defect images; a feature set and labels of the to-be-classified fabric defect images are used as a training set to train an ELM classifier to obtain ELM classifier parameters; and fabric defect image classification is realized through the trained classifier according to Bayesian probability fusion, and a fabric defect image classification result is output. The detection and classification method provided by the invention has a higher accuracy rate.

The invention provides a real-time visual object tracking method having shielding sensing mechanism, tracking results based on filter tracking algorithm based on correlation and Bayesian probability model tracking algorithm based on color statistical information are fused, position prediction of a target is performed, one-dimensional scale pyramid is independently trained to perform target scale estimation, in addition, a shielding sensing mechanism is designed, the mechanism can judge whether a target is partly or completely shielded, then an update strategy of the model and which kind of model utilized to track new frame are determined, tracker drifting caused by pollution of background information is avoided, and according to the above steps, the tracking algorithm is used for tracking the target persistently and iteratively. The above algorithm can be used for accurately tracking the target in real time, and a plurality of problems including deformation, rotation, shielding, size variation, illumination variation, motion blur and the like can be effectively handled.

The invention discloses a scene text identification method based on a Bayesian probability frame, and relates to vision and mode identification of a computer. The scene text identification method based on the Bayesian probability frame comprises the steps that S1, scene image text is input; S2, character detection and recognition are carried out; S3, a detection-identification candidate grid is built, particularly, a candidate character area, a corresponding character class and a corresponding identification score are stored in the detection-identification candidate grid, each detection-identification route in the detection-identification candidate grid corresponds one text detection and identification result, and a route evaluation function is designed to evaluate each detection-identification route in the candidate grid; S4, the best detection-identification route is searched for from the candidate grid through a dynamic planning algorithm, that is to say, the identification result is obtained; S5, the text identification result is output. The scene text identification method based on the Bayesian probability frame achieves probability modeling and parameter learning of scene text identification integrating detection and identification.

A controller receives outputs form a plurality of sensors such as a camera, LIDAR sensor, RADAR sensor, and ultrasound sensor. Sensor outputs corresponding to an object are assigned to a tracklet. Subsequent outputs by any of the sensors corresponding to that object are also assigned to the tracklet. A trajectory of the object is calculated from the sensor outputs assigned to the tracklet, such as by means of Kalman filtering. For each sensor output assigned to the tracklet, a probability is updated, such as using a Bayesian probability update. When the probability meets a threshold condition, the object is determined to be present and an alert is generated or autonomous obstacle avoidance is performed with respect to an expected location of the object.

The invention provides an optimal protection policy analysis system and method based on an improved ant colony algorithm. The system comprises a network risk assessment subsystem, an attack path prediction subsystem and a security policy management subsystem. The method comprises the following steps: 1) importing a Bayesian probabilistic model to describe the complex relationship between multiple steps of atomic attacks, and constructing an attack diagram by using the Bayesian belief network to evaluate a network security risk; 2) simulating a decision making process of an attacker in combination with the quantitative indexes of the protection cost and the gain, and predicting a possible attack path; and 3) using the improved ant colony algorithm capable of calculating the quasi-optimal solution within a polynomial time to solve the security policy selection problem on the attack path. By adoption of the optimal protection policy analysis system and method, a risk solution can be provided for a network security administrator in time, the attack path prediction and protection strategy selection problems in network security protection are solved, and the security of the whole network under the condition of limited protection resources is effectively protected.

The invention discloses a multi-document automatic abstract generation method based on phrase subject modeling. Multiple sample documents are subjected to word segmentation to obtain phrases and frequency of occurrence of the phrases, and the documents are expressed in the form of a phrase bag; joint probability distribution of the documents is calculated on the basis of an LDA subject model, the LDA subject model is converted into a phrase subject model, then a Gibbs sampling algorithm is used for estimating implicit parameters in the phrase subject model according to Bayesian probability, and lastly probability distribution of the subject in words is obtained; the tested documents are subjected to word segmentation, the subject weight and word frequency weight of obtained sentences are calculated and obtained, the final weight of the sentences is obtained by means of weighting calculation, and abstract content is generated according to the final weight. The method is more standard and precise, the relationship between different words is taken into consideration, the subject weight of the sentences is introduced, and the generation result better conforms to the practical essay abstract writing conditions of people after the subject weight of the sentences is introduced.

The invention discloses a method and a system for deducing a sudden event situation based on a case. The method comprises the steps of generating a case script of the case, abstracting the case script, extracting a case event link and storing the case event link in a server database; by a client, obtaining the case event link from the server database, analyzing the case event link, generating a logic relationship and specific information of the event in the case, generating a scalable vector diagram of the event based on the logic relationship and the specific information and adding the scalable vector diagram to a page of a webpage; and adjusting the scalable vector diagram based on the logic relationship, finishing the showing of the sudden event situation by combining with a visualization technology, and finishing deduction of the event automatically by adopting a Bayesian probability statistics principle and by combining with the internal logic probability of the event. According to the method and the system for deducing the sudden event situation based on the case, a visual showing and editing function of the event link is provided, and assistance is offered for non-specialized persons to edit and drill the script.

A method and system for truth discovery may implement a methodology that accounts for accuracy of sources and dependency between sources. The methodology may be based on Bayesian probability calculus for determining which data object values published by sources are likely to be true. The method may be recursive with respect to dependency, accuracy, and actual truth discovery for a plurality of sources.

A barcode reader and processing circuitry are disclosed for identifying a position of a cylinder rod, for example, along a path of movement. The barcode is marked on a surface of the rod and constitutes a series of markings extending along a portion of the length of the rod. The markings represent encoded information based on numbers calculated with a random number generator, and are grouped into subsets, each of which corresponds to a particular rod position. A portion of the rod including a subset of barcode markings is exposed to light emitted from the reader. Light reflected off the barcode is sensed with an array of photosensors provided in the reader. The photosensors generate electrical signals, and using Bayesian probability, a likely subset of markings is determined corresponding to the received electrical signals. Once identified, the markings subset is associated with a location corresponding to the position of the rod. A barcode reader is also described that is relatively easy to assemble and provides uniform illumination of the barcode.

The invention discloses a personalized information retrieval method and device. The method comprises generating retrieval results according to an assigned user's retrieval task; subjecting each query term in the user's retrieval task and each query word in each query term to Bayesian probability matrix decomposition to obtain query-document relevancy; acquiring the user's long-term behaviors and short-term behaviors, and acquiring user-document attention according to the user's long-term behaviors, short-term behaviors and behavior weighted parameters; generating the user's sorting weighted parameters, and determining relevancy of each document in the retrieval results according to the query-document relevancy, user-document attention and sorting weighted parameters; sorting the documents in the retrieval results according to the relevancies of the documents in the retrieval results, and returning to the user. The personalized information retrieval method and device have the advantages that the accuracy of a personalized information retrieval sorting model can be improved and retrieving may be independent of specific contents of query and documents.