32 results about "Gibbs sampling algorithm" patented technology

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 relates to a network safety log template extraction method and device. The method comprises steps: firstly, original safety logs are subjected to data cleaning, and log information after filtering of time and IP addresses is obtained; secondly, the log information which does not containing time and IP addresses are subjected to clustering, and logs with similar formats are classified to the same class; thirdly, for logs in the each class, template words describing formats in the logs are extracted, and the template of the log format is obtained. Concretely, a DBSCAN algorithm or an OPTICS algorithm is employed to cluster logs with similar formats, and template words describing formats in logs are extracted by utilization of an LDA Gibbs sampling algorithm. The device comprises a data cleaning unit, an information clustering unit and a template extraction unit. Priori knowledge is not needed, a template of a network safety log format can be obtained automatically, the system load can be reduced, and operational efficiency and accuracy are raised.

The invention discloses an improved-Gibbs-sampling-based Beidou satellite selection method comprising the following steps: calculating an azimuth and an elevation angle of a Beidou visible satellite at a monitoring point to obtain state matrixes of n Beidou satellite, and using a GDOP value as an objective function of Beidou satellite selection; and on the basis of an optimized Gibbs sampling algorithm, searching for a Beidou satellite combination enabling the objective function value to be optimized to obtain an optimal solution. On the basis of the adaptive sweeping type improvement strategy, a phenomenon of falling into the local optimization of the algorithm is avoided and convergence to the optimal Beidou satellite selection scheme is realized well. According to the invention, the optimized Gibbs sampling algorithm is applied to the Beidou satellite selection process and the inherent advantages, including low complexity, fast convergence speed and good performance of the Gibbs sampling algorithm are utilized fully, so that the quality and efficiency of the Beidou satellite selection are improved effectively. The improved-Gibbs-sampling-based Beidou satellite selection method has the broad application prospects.

The invention discloses a knuckle recognition method based on an infinite Dirichlet process mixed model, which is carried out according to the following steps: step 1, on the basis of the local Markov assumption, the learning problem of conditional random measure is transformed into a random clustering learning problem; step 2. Use the infinite Dirichlet process mixture model to describe the probability density, and express the number of clusters as a random state; step 3, use the Gibbs sampling method to iteratively learn the density structure in the form of hierarchical probability; step 4, based on the Dirichlet process mixture model The collapsed Gibbs sampling algorithm DPMM uses the sample set for model training and learning, and uses a fixed threshold to recognize the knuckles of the hand image. The invention refines the description of the biological structure of the hand, the detection result is stable and reliable, and the calculation efficiency is high.

The present invention relates to a small sample reliability evaluation method based on a Bayesian theory. The method comprises the following steps: (1) performing system initialization, and inputtingpre-test information; (2) classifying the pre-test information; (3) analyzing an existing conversion method of similar system information, and using the D-S evidence theory and the F-HS based algorithm to perform analysis respectively; (4) using a hybrid pre-test distribution model; (5) determining a Bayesian reliability model of field test information; (6) integrating the field test information with the pre-test information through the Bayesian method to obtain an effective distribution model; (7) using a Gibbs sampling algorithm to obtain a sample value of a post-test distribution function;and (8) performing evaluation and estimation on reliability parameters to obtain estimated values of the reliability parameters. The small sample reliability evaluation method based on the Bayesian theory proposed by the present invention shows the superiority in both the effect of pre-information conversion and the accuracy of small sample reliability evaluation.