33results about How to "Improve prediction success rate" patented technology

The invention relates to oil-gas prospecting technical field, especially using the petroleum inclusion and asphalt to distinguish the oil reservoir and fulfill the migration tracing. With the statistical data from the petroleum inclusion and asphalt, the invention may distinguish the oil reservoir during the drilling (including the prospecting and development) or later; and determine the petroleum migration direction, and trace its routing, search the prevailing migration channel etc. under multi-stage migration and accumulation, fulfill the multi-stage and multiple genesis statistics for petroleum inclusion, trace the oil-gas migration during different geologic phase, determine the oil-gas migration direction and prevailing channels during different stages, and the variation of physical and chemical conditions during migration, thereby recovers the history of oil-gas migration and accumulation, provides the important basis for finding the oil-gas distribution law and advantageous areas for oil-gas migration and accumulation.

The invention discloses a method and device for building a model and pushing a message. The method for building the model comprises the following steps: obtaining user behaviour information of a user accessing a preset application in recent K days and K+1 days; constructing a training set according to the user behaviour information of the user in recent K days and K+1 days; and building a predication model according to the training set, wherein the predication model is used for determining the pushing time of the user about the preset application in the N+1th day according to the user behaviour information of the user accessing the preset application in the recent N days; and both K and N are preset parameters, which are integers more than 1. By means of the technical scheme, aiming at different users, the pushing message can be sent in a customized manner; and the opening rate for the user to open an application programme or the pushing message can be increased.

The invention provides a method for predicting a period of re-enriching remaining oil of a water flooding reservoir into a new reservoir. The method comprises the following steps of (1) gaining a value of water saturation of a researched oil reservoir; (2) selecting a corresponding oil and water delivery speed formula of the water flooding reservoir according to the value of the water saturation; and (3) calculating the maximum time of re-enriching the remaining oil of the water flooding reservoir into the new reservoir, namely the period of re-enriching the remaining oil of the water flooding reservoir into the new reservoir, by using the selected oil and water delivery speed formula of the water flooding reservoir. The method for predicting the period of re-enriching remaining oil of the water flooding reservoir into the new reservoir points out the direction for strategic adjustment of low-efficiency oil reservoir development at an ultra-high-water content stage, periodic oil reservoir injection-production development, capacity recovery and re-development of an oil reservoir which is about to be abandoned, and the like. Therefore, the method is wide in popularization and application prospect and high in economic benefit and social benefit.

The invention discloses a Web service quality prediction method based on random walk. The Web service quality prediction method comprises the following steps: on the basis of a user position, calculating a distance between users; selecting adjacent users for each user to form a neighborhood, and connecting neighborhood users to construct a user relationship network; calculating a similarity between the users and the similarity between services in the user relationship network; calculating the weight of each edge in the user relationship network; aiming at a target service requested by a source user to independently look up the random walk about the target service i for multiple times from the source user, randomly selecting a next-hop node by each hop in each-time random walk according to a certain probability, returning a QoS value after each-time walk is finished; and synthesizing all QoS values to calculate a quality prediction value, which is relative to the target service, of the source user. The Web service quality prediction method can balance coverage rate and the prediction precision of Web service quality prediction and has the advantages of being accurate in prediction, high in success rate, good in universality and favorable in evaluable confidence coefficient and interpretability.

The invention discloses a method for predicating drug-target combination based on the grey theory and molecular fingerprints. The method includes generating protein pseudo amino acid compositions based on a gray theory gray model (GM) (1,1) and converting target protein sequences into 21-dimensional spatial vectors combined with the protein sequence pseudo amino acid compositions; describing drug molecules into a 256-dimensional spatial vector through a drug molecular fingerprint software; combining the 21-dimensional spatial vectors of the protein sequences and the 256-dimensional spatial vector of the drug molecules into 277-dimensional spatial vectors to serve as drug-target combination descriptors; and training sets are trained through a fuzzy K-nearest neighboring method to obtain optimum parameters of a predictor and the drug-target combination descriptors are input into the predictor to predicate whether a drug is associated with a target. According to the method, three-dimensional structures of proteins are not needed to be measured, whether the drug and the proteins can be combined can be predicated by only adding drug molecular fingerprints on a protein one dimensional sequence, and the predication success rate is high.

The invention relates to a data correcting method applying a petroleum inclusion to carry out oil and gas migration tracking and ancient and present oil layer recognition, which is characterized in that for the petroleum inclusion of each cement, a correction formula of GOI and FOI statistic data is disclosed in the specification, wherein GOImc is the relative number of a corrected single cement (such as silicious cement) and an oil-bearing inclusion in a fissure thereof, and the GOIm is the original data; Cm represents the content of the cement and is an average value; and CC represents the content of carbonaceous asphalt and is the average value of the content of the carbonaceous asphalt. Similarly, the correction of FOI can be realized, the FOImc is the frequency of occurrence of the m cement after correction and the oil-bearing inclusion in the fissure thereof, and the FOIm is the original data. Microscopic observation is carried out on ordinary sheets or casting sheets, the contents of the cement and the carbonaceous asphalt corresponding to the petroleum inclusion are counted, and the number of statistical points is between 100 and 1000; and a fluorescence microscope is utilized to observe the dissolution phenomenon of the carbonaceous asphalt by oily asphalt so as to identify the carbonaceous asphalt, and the formula (2) and the formula (3) are utilized to correct the GOI and the FOI.

The invention discloses an accurate digital indicator reading detection method in the technical field of mode recognition and artificial intelligence. The method comprises the following steps of: (1)data obtaining: shooting an instrument image which comprises a reading area by adoption of a camera device; (2) data processing: manually annotating an instrument reading value and carrying out randomrotation, stretching and translation transformation on the instrument image; (3) deep network model construction and training: inputting the instrument image and corresponding annotation data into adeep network model to carry out training; and (4) instrument reading detection: inputting the instrument image and returning a standard instrument reading image by a system. The method is capable of overcoming deficiencies of existing instrument reading extraction methods, sufficiently utilizing parameter learning ability of the deep network model, and accurately detecting digital indicator readings through distribution of learning data samples on the basis of adversarial learning ability of the deep network model and physical meaning of backpropagation residuals, has the characteristics of being strong in practicability and high in correctness, and has a relatively good practical value.

The invention relates to a soil adsorbed hydrocarbons cryogenic temperature desorption apparatus and its method.The method comprises following steps: injecting water without gas into the apparatus; mixing water and soil to mud absorbed into the apparatus; then injecting water without gas and adjusting temperature head to normal temperature and normal pressure; collecting desorption gas sample and detecting them in gas-chromatography.

The invention provides a sandstone bottom water reservoir water cone quantitative-description method. The sandstone bottom water reservoir water cone quantitative-description method comprises the steps of 1 collecting an oil reservoir characterization parameter of a researched oil reservoir, 2 describing a model formula by selecting a corresponding water cone or a water ridge according to well type and 3 utilizing the selected water cone or the water ridge to describe the model formula and calculate the morphological parameter and volume of the water cone or the water ridge. By means of the sandstone bottom water reservoir water cone quantitative-description method, the morphologies of the water cone or the water ridge of the oil reservoir at different position in different periods can be predicted, and the residual oil distribution rule of the sandstone bottom water reservoir can be determined. The sandstone bottom water reservoir water cone quantitative-description method has the advantages of being high in prediction success rate, simple and convenient to used and practical.

The invention relates to the field of computer components, in particular to a six-level assembly line processor based on an RISC-V instruction set, which comprises a six-level assembly line, a first register, a data memory and an instruction memory, and is characterized in that an instruction fetching module, a decoding module, a distribution module, a transmitting module, an execution module and a write-back module of the six-level assembly line are connected in sequence; the data memory is connected with the execution module and the write-back module, the first register is connected with the distribution module and the write-back module, and the instruction memory is connected with the instruction fetching module. The characteristic of out-of-order execution of double-instruction decoding emission is achieved, and a reference model is provided for superscalar parallel design of the RISC-V processor. And a Gshare branch prediction circuit is provided, so that the branch prediction success rate is improved. The register is renamed, so that assembly line cavitation caused by reading after writing and writing after writing is remarkably reduced. And a complete branch prediction failure repair circuit is provided, so that the time loss caused by branch prediction failure is reduced.

The invention discloses a crop cultivation mode analysis optimization method with a real-time database and an advanced control algorithm. Physiological factors and environmental factors of crop cultivation are introduced, and through a canonical correlation analysis method and a support vector machine, environmental parameters and physiological parameters are analyzed and processed to obtain an optimal parameter combination; sample data corresponding to the optimal parameter combination are input to the support vector machine to be trained, a crop cultivation model capable of predicting the crop growing state is built, and the obtained crop cultivation model is utilized to optimize a crop cultivation mode. The crop cultivation mode analysis optimization method with the real-time database and the advanced control algorithm simplifies the crop cultivation mode, can solve the problem in the prior art that correlation calculation is too complicated, can accurately predict the growth of crop that is cultivated, and optimize a crop cultivation process.

The invention discloses a concealed resource prediction method and a rock electromagnetic logging system which are applied to the technical field of mineral and oil-gas resource exploration. Accordingto the rock electromagnetic logging system, by the adoption of a broadband (0.001-10,000 Hz) sweep frequency mode, rock broadband complex impedance measurement can be realized, and potential mineralor oil-gas resources are predicted through acquired rock electromagnetic characteristic factors separately or in combination with geochemical characteristics.

The invention discloses a logarithm normalization method for improving local prediction ability of a neural network. The method comprises the following steps of: determining input and output of the neural network according to a prediction object, specifically, generating an input value of a training sample as a neural network input value through a sampling method, and calculating an output value of each training sample through a computational fluid mechanics method, namely, designing a target value, as a neural network output value; performing normalization for the neural network input value through a linear normalization method; determining a threshold value of the neural network output value, and performing normalization for the neural network output value through a logarithm normalization method. Compared with the conventional linear normalization method, the logarithm normalization method provided by the invention improves prediction success rate of the neural network by 17.1%.

The invention provides a conditional jump instruction processing method. The method comprises the following steps: determining an instruction fetching address, and fetching a target instruction from an instruction cache according to the instruction fetching address; analyzing the target instruction to obtain an instruction type; if the instruction type is a conditional jump instruction, controlling the conditional jump instruction to jump according to a preset jump rule; when the conditional jump instruction jumps normally according to the preset jump rule, sending corresponding instruction information to a first cache; and when the conditional jump instruction jumps abnormally not according to the preset jump rule, sending the corresponding instruction information to a second cache. According to the invention, the problem that a dynamic prediction method is easily influenced by an algorithm, resources, a compiler and application adaptation can be solved, and the instruction execution efficiency of a processor is improved. The invention further provides a conditional jump instruction processing system, a computer readable storage medium and electronic equipment, which have the above beneficial effects.

The invention provides an RNA sequence description method that combines local and global features, constructs an RNA sequence physicochemical matrix PCM by combining nucleotide duplex structure information and other physical and chemical properties, and the matrix contains RNA sequence structure information and sequence order information. By establishing a gray model for each row of data in the matrix and calculating its autocorrelation factor and cross-correlation factor between rows to obtain the local information of the sequence, convert the PCM into a two-dimensional image, and calculate the geometric distance of the image to obtain the global sequence information. Combining the obtained factors describing sequence local information and global information, a discrete vector description model of RNA sequence is obtained. The method is used in the prediction of RNA attribute types and RNA modification sites, can significantly improve the prediction success rate of related predictors, and has broad application prospects.