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

The invention discloses a risk assessment method for unqualified products, and the method comprises the following steps: (1) carrying out the data preprocessing, data screening and feature extraction of pre-collected product large-class sample data, and obtaining feature parameters corresponding to the unqualified products; (2) randomly segmenting data of all samples into a training set and a verification set according to a proportion of 7: 3; (3) establishing a Logistic regression model according to the characteristic parameters, evaluating the Logistic regression model by adopting an ROC curve and an AUC value, and calculating an optimal truncation value of an unqualified product probability according to a Youden index; and (4) inputting the characteristic parameters of the to-be-tested product into the Logistic regression model to obtain the disqualification condition of the to-be-tested product. The method is accurate in prediction result, high in speed and good in adaptability to a large amount of detection data.

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 discloses a post-drilling data processing method. The method comprises the following steps: data acquisition is carried out, a pre-drilling calculation analysis result for an exploratory well is acquired, and post-drilling data of the exploratory well are acquired; post-drilling data are pretreated, the post-drilling data are calculated and analyzed to acquire a post-drilling calculation analysis result corresponding to the pre-drilling calculation analysis result; comparative analysis is carried out, based on the comparative analysis on the pre-drilling calculation analysis result and the post-drilling calculation analysis result, a comprehensive analysis processing result for the post-drilling data is acquired, wherein the comprehensive analysis processing result comprises rationality judgment on the pre-drilling calculation analysis result and / or reasons for deviation of the pre-drilling calculation analysis result. The method of the invention can reduce uncertainty of the pre-drilling geological data calculation analysis result, the geological state prediction success rate can be improved, and the oil-gas exploration risks are finally reduced.

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 mobile robustness optimization method and device based on mobile state information and adjacent region proximity and belongs to the technical field of wireless communication. In the method, a micro base station controller estimates real-time mobile state information of mobile UE according to cyclic measuring reports obtained from the mobile UE; then the mobile UE is subjected to switching control according to an adjacent region deploying environment, and frequent switching between a macrocell and a microcell of the mobile UE is avoided; and finally, during a switching process, switching preparing operation to a preparing target cell is brought forward, and according to the moving speed of the mobile UE and the PDCCH interruption point of the service microcell, the switching condition for avoiding ping-pong switching is computed. Compared with the prior art, the prediction success rate of the preparing target cell is improved, frequency spectrum resource using rate is improved, frequent switching between the macrocell and the microcell of the mobile UE is lowered, then network signaling expenditure is reduced, SToS rate is not increased, meanwhile, the switching success rate of the mobile UE can be improved, and then user experience is improved.

The present invention relates to the technical field of obstetrics and gynecology, in particular to a method for predicting the success rate of trial delivery of a scarred uterus, by analyzing the prenatal parity of pregnant women, prenatal body mass index, dilation of the cervix, and premature rupture of membranes. Data collection, and then use P=1 / [1+exp (‑0.785+1.3259x1‑1.5061x2‑1.2449x31‑1.4211x32+0.7617x4)]*100% to calculate, and predict the success rate of vaginal trial delivery according to the P value , to provide a reference index for pregnant women and doctors, data collection is easy, and the prediction success rate is high, especially suitable for the prediction of vaginal trial labor of pregnant women with scarred uterus in our country.

The present invention provides a new solution to the imbalance problem of the protein training set fused with genetic information. The amino acids in the conserved region in the protein P sequence are kept unchanged, and the amino acids in the non-conserved region are sequentially converted according to the probability of mutation of the PSSM matrix to other amino acids. into other amino acids, so that 20 virtual proteins containing the genetic information of protein P can be obtained, and the small number of subsets in the unbalanced data set can be expanded to make the unbalanced data set into a balanced data set, which is conducive to training related predictors , which can improve the prediction success rate of the predictor. Different from the existing methods for solving unbalanced data, the present invention can integrate protein evolution information and directly expand from the sequence instead of interpolating in the discrete digital model describing the sequence information, which has obvious biological significance, so it can significantly improve The prediction success rate of related predictors has a wide application space.

The invention relates to a novel protein sequence representation method based on gene ontology information, comprising: using BLAST program to search Swiss-Prot database for all similar protein sequences of protein sequence P, inputting all proteins in a training dataset into GO (gene ontology) database, and searching for GO information of each protein; searching the gene ontology database for targeting gene ontology information of P protein; defining the P protein as discrete vectors of M elements according to M labels that a prediction problem has. The protein GO information in a sequence set is fused into novel P protein vector description, and the dimensionality of the GO method is reduced greatly; by applying the method to protein subcellular multi-label positioning prediction and antibacterial peptide functional multi-label prediction, it is possible to significantly increase the prediction success rate of a related predictor; the novel protein sequence representation method based on gene ontology information has a promising application prospect.

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 ray impedance inversion method orienting an angle partial stack gather and a device thereof, and relates to the technical field of geophysical prospecting for petroleum. The method comprises the steps that a ray impedance equation is established based on the start angle and the end angle of the ray impedance inversion angle partial stack gather; a ray impedance pseudo well curve is obtained based on longitudinal wave velocity, transverse wave velocity, a density logging curve and the angle range of the angle partial stack gather through the ray impedance equation; angle wavelets are obtained based on the angle partial stack gather and the ray impedance pseudo well curve; a constraint model reflecting the stratum form is obtained according to the angle partial stack gather, stratum information and the ray impedance pseudo well curve; model constraint sparse pulse inversion is performed according to the angle wavelets, the angle partial stack gather and the constraint model so that the ray elastic impedance data of different angle ranges are obtained; and longitudinal wave impedance and transverse wave impedance are obtained based on the ray elastic impedance data through extraction. The precision of the inversion result can be effectively enhanced.