Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

460 results about "Stratified sampling" patented technology

In statistics, stratified sampling is a method of sampling from a population which can be partitioned into subpopulations. In statistical surveys, when subpopulations within an overall population vary, it could be advantageous to sample each subpopulation (stratum) independently. Stratification is the process of dividing members of the population into homogeneous subgroups before sampling. The strata should define a partition of the population. That is, it should be collectively exhaustive and mutually exclusive: every element in the population must be assigned to one and only one stratum. Then simple random sampling or systematic sampling is applied within each stratum. The objective is to improve the precision of the sample by reducing sampling error. It can produce a weighted mean that has less variability than the arithmetic mean of a simple random sample of the population.

Stratified sampling device for applied statistics

The present invention relates to the technical field of stratified sampling devices for applied statistics, and particularly to a stratified sampling device for applied statistics. The stratified sampling device comprises: a sampling device, wherein a plurality of water intake pipes are inserted into the periphery of the sampling device, the sampling device is provided with a closed cavity in the middle of each water intake pipe, the sampling device is provided with a mounting cavity at the bottom of the closed cavity, an electric push rod is provided in the mounting cavity, and the electric push rod extends out of the mounting cavity and is then connected to a closing plate in the closed cavity; and a support column, wherein a lead screw is provided in the middle of the support column, a connecting block is slidably connected on the lead screw, and the other end of the connecting block is connected to the sampling device. The beneficial effect lies in: in the stratified sampling device for applied statistics provided by the present invention, after sampling bottles are mounted, a lead-screw electric motor at the top of the support column starts to drive the lead screw to rotate, the connecting block on the lead screw thus drives the sampling device to ascend and descend in water, and according to the depth of the water, the electric push rod starts to drive the closing plate in the closed cavity to retract, thereby enabling the closing plate to open the water intake pipe.
Owner:CHONGQING COLLEGE OF FINANCE ECONOMICS

Target detection tracking method and system based on automatic detection

The invention discloses a target detection tracking method and system based on automatic detection, and relates to the technical field of target detection tracking, and the method comprises the following steps: carrying out the target position and category labeling of an image through a labeling tool based on an image data set of a typical target, and obtaining a training data set with a label, establishing a target detection model according to a YOLO architecture, and introducing a dynamic weighted loss function; according to the training data set, a training batch is generated by using a hierarchical sampling strategy, an optimized training sample is obtained through an online difficult case mining mechanism, and iterative training is performed on the target detection model based on the generated training batch and the optimized sample; and performing target detection on the real-time video stream by using the trained target detection model to obtain the tracking position and category information of each frame. According to the method, the problem of class imbalance can be effectively solved, the detection capability of rare targets is improved, and meanwhile, the overall performance of the model is improved through difficult case mining.
Owner:HEFEI HEZHIXIN MICROELECTRONICS CO LTD

Method for monitoring and evaluating carbon sequestration capability of island ecosystem

The invention relates to the technical field of carbon sequestration capacity monitoring, in particular to a carbon sequestration capacity monitoring and evaluating method for an island ecosystem, which comprises the following steps: S1, multi-source data fusion and habitat partitioning: dividing an island habitat type partitioning map; s2, stratified sampling and carbon reserve actual measurement: measuring the carbon content, and constructing a carbon reserve actual measurement database; s3, constructing a dynamic carbon sink model: carrying out a whole growth cycle environment simulation experiment on sea grass and algae in different habitats, and establishing the dynamic carbon sink model taking habitat types as classification units; s4, carbon sequestration capability quantification and bottleneck diagnosis: calculating carbon sequestration contribution values of various habitat types, identifying low-efficiency carbon sequestration areas through contribution value comparison, and analyzing environmental limitation factors of the low-efficiency carbon sequestration areas; and S5, integrating a sink increasing technology and generating a path: designing a habitat specific sink increasing scheme for the identified low-efficiency carbon sink area. According to the invention, high-precision monitoring, intelligent evaluation and system optimization of the island ecosystem carbon sequestration capability are realized.
Owner:OCEAN UNIV OF CHINA +1

Offshore red tide multi-parameter real-time monitoring method and monitoring system based on buoy array

The invention discloses an offshore red tide multi-parameter real-time monitoring method and monitoring system based on a buoy array. The method comprises the following steps: S1, performing array deployment and topology configuration according to a sea area sensitivity degree, tide characteristics and a historical red tide trajectory; according to the invention, through risk differentiation buoy array deployment and redundancy topology, traditional isolated monitoring limitation is solved, and sea area full coverage and red tide diffusion tracking are realized; physical-chemical-biological multi-source sensing and in-situ detection are combined with data fusion, so that the problem of single parameter is solved, and the monitoring accuracy is improved; by means of liftable stratified sampling and in-situ algae species and toxicity analysis, the vertical monitoring blank is filled up, and the complete life cycle process of red tide from occurrence to regression is completely captured; through linkage of edge-cloud multi-model early warning and field equipment, a'monitoring-early warning-disposal 'closed loop is formed, and the red tide monitoring real-time performance and the emergency response capability are greatly improved.
Owner:WEIHAI ZHENGMING SEA TECH DEV CO LTD

Artificial intelligence auxiliary antenna design method based on Gaussian regression process

The invention provides an artificial intelligence auxiliary antenna design method based on a Gaussian regression process, and the method comprises the steps: employing a script interface to interact with a simulation platform, automatically completing the parameterized modeling process of an antenna model, and setting an excitation condition and a radiation boundary; a multi-dimensional space stratified sampling method is adopted to generate a uniformly distributed sample point set in the multi-dimensional parameter space, and electromagnetic simulation is carried out; preprocessing the data output by simulation to form a data structure suitable for machine learning model processing; integrating the trained MLNN model into a grey wolf optimization algorithm, and optimizing antenna design parameters; and performing electromagnetic simulation on the optimal parameter obtained by optimization, comparing a prediction result with actual simulation data, and verifying the performance of the optimization engine in the target frequency band. According to the method, the problems of insufficient sample generation efficiency, insufficient parameter sensitivity analysis, limited model generalization ability, low convergence speed of an optimization algorithm and the like in the prior art can be solved.
Owner:BEIJING INSTITUTE OF TECHNOLOGY (ZHUHAI) +1

Downstream analysis feedback-oriented data cleaning method and system and electronic equipment

The invention discloses a downstream analysis feedback-oriented data cleaning method, which comprises the following steps of: constructing a multi-layer perceptron classifier to execute classification, and generating a soft label; dirty data is combined with a soft label to input an error detection model, a preliminary error detection result in a feature vector form is output, and a clustering result is obtained through clustering weighting; selecting partial data for training, inputting a plurality of isomorphic error detection models after stratified sampling, and voting an integration result; processing error detection output and dirty data by using a neural network repairing model of a cross attention mechanism, and outputting correction data through feature extraction and optimization fusion; and the downstream model feeds back an optimization repair process to obtain final cleaning data. According to the method, data is detected mainly through a method based on deep learning, dirty data in the data are restored, and the data are more accurate and credible. According to the method, customization processing can be carried out on the data according to the requirements of downstream tasks, and it is ensured that the cleaned data can be directly applied to the tasks.
Owner:ZHEJIANG UNIV

Sampling device for water resource quality detection

The invention relates to the field of water resource quality detection, in particular to a sampling device for water resource quality detection. The invention provides a water resource quality detection sampling device which comprises a sampling barrel, a barrel cover is arranged at an opening of the sampling barrel, clamping seats are symmetrically distributed on one side, close to the barrel cover, of the sampling barrel, clamping pins are arranged in the clamping seats, the barrel cover is locked through cooperation of the clamping pins and the clamping seats, and the sampling barrel is connected with the clamping seats. A plurality of sampling box bodies for containing water samples are vertically arranged in the sampling barrel, a water pump is arranged in the sampling barrel, and a water inlet pipe of the water pump extends to the outer side of the sampling barrel. The stratified sampling box body and the independent flow guide assembly are combined with the servo motor to drive the staggered box cover, so that accurate stratified collection of water samples at different depths is realized, and mixed pollution is avoided; a water flow switching assembly is additionally arranged, a pipeline is automatically flushed after sampling, a residual water sample is discharged into a liquid storage box, cross contamination is reduced, independence and representativeness of the water sample are guaranteed, and detection accuracy is improved.
Owner:SHENBIAO LAB (SHENZHEN) CO LTD

High-adaptability multi-type geological exploration sampling device and accurate sampling method

The invention belongs to the technical field of geological exploration, particularly relates to a high-adaptability multi-type geological exploration sampling device and an accurate sampling method.The high-adaptability multi-type geological exploration sampling device and the accurate sampling method aim to solve the problems that a drill bit is stuck and stratified sampling is difficult, and the high-adaptability multi-type geological exploration sampling device comprises a base, a vertical plate, a lifting plate and a driving cylinder with the drill bit; a first annular groove is formed in the outer wall of the driving cylinder and matched with the first circular ring, downward-pressing drilling and upward-lifting resetting of the drill bit are achieved, a second annular groove is formed in the drill bit, and the flow direction of hydraulic oil is controlled through a second circular ring connected with a fixing rod; during lifting, hydraulic oil is injected into the second hydraulic channel, the trapezoidal collecting box extends out to scrape hole wall samples, the electromagnet and the iron block are matched to adjust the position of the second circular ring, the working states of the teeth and the collecting box are switched, drill bit jamming is reduced, hole wall sampling at any depth is supported, and exploration efficiency is improved.
Owner:四川省地质矿产勘查开发局攀西地质队

Long-time-sequence remote sensing crop classification method

The invention discloses a long-time-sequence remote sensing crop classification method. The method comprises the following steps: extracting spectral features of a to-be-detected region in different periods; constructing an NDVI phenological characteristic curve of the typical crops; judging the crop category based on the similarity between the curves, and generating a crop classification map of 10 meters of the to-be-detected region; performing upscaling and stratified sampling processing on the crop classification result of the sentinel No.2 satellite, and constructing training and verification samples of Landsat satellite crop classification; carrying out space-time fusion on the Landsat multiband spectral image, and constructing crop supervision classification input based on a characteristic spectrum of the synthesized image and a derived vegetation index; and constructing a crop classifier of a Landsat satellite based on the random forest model, and generating a crop classification map of the to-be-detected region. According to the method, crop classification is carried out by fully utilizing the advantages of high resolution of the sentinel No.2 satellite and long sequence of the Landsat satellite, and long-sequence crop classification drawing of the to-be-detected area can be realized.
Owner:INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS +2

UWB-NLOS identification and ranging correction method based on DEA-PSO hybrid optimization LightGBM algorithm

The invention provides a UWB-NLOS identification and distance measurement correction method based on a DEA-PSO hybrid optimization Iight GBM algorithm, and the method comprises the following steps: 1, building multi-dimensional features after data stratified sampling, obtaining an HCNLOS data set through two times of label screening, and generating initial positioning; step 2, adopting a DEA-PSO hybrid optimization algorithm to realize LightGBM hyper-parameter automatic adjustment and optimization, and improving parameter adjustment efficiency through dynamic fitness evaluation and collaborative search; 3, constructing a classification model and a regression model in parallel; and 4, fusing the corrected NLOS data and the original LOS / HCNLOS data, and outputting a high-precision coordinate through a positioning algorithm, thereby effectively improving the positioning robustness in a complex environment.
Owner:LIAONING TECHNICAL UNIVERSITY

Soil stratified sampling device for hydrogeological survey

The invention discloses a soil stratified sampling device for hydrogeological survey, and relates to the field of geological survey, the soil stratified sampling device comprises a support frame, and four universal wheels with brakes are symmetrically and movably mounted on the bottom side of the support frame. When the soil sampling device is used, a driving motor and a lifting motor are started, a connecting drill bit rotates and descends to drill into soil, a first sampling pipe, a second sampling pipe and a third sampling pipe are driven to be located at different depths in the process, and a sampling scraper blade is in a closed state, so that the soil at different depths can be prevented from being mixed into a sampling groove; the lifting motor is closed, the sampling motor at the corresponding position is started, the sampling scraper is unfolded and makes contact with surrounding soil, at the moment, the sampling scraper can be driven to scrape the surrounding soil under the action of the driving motor, the soil falls into the sampling groove or the sampling baffle, and then the sampling scraper is controlled to retract into the sampling groove. Meanwhile, the soil can be collected into the sampling groove, and sampling of soil of different layers is completed.
Owner:CHINA COAL ZHEJIANG SURVEY & DESIGN CO LTD

Farmland soil heavy metal inversion method fusing spectrum layering and machine learning

The invention discloses a farmland soil heavy metal inversion method fusing spectrum layering and machine learning, and belongs to the technical field of image processing. The method comprises the following steps: acquiring a to-be-measured spectral variable of soil in a survey area; and inputting the to-be-detected spectrum variable into the soil heavy metal inversion model, generating a heavy metal prediction concentration spatial distribution diagram of the investigation area output by the soil heavy metal inversion model, and constructing the soil heavy metal inversion model based on the calibration sample set and the verification sample set. The calibration sample set and the verification sample set perform stratified sampling determination on a plurality of training samples based on soil component information of sample soil, and the training samples comprise important spectral variables and heavy metal concentrations of the sample soil; and determining a heavy metal concentration spatial distribution diagram of the farmland in the investigation area according to the heavy metal predicted concentration spatial distribution diagram. According to the method, the accuracy of soil heavy metal concentration monitoring is remarkably improved, and the monitoring cost and time are effectively reduced.
Owner:INST OF AGRI RESOURCES & REGIONAL PLANNING CHINESE ACADEMY OF AGRI SCI

Network intrusion detection method and system, terminal and storage medium

The invention provides a network intrusion detection method and system, a terminal and a storage medium, and the method comprises the steps: determining a target sampling strategy according to the sample number of preprocessed data, and carrying out the sampling of the preprocessed data according to the target sampling strategy, and obtaining a sampling sample; inputting the sampled sample into a network intrusion detection model for feature extraction, deep information extraction and multi-head attention mechanism calculation to obtain a global feature vector; performing data type prediction according to the global feature vector, and calculating model loss based on a weight balance loss function dynamically adjusted based on the category number; training the network intrusion detection model according to the model loss; and inputting to-be-detected network data into the converged network intrusion detection model for intrusion detection to obtain a network intrusion detection result. According to the embodiment of the invention, the pre-processed data is sampled in a layered sampling mode, so that the phenomenon of unbalanced sample data category is effectively prevented.
Owner:JIANGXI TONGFU TECH

Lake and reservoir inflow and outflow scheduling scheme optimization method and system based on three-dimensional hydrodynamic force-water quality-water ecological model

The invention discloses a lake and reservoir inflow and outflow scheduling scheme optimization method and system based on a three-dimensional hydrodynamic force-water quality-water ecological model, and the method comprises the steps: firstly constructing a three-dimensional hydrodynamic force-water quality-water ecological model of a target lake and reservoir, setting a scheduling constraint, a scoring constraint, a scoring single item, and a comprehensive scoring weight by a decision maker, stratified sampling and one-dimensional uniform random sampling are carried out on each flow range of a water replenishing port, a flow outlet and a water suction pump station which can be regulated and controlled manually according to scheduling constraints set by a decision maker, and a plurality of inflow and outflow scheduling schemes are generated; and performing big data mining on the model calculation results of all the access flow scheduling schemes, performing comprehensive scoring according to the scoring single item and the comprehensive scoring weight, and selecting the access flow scheduling scheme with the highest comprehensive score as an optimal scheduling scheme for recommendation. The method has the advantages of high simulation precision, flexible decision support, multi-target comprehensive balance and the like, and a scientific and efficient technical solution can be provided for lake and reservoir inflow and outflow scheduling.
Owner:BEIJING YINGTELIWEI ENVIRONMENTAL TECH CO LTD

Universal crop genotype-to-phenotype prediction method based on multi-task learning

A crop genotype-to-phenotype general prediction method based on multi-task learning comprises the following steps: collecting genotype and phenotype data sets of five crops, performing quality control and coding on genotype data, and performing standardization processing on phenotype data; dividing the data set into a training set and a test set in proportion by adopting a stratified sampling method; by taking genotype data as input and phenotype data as output, a universal prediction model based on a multi-task learning framework is constructed, information sharing among different phenotype prediction tasks is realized, and potential relations between complex phenotypes and genotypes are accurately captured. According to the method, phenotypes can be accurately predicted according to genotypes in the early stage of breeding, so that materials with target characters are rapidly screened, the breeding period is effectively shortened, the breeding efficiency is remarkably improved, and an efficient and universal technical scheme is provided for crop breeding and related research.
Owner:NORTHWEST A & F UNIV

Water quality detection sampling device

The invention relates to the technical field of water quality detection, in particular to a water quality detection sampling device which comprises a shell of a cylindrical hollow structure. The floating body is installed in the middle of the outer side of the shell and used for driving the shell to float on the water surface; the filtering frame is installed at the lower end of the outer side of the shell, filtering holes are evenly formed in the filtering frame, and the filtering frame is used for filtering floating objects on the upper layer of the water surface; the sampling assembly is mounted in the shell and is used for sampling water at different depths in the multiple areas one by one; when the sampling barrel descends to the lowest position, the sampling barrel is just opened, the sampling barrel is vertically inserted into water, the collecting barrels at different heights on the sampling barrel can achieve the stratified sampling function, and the multiple sampling barrels can conduct stratified sampling on different sampled water quality one by one.
Owner:INNER MONGOLIA VOCATIONAL OF CHEM ENG

Flue-cured tobacco moisture and pigment content prediction method and system based on chain type integrated model

The invention belongs to the technical field of model prediction, and particularly relates to a flue-cured tobacco moisture and pigment content prediction method and system based on a chain type integrated model.The method comprises the steps that near infrared spectrum original data of tobacco leaves in different baking stages are collected to serve as spectrum data, and preprocessing is conducted through multiple preprocessing methods; dividing a training set and a test set by applying a stratified sampling mode; combining data obtained by different preprocessing methods with different regression methods, respectively constructing full-band prediction models, and screening the preprocessing methods; combining the screened preprocessing method with different characteristic wave band screening methods, respectively constructing a characteristic wave band prediction model, and determining an optimal sub-model for moisture and pigment content prediction; and combining a plurality of same optimal sub-models according to a cascading mode to construct a chain type integrated model, continuously carrying out iterative optimization, and outputting a final prediction result. According to the method, the moisture content and the pigment content in the tobacco leaf curing process can be subjected to multi-index prediction at the same time.
Owner:TOBACCO RESEARCH INSTITUTE OF CHINESE ACADEMY OF AGRICULTURAL SCIENCES (QINGZHOU TOBACCO RESEARCH INSTITUTE OF CHINA NATIONAL TOBACCO COMPANY)

High-precision water quality stratified sampling device

The invention relates to the technical field of water quality detection, in particular to a high-precision water quality stratified sampling device which comprises a main body, a plurality of fixing plates are arranged at the top of the inner side of the main body, a plurality of mounting cabins and a control mechanism are uniformly arranged on the outer wall of the circumference of the main body, and the control mechanism is arranged between the fixing plates. The control mechanism comprises a fixing shell fixedly connected with the inner wall of the fixing plate, a speed reducer is arranged at the bottom of the fixing shell, a rope disc is rotationally connected into the fixing shell, the input end of the top end of the speed reducer is fixedly connected with the bottom of the rope disc, and the output end of the bottom of the speed reducer is fixedly connected with a butt joint rod and a sampling mechanism. High-precision stratified sampling is realized through close cooperation of the control mechanism and the sampling mechanism. The rope disc is driven by a water body to rotate and drives the butt joint rod through the speed reducer, the pressing face of the butt joint rod acts on the butt joint face of the check block, opening and closing of a water inlet of the water taking box are accurately controlled, and accurate sampling is conducted on different water layers according to needs.
Owner:ZHEJIANG SHANQINGSHUILV CONSTR CO LTD

Stable adsorption configuration global search method based on similar distribution and active learning

The embodiment of the invention discloses a stable adsorption configuration global search method based on similarity distribution and active learning. The method comprises the following steps: S1, constructing a knowledge-embedded high-throughput adsorption structure model; s2, constructing an adsorption structure small sample data set based on clustering stratified sampling; s3, in combination with the adsorption structure small sample data set and the Gaussian process regression model, establishing an adsorption energy prediction-oriented proxy model; and S4, training the proxy model, and searching a stable adsorption configuration. The embodiment of the invention discloses a stable adsorption configuration global search method based on similar distribution and active learning. By means of catalytic configuration high-throughput modeling containing knowledge embedding, small sample data set construction based on clustering stratified sampling, a small sample recommendation model based on Gaussian process regression and hybrid convergence criteria considering data distribution and agent model prediction, efficient searching of a stable adsorption configuration in a large structure space is achieved.
Owner:ZHENGZHOU UNIV

Layered sampling equipment for land engineering investigation and sampling method thereof

The invention discloses stratified sampling equipment for land engineering investigation and a sampling method of the stratified sampling equipment, and belongs to the technical field of land engineering investigation. The device comprises a fixing plate, the lower side of the fixing plate is rotationally connected with a threaded rod, the outer side of the threaded rod is in threaded connection with two threaded cylinders, the outer sides of the threaded cylinders are fixedly connected with mounting plates, a rotating cylinder is jointly rotationally connected between the two mounting plates, and a plurality of stop rods slidably penetrate through the interior of the rotating cylinder; the lower side of the fixing plate is fixedly connected with a motor, and the output end of the motor is fixedly connected with a rotating shaft. According to the invention, the whole device is firstly placed in an excavated sampling point, the cutting plate is opposite to a soil layer to be sampled, and then the motor is started, so that the sampling operation can be automatically carried out, the workload of workers is reduced, the operation is simple, and the working efficiency is relatively high.
Owner:SHAANXI ESTATE DEV SERVICE CORP

Multifunctional random extraction method, system and device for NQI field expert database

The invention discloses an expert matching method based on structured data, and the method comprises the following steps: firstly, calculating the objective weight and subjective weight of each result index through information and a Delphi method based on the structured data in a database, and generating an expert comprehensive academic score; then, keyword information input by a front end is obtained, the matching degree of keywords and expert fields is calculated by adopting a word frequency-inverse document frequency algorithm, and experts are grouped according to the matching degree; and finally, according to the scale of the expert database and the number of persons to be extracted, adaptively selecting a random extraction or hierarchical distribution extraction mode to determine a target expert. According to the method, through a dynamic weight calculation model and an intelligent stratified sampling mechanism, the accuracy and flexibility of expert matching are realized, the score is continuously optimized in combination with the historical participation performance data of the experts, and the scientificity and practicability of expert extraction are improved.
Owner:NATIONAL INSTITUTE OF METROLOGY CHINA

Rapid Shapley value estimation method based on stratified sampling

The invention relates to a rapid Shapley value estimation method based on stratified sampling, and belongs to the technical field of fault diagnosis. According to the method, firstly, an improved evaluation index is defined, the contribution degree of features is measured through the distance between output of a target neuron and output of a reference neuron, and the nonlinear attribute of SoftMax is better met. And secondly, the potential relationship is measured by using the interaction relationship among the members, and the Shapley value of each feature is estimated through stratified sampling and a nonparametric test method, so that the calculation complexity is effectively reduced. Finally, the invention provides an explanation result accuracy quantitative evaluation method, the accuracy of the explanation result is evaluated through activation and inactivation characteristics, and a quantitative index is provided for evaluation of the explanation method. Experimental results show that compared with traditional SHAP, the method is more suitable for explanation of high-dimensional samples and has a better explanation result, and the blank of an explanation result accuracy quantitative evaluation method in the field of fault diagnosis is filled.
Owner:CHONGQING UNIV

Intelligent sampling and quality inspection system and method for grain depot

The invention discloses a grain depot intelligent sampling quality inspection system and method, and relates to the technical field of intelligent grain depot detection, and the method comprises the steps: collecting environment data, three-dimensional point cloud data and surface acoustic wave group velocity data, and carrying out the inversion calculation through a Love wave equation, and obtaining a three-dimensional density distribution diagram; calculating the density gradient in the vertical direction according to the three-dimensional density distribution diagram to obtain a stratified sampling parameter table; according to the stratified sampling parameter table, stratified puncturing is conducted on the grain pile, actual sampling depth data are generated, and a three-way sample valve is used for conducting sample distribution detection on the actual sampling depth data. Through layered detection of the three-way sample valve and near-infrared X-ray combined analysis, accurate detection of moisture, protein and heavy metal content is realized, and a detection result is matched with a density distribution diagram through a constrained Kriging interpolation method to generate a quality safety index.
Owner:THERMOWAY (HUBEI) INTELLIGENT TECH CO LTD

Soil detecting and sampling device for grain planting

The invention belongs to the technical field of soil sampling, and particularly relates to a grain planting soil detecting and sampling device which comprises a base, a supporting column is fixedly arranged on one side of the base, a top plate is fixedly arranged on the supporting column, a driving mechanism and a stratified sampling mechanism are arranged on the top plate, and a soil sample storage mechanism is arranged on the base. The stratified sampling mechanism comprises a sampling drill rod assembly and a pressure difference sampling assembly, the pressure difference sampling assembly is installed on the top plate, and the sampling drill rod assembly is located below the top plate and connected with the pressure difference sampling assembly; by combining the pressure difference principle with the longitudinally-arranged temporary storage cavities, soil of different depths can be collected in a layered mode in one-time sampling operation, after the soil is collected, the soil of different depths is sprayed into the corresponding sample receiving plates to be stored according to the pressure difference principle again, sample mixing is effectively avoided, accurate layered sampling is achieved, and the sampling efficiency is improved. And the collected soil sample is kept in an original state.
Owner:DEZHOU VOCATIONAL & TECHN COLLEGE

Urban space evolution detection method and system based on machine learning

InactiveCN120781021AFeature vectorEngineering
The invention relates to the technical field of data processing, and discloses an urban space evolution detection method and system based on machine learning. The method comprises the following steps: carrying out space-time stratified sampling on city multi-scale data to obtain block-level, district-level and city-level feature tensors; calculating a collaborative weight matrix through a multi-scale space attention mechanism; a collaborative feature vector is obtained through mapping of a collaborative order degree encoder; performing time sequence interpolation processing by using a space-time cooperation trajectory decoder to obtain a dynamic cooperation index; and performing multi-level fusion on the dynamic collaboration index to obtain a collaboration measurement result. The problem that urban spatial evolution detection is lack of multi-scale cooperative relationship quantitative analysis and dynamic evolution trajectory recognition capability is solved.
Owner:HARBIN INSTITUTE OF TECHNOLOGY (SHENZHEN) (INSTITUTE OF SCIENCE AND TECHNOLOGY INNOVATION HARBIN INSTITUTE OF TECHNOLOGY SHENZHEN)

Sewage detection stratified sampling device

The utility model relates to the field of sewage sampling, and discloses a sewage detection stratified sampling device, which comprises three sampling buckets, a support column and a sliding rod, the sampling buckets are internally provided with mounting cavities, the middle part of the support column is provided with a sliding chute, and the sliding rod slides in the sliding chute. The lower ends of the supporting columns penetrate through the sampling barrel and extend out of the sampling barrel, a plurality of sliding cavities are formed in the two sides, close to the sampling barrel, of the middle of the outer wall of a supporting column rod body, and connecting rods are fixedly connected to the two sides, close to the sampling barrel, of the middle of the outer wall of a sliding rod body. According to the stratified sampling device, when the stratified sampling device is used for sampling sewage, the stratified sampling device is in threaded connection with the filter screen on the outer wall of the threaded pipe, water plants, floating objects or other garbage in a sampling water area can be effectively prevented from blocking the water taking hole, the sampling process is ensured to be smoothly carried out, and in addition, the threaded connection mode is convenient to mount and dismount and simple and convenient to operate. And when the filter screen needs to be replaced or equipment needs to be maintained, a worker can quickly finish operation, so that the maintenance cost is reduced.
Owner:TIANJIN HONGCHENG ZHENGXIN ENVIRONMENTAL PROTECTION TECH CO LTD

Complex industrial contract information extraction system and method based on large language model

The invention provides a complex industrial contract information extraction method and system based on a large language model, and the method comprises the steps: S1, carrying out the clustering analysis of different obtained industrial contract texts, and determining a cluster where each contract text is located; s2, stratified sampling is carried out on each cluster, and representative samples are obtained; s3, carrying out key content extraction meeting a preset requirement on the representative sample corresponding to each cluster; obtaining a corresponding structured text based on the extracted key content meeting the preset requirement; s4, labeling the structured text to obtain labeled data; step S5, acquiring annotation data of all clusters to generate an annotation data set, and performing fine tuning on the large language model by using the annotation data set to obtain a fine-tuned large language model; and S6, deploying the fine-tuned large language model into a production environment, and extracting key information, meeting a preset requirement, of the unlabeled contract text.
Owner:欧冶工业品股份有限公司 +2

A method, device and computer readable storage medium for cell heterogeneity analysis of large-scale single-cell sequencing data

The application discloses a cell heterogeneity analysis method, device and computer readable storage medium for large-scale single cell sequencing data. The application combines kernel non-negative matrix factorization with deep neural network to construct a clustering method called KNMF-DNN. The method selects some representative subsets from the overall data, and uses the kernel non-negative matrix factorization method to cluster the subsets; then, the representative subsets are used as a training set, the remaining data is used as a test set, the labels obtained by clustering are used to classify the remaining data through the deep neural network, so that the clustering of the whole data is realized. In addition, the application also uses the idea of stratified sampling to select the most representative samples, and the best division of the clustering samples and the classification samples is determined by the KL divergence. Experimental results show that, compared with other latest methods, the KNMF-DNN has more excellent results on three real scRNA sequencing data sets.
Owner:RENMIN UNIVERSITY OF CHINA

Soil solidification effect detection and analysis system and method

The invention discloses a soil solidification effect detection and analysis system and method, and relates to the field of soil solidification effect detection.The soil solidification effect detection and analysis method comprises the steps that based on the sedimentation characteristics of a solidification material, soil is divided into a bottom sedimentation sudden change zone, a middle sedimentation gradual change zone and a top sedimentation stable zone in the depth direction, and the minimum operation thickness of a sampling tool serves as the reference layer thickness; dividing non-equidistant sampling layers according to different multiples in each settlement zone to intercept columnar soil samples; performing an initial dynamic load test on the original soil sample, then performing circulating saturated water drying treatment on the same soil sample, and applying the completely same dynamic load mode again; a water circulation effect in a natural environment is simulated by performing vacuum water saturation and constant-temperature drying circulation treatment on a soil sample for multiple times and combining a dynamic load test after water damage. Through non-equal-interval stratified sampling and multi-round dry-wet cycle testing, stratified dynamic and high-precision quantitative evaluation of the water damage resistance of soil solidification is achieved.
Owner:中交综合规划设计院有限公司

Merchant automatic classification method and system

The invention provides a method and system for automatically classifying merchants, and the method comprises the steps: obtaining multi-source data, and generating a structured merchant data set; performing feature engineering on the merchant data set, constructing a merchant multi-dimensional feature vector, and performing multi-dimensional marking on the multi-dimensional feature vector to obtain marked merchant feature data; performing data set division and preprocessing on the commercial tenant feature data by using random sampling, stratified sampling and cross validation technologies to obtain a model training data set, training a commercial tenant classification model, and generating a trained commercial tenant classification model; classifying the commercial tenants by using the trained commercial tenant classification model, and calculating scores of the commercial tenants based on a set grade division standard and a scoring rule system to obtain a commercial tenant grade scoring result; and comparing the commercial tenant grade scoring result with a preset risk threshold value, and generating a commercial tenant classification and risk assessment result. According to the invention, the accuracy of commercial tenant classification can be effectively improved.
Owner:BEIJING YULORE INNOVATION TECH