58results about How to "Avoid Data Bias" patented technology

The invention discloses an experiment device used for researching effects on fish behaviors imposed by light and the application thereof. The experiment device comprises a water tank, a double-layered sleeve, a transparent waterproof glass reinforced plastic lampshade , an LED lamp pipe set, a motor, a circuit control system and a camera. According to the invention, by arranging a PVC cylinder of the double-layered sleeve, the experiment is carried out when fish freely swims in an experiment water tank, so via light transmitting holes in the PVC cylinder, experimental fish can freely swim in and out of the cylinder, and an ideal habitat is provided for the experimental fish, thereby simulating the free choices to light wave length and illuminance of the fish in natural sea and industrial aquaculture conditions in a more real manner. Thus, data deviation caused by harmful effect to the fish behaviors by improper structure and operation is avoided. Via the experiment device, many kinds of experiments on effect on fish behaviors imposed by light can be well finished, and the experiment device can satisfy many experimental objectives and requirements for researching effect on fish behaviors imposed by light.

The embodiment of the invention provides a method and a system for acquiring road traffic condition information based on mobile phone signaling data. The method comprises steps: in a preset time period, GPS data of all vehicles in a target road section and / or mobile phone signaling data of mobile phone users in all vehicles are acquired; according to all the GPS data and / or all the mobile phone signaling data, the vehicle speed in the target road section is determined; and according to the vehicle speed, the road traffic condition information in the target road section is acquired. According to the method and the system provided in the embodiment of the invention, in order to acquire the road traffic condition information in the target road section, the GPS data are collected, the mobile phone signaling data are also collected, the vehicle speed in the target road section is determined according to the data in two different sources, and the problems of data deviation and poor risk resistance caused by using data in a single source can be avoided.

The invention relates to a data cleaning method for improving the accuracy of target data, which comprises the following steps: (1) finding out a plurality of original field items of A1, A2, ..., An,which are related to the target data from a data source; (2) establishing a logic processing model: the quantity of the target data is equal to f(M1(q1, G1(A1)), M2(q2, G2(A2)),...Mn(qn, Gn(An))), wherein, A1, A2...An are the original field items respectively, G1(A1), G2(A2)...Gn(An) are attribute functions which respectively reflect the data attribute of each original field item, and q1, q2...qnare weight value of each original field item respectively; M1(q1, G1(A1), M2(q2, G2(A2))...Mn(qn, Gn(An)) are influence functions which influence each original field component of the target data value, f() is the determination function for determining the target data value according to each influence function; and (3) finding out all the original field items during each time of data cleaning, anddetermining the value of the target data according to the logic processing model. The data cleaning method can improve the accuracy of the target data which is cleaned out from the data source.

The invention discloses a progress monitoring method used for laboratory engineering construction. The method comprises the steps of: creating a building information model of a laboratory, wherein the building information model comprises a 3D building model, and the 3D building model comprises a component model; creating information of a laboratory engineering construction progress plan, and correlating the engineering construction progress plan and the component model of the 3D model; arranging an Internet-of-Things tracking device in the laboratory engineering construction, and acquiring locations of construction devices and construction materials which are tracked by the Internet-of-Things tracking device in real time and in laboratory engineering ; updating engineering construction progress according to location information of the construction devices and the construction materials; and comparing the engineering construction progress and the engineering construction progress plan, and issuing early warning information to a client when the difference between the engineering construction progress and the engineering construction progress plan is greater than a preset threshold value. The method improves the efficiency of monitoring and the level of automation.

The invention discloses a probe assembly for monitoring seawater quality, which belongs to the technical field of environment monitoring. The probe assembly comprises an outer shell, a protection shell and a protection sleeve, wherein the bottom of the outer protection shell is provided with a storage box, the bottom of the storage box is provided with an openable and closable bottom cover, poresare uniformly formed in the bottom cover, the protection sleeve is arranged outside the protection shell in a sleeving manner, and the protection sleeve consists of a low-density layer and a high-density layer sleeving the low-density layer. By adopting the probe assembly for monitoring the seawater quality, the material scattering the smell disliked by a breeding object is placed in the storage box on the bottom of the protection shell, after the material is independently dissolved, the material permeates into the protection sleeve, and the protection sleeve sleeves the outer side of the protection shell, so that in a monitoring process, the breeding object does not approach the probe assembly, the data deviation caused by the swinging of the breeding object around the probe assembly in the monitoring process can be avoided, and the damage problem of the probe caused by the collision of the breeding object can be avoided.

The invention discloses an information data model analysis method, which belongs to the technical field of data processing. The information data model analysis method comprises the following steps: generating a corresponding analysis scene and an analysis result according to case-related basic information, selecting a target analysis result, and generating a target analysis task list; acquiring to-be-analyzed data by using a database querier, preprocessing the acquired data to generate intermediate data, establishing a query model, and storing the intermediate data and the query model in a memory; and reading the corresponding analysis models from the algorithm library one by one according to the analysis task list, calling an analysis execution module to analyze the data in the referencedmemory according to strategy trees of the analysis models to generate data features, and storing and displaying the data features on a graphical interface. According to the invention, the problems oflarge time consumption and large workload of case handling personnel due to multiple rounds of analysis, data deviation and repeated work of information analysis of information data such as bank bills, telephone bills, tracks, tax receipts and the like in the existing case handling process are solved.

The invention belongs to the field of biomedicine and relates to a miRNA molecular marker for diagnosing type 2 diabetes mellitus and application thereof. The steps are as follows: extraction of total RNA in blood and preparation of cDNA; fluorescent quantitative PCR detection of hsa-miR-149-3p level; evaluation of the miRNA molecular marker. According to the invention, large sample tests indicate that novel miRNA marker hsa-miR-149-3p in peripheral blood is significantly increased in the type 2 diabetes mellitus, and has a molecular level positively correlated with the blood glucose level of a subject. A detection method is simple and easy to operate, has little trauma to the subject, is high in sensitivity and specificity, and can be used for the diagnosis of type 2 diabetes mellitus at early stage and different stages. On this basis, the invention develops a primer, a kit and the detection method for detecting hsa-miR-149-3p in the blood, and has obvious clinical application value.

The invention relates to the technical field of risk area evaluation of bursting liability in underground mining of coal mines, and discloses a pre-evaluation device and method for a coal seam rock burst risk advanced drilling test. The device comprises a drilling mechanism comprising a drill bit, a directing piece and a stabilizer connected in sequence, a measurement mechanism while drilling usedfor acquiring one or more of ground stress, surrounding rock strength, an electromagnetic radiation signal and a natural gamma signal, an in-situ measurement mechanism used for acquiring one or moreof a temperature curve, hole wall deformation degree and position, coal seam thickness, structure occurrence and a natural resistivity curve, and a directional coring mechanism used for acquiring a rock core; the stabilizer is detachably connected with a pipe column; the measurement mechanism while drilling is detachably connected with each of the pipe column and the stabilizer; the in-situ measurement mechanism is detachably connected with the pipe column; and the directional coring mechanism is detachably connected with the pipe column. According to the pre-evaluation device, accurate evaluation and determination of a bursting risk area provide basic measurement data for accurate pressure relief in advance before coal mining.

The invention relates to a multi-step communication distributed optimization method based on a BB step length, and belongs to the technical field of large-scale machine learning and information processing. The method comprises the following steps: 1), determining a problem objective function and a strong convex coefficient and a smooth coefficient thereof; 2), constructing a directed strongly-connected unbalanced communication network, and generating a row-column random weight matrix by adopting a uniform weight strategy according to the adjacency matrix of network topology; 3), initializing all variables of the system; 4), calculating a BB step length by utilizing variable information currently stored by the node; 5), enabling the system to calculate the number P of internal circulation times according to the related parameters; and 6), enabling the nodes to receive information of the neighbor nodes and updating the variables in combination with the information stored in the nodes until the variables converge, and obtaining the optimal value of the problem objective function. The method can solve the problem of unconstrained optimization under the directed strongly-connected unbalanced communication network, and the convergence rate and gradient estimation accuracy of the algorithm are improved.

The invention provides an agricultural whole industry chain traceability system based on a block chain. The system comprises a technical support layer, a digital rural layer, an agricultural source supply chain layer, a supply and marketing layer and a nutrition and health layer, wherein the cooperation agencies, farmers and large enterprises of the digital rural layer and the agricultural sourcesupply chain layer have an association relationship, and the agricultural source supply chain platform is associated with an agricultural big data platform and can provide a corresponding interface tobe associated with other e-commerce platforms, i.e., the traceability end information is associated with the commodities of the other e-commerce platforms; the method has the advantages that the trust problem is solved, human errors are avoided, a traditional information island is broken through, namely decentralization and tamper resistance are achieved, an agricultural product health system iscreated, and income and efficiency are increased for the government industry.

The invention discloses a score card fusion method and device, computer equipment and a storage medium, and belongs to the technical field of risk control. The method comprises the steps that an expert score card and a machine learning score card are acquired, and feature variables used by the expert score card are different from feature variables used by the machine learning score card; the expert score cards and the machine learning score cards are respectively fused on the plurality of weight combinations to obtain a plurality of fused score cards; and the plurality of fusion score cards are verified by using the verification sample, and the optimal fusion score card is screened out. Compared with the prior art that variables selected by experts are directly added into the machine learning model, the process is repeated until the variables meeting the model conditions are selected, the problem that the repeated workload is large can be avoided, the problem of sample data deviation can be effectively avoided, and the model prediction capacity is improved.

The invention discloses a method and a system for realizing the generation and data updating of an underground pipe network GIS system, comprising the steps of selecting and running a corresponding GIS application unit according to actual operation requirements; scanning a traceability code on an underground pipe or pipe to be operated to obtain original information, wherein the underground pipe or pipe is embedded with an RFID chip and a GPS chip; receiving inputted new operational information on the basis of the original information; the original information and the input new operation information being sent to the database of GIS system by wireless transmission; running the GIS generation program on the server side, all the information in the database being gathered to form a complete underground pipe network GIS system, so as to achieve the generation of underground pipe network GIS system and data updating. The method and the system of the invention can automatically generate a new GIS system when installing or repairing pipelines, which ensures that the GIS system is always up to date, has strong practicability and is suitable for wide-range popularization and application.

The invention relates to a portable corrosion-proof micro-organism growth-proof tide gauge which comprises a bracket and an underwater probe; the probe is fixedly arranged on the bracket; the bracket is a structure consisting of three conical bar frames; the underwater probe comprises an underwater probe shell, a water inlet cover and a cable joint; a sensor and a circuit board are arranged in the underwater probe shell; the water inlet cover is coaxially arranged at the lower end of the underwater probe shell; a plurality of water inlet grooves are axially arranged on the outer cylindrical surface of the water inlet cover; the underwater probe shell is coaxially provided with a red copper uniform flow cover through an insulating sleeve on a water inlet nozzle arranged in the water inlet cover; and water inlets are uniformly distributed on the outer cylindrical surface and the lower bottom surface of the uniform flow cover. In the invention, the tide gauge is simple in integral structure, scientific and reasonable in bracket design, quick and firm to mount and convenient to carry; sacrificial anodes are respectively arranged on the underwater probe uniform flow cover and the water inlet cover, and the corrosion-proof effect is excellent; the uniform flow cover arranged at the water inlet nozzle is capable of guaranteeing the data acquisition precision of the sensor, simultaneously the growth of the marine micro-organism on the surface of the underwater probe is effectively avoided, and the precision of the data acquisition of the tide gauge is effectively guaranteed; and the tide gauge is simple in integral structure and scientific and reasonable in design.

The invention relates to the technical field of sterilization monitoring and particularly relates to a disinfectant sterilization effect monitoring device convenient to sample. The device comprises a sampling mechanism, wherein a monitoring mechanism is mounted in the sampling mechanism, a supporting mechanism is connected to the sampling mechanism, and a protection mechanism is mounted in the supporting mechanism; a placing mechanism is mounted on the supporting mechanism, a taking mechanism is mounted in the supporting mechanism, a cleaning mechanism is mounted in the sampling mechanism, and a control mechanism is connected to the interior of the sampling mechanism; the prepared disinfectant liquid is conveniently extracted and detected under the action of the sampling mechanism and the monitoring mechanism, data are obtained, the sterilization effect is checked, the sampling mechanism is conveniently and stably placed under the action of the supporting mechanism, and samples are conveniently and rapidly switched and need to be extracted under the action of the placing mechanism; under the action of the control mechanism and the cleaning mechanism, interiors of the sampling mechanism and the monitoring mechanism are conveniently washed, and data errors caused during subsequent detection are prevented.

The invention discloses an information processing method and application of project management. The method comprises the steps that under the condition that a to-be-processed engineering project is determined, basic information of the engineering project is acquired; a closed-loop management flow is generated according to the basic information of the engineering project; based on the closed-loop management flow, flow nodes are sequentially sent to an appointed account; and when an interaction instruction sent by the specified account based on the process node is received, the basic informationof the engineering project is updated according to the interaction instruction. Through the method and the application, the problems that a universal project management system in the related art cannot meet the requirements of a user account and is low in processing efficiency are solved.

The invention relates to a method for positioning a bottom section and a broadside section on a heavy lift vessel, which comprises the following steps of: detecting data of the bottom section and the broadside section of the heavy lift vessel by erecting a total station, comparing the detected data with a theoretical value to obtain a three-dimensional change reference value, and obtaining field positioning data of the bottom section and the broadside section; measuring equipment is fully utilized to be combined with three-dimensional analysis software to achieve detection of positioning data of the bottom and the broadside of the heavy lift vessel, the total station is easy and convenient to operate and high in measuring precision, field point position collection can be conducted on bottom data and broadside segments through the total station, and after the collected data are matched with theoretical data of the bottom and the broadside respectively, the positioning data of the bottom and the broadside can be obtained. Therefore, the change conditions of the bottom data and the broadside data can be directly judged on site. According to the method, the positioning precision is high, the change condition of bottom and broadside data can be directly judged on site, the labor intensity of constructors is reduced, and the safety coefficient of the constructors is increased.

The invention discloses a processing method for fitting and sampling telemetry data fitting and sampling of a data acquisition and monitoring control system, and relates to the technical field of data processing. The invention aims to solve the technical problem to ensure that the data sampling accuracy is improved in the telemetry data sampling in a power system. The method comprises the following steps: 1) taking at least 12 sample data; 2) processing extreme values of the data: removing the maximum and the minimum; 3) solving a weighted value of fitting data; fitting the sample data E(Xi) according to a weighing formula; 4) computing a sampling error, wherein the sampling error e is equal to Xio-E(Xi); 5) setting a warning sign for acquired abnormal data; and for the given error limit Ee, if e is greater than Ee, transmitting the data exception warning, and using the fitting data to replace the sample data; and 6) saving the fitting data, repeating the steps 1) to 5), and performing fitting the sampled telemetry data. The method has the characteristics of avoiding data deviation and data exception, and improving the data sampling accuracy.

The invention provides an immune thrombocytopenia lncRNA marker, a kit and an application thereof. The lncRNA marker is LOC100506314, and its base sequence is shown in SEQ ID NO: 1. The base sequenceof the primer of the lncRNA marker is shown in SEQ ID NO: 2 and SEQ ID NO: 3. LncRNA markers and primers thereof are used for preparing drugs for inhibiting immune thrombocytopenia or for preparing immune thrombocytopenia auxiliary diagnostic kits. The lncRNA marker can be used for judging the progress and severity of immune thrombocytopenia. The LOC100506314 primer and diagnostic reagent kit anddetection method developed on the basis of LOC100506314 are easy to operate, less invasive, high sensitivity and specificity, and can be used to monitor the disease progression clinically. The LOC100506314 primer and diagnostic reagent kit has an auxiliary diagnostic effect on the disease onset, drug efficacy and prognosis, and has obvious clinical application value.

The invention relates to a lightweight SAR image ship detection method based on stripe pruning, and belongs to the technical field of image processing. Extracting feature maps of different depths by using a strip convolution-based lightweight residual convolutional network; fusing different depth features by using a jump connection feature pyramid network to obtain fused depth feature maps of different scales; and predicting the probability of the corresponding position of each feature point and the vertical distance from the position of the feature point to the edge of the target by using a regression model based on the key points, and finally obtaining the position and probability of the target. According to the method, the lightweight residual convolutional network is obtained through one-time end-to-end training by using the lightweight residual convolutional network based on strip convolution, so that the training efficiency and the model robustness of the model are improved; the dimensionality of pruning is increased through strip pruning, and the parameter quantity and the calculated quantity of the model are further compressed.

The invention discloses an inspection report visual configuration method in a medical inspection LIMS system, and relates to the technical field of information management. The examination report visual configuration method in the medical examination LIMS system comprises the following steps of template editing and data binding, data uploading and data previewing, report batch generation and report comprehensive management. According to the visual configuration method for the inspection report in the medical inspection LIMS system, visual configuration of the inspection report is achieved through template editing and data binding, data uploading and data previewing, batch generation of reports and comprehensive management of the reports, data deviation of a computer in the process of generating the inspection report can be effectively avoided, the report template does not need to be developed by professional personnel, high matching of different report data and the template can be achieved through binding of the template and the data, a user can effectively obtain the needed report data through the configuration method, and the configuration method is simpler and more efficient.

The invention discloses a compression resistance detection device for production and processing of wooden furniture. The compression resistance detection device comprises a bottom plate, grooves are machined in left and right sides of the inner wall of the bottom plate, lifting mechanisms are mounted on inner walls of the two grooves, outer walls of the two lifting mechanisms are fixedly connectedwith supports, and tops of the two supports are fixedly connected with left and right sides of the bottom of the bottom plate respectively. By cooperation of long grooves and a press mechanism, the compression resistance detection device for production and processing of wooden furniture solves the problem that the existing compression resistance detection device can only detect one piece of wooden furniture during use, work is simple, the working time is shortened, two sets of wooden furniture can be compared conveniently, data deviation is avoided, and later work is guaranteed; besides, by cooperation of a sliding mechanism and a clamping mechanism, the problem about change according to the size of the wooden furniture is solved, the requirement of different specifications of wooden furniture is met, the usability is improved, and the device is convenient to promote.

The invention discloses a progress monitoring method used for laboratory engineering construction. The method comprises the steps of: creating a building information model of a laboratory, wherein the building information model comprises a 3D building model, and the 3D building model comprises a component model; creating information of a laboratory engineering construction progress plan, and correlating the engineering construction progress plan and the component model of the 3D model; arranging an Internet-of-Things tracking device in the laboratory engineering construction, and acquiring locations of construction devices and construction materials which are tracked by the Internet-of-Things tracking device in real time and in laboratory engineering ; updating engineering construction progress according to location information of the construction devices and the construction materials; and comparing the engineering construction progress and the engineering construction progress plan, and issuing early warning information to a client when the difference between the engineering construction progress and the engineering construction progress plan is greater than a preset threshold value. The method improves the efficiency of monitoring and the level of automation.

The invention belongs to the technical field of biology, and particularly relates to equipment capable of providing self-protection for a cell test tube for a biological cell test. A contraction plateis movably connected to the interior of a workbench, a clamping plate is movably connected to the outer surface of the contraction plate, a telescopic spring is movably connected to the outer surfaceof the clamping plate, a limiting column is fixedly connected to one side of the telescopic spring, a limiting base is fixedly connected to the outer surface of the contraction plate, a limiting groove is formed in the outer surface of the limiting base, a conductor base is arranged on the position, located in the workbench, of the outer surface of the contraction plate, an electromagnetic coil is fixedly connected to the interior of the workbench, a protective base is fixedly connected to the outer surface of the workbench, and an electromagnetic base is fixedly connected to the interior ofthe protective base. The problem of overlarge centrifugal force possibly caused by a stall phenomenon can be prevented in use, so that the defect of cell tissue damage caused by the centrifugal forcecan be avoided, and the correctness of cell tissue sample data is ensured.

The invention relates to a direct current resistance testing device and method of an intelligent transformer. The device is composed of a machine box carrier and an inner core which is arranged in themachine box carrier, and the inner core comprises a central processing unit and a constant current source circuit, a signal collecting module, a measuring protecting module, a relay switching unit and a human-computer interface module which are connected with the central processing unit; the constant current source circuit comprises a three-phase constant current source generating controller, thesignal collecting module comprises a current signal collecting circuit and a voltage signal collecting circuit, and conventional measuring and one-button measuring are switched by the relay switchingunit according to a control demand. When testing is conducted, a testing method is selected on a device side according to the linking manner of a tested winding group, the measuring of all winding groups of primary and secondary winding groups is completed in one attempt, the working efficiency is improved, and the potential safety hazards of the falling of wires caused by climbing height are greatly reduced.

The invention relates to the technical field of data processing and intelligent building design, in particular to a fabricated building design method and system and a cloud platform. By determining the component information set, whether the construction safety information of the to-be-analyzed building component changes or not can be judged, and if not, the fabricated modeling model is created through the building design terminal, so that the problem of multiple data modifications during creation of the fabricated modeling model can be avoided, and the efficiency of creating the fabricated modeling model is improved; if so, the structural component data is updated and designed, so that the fabricated modeling model can be more accurately created; the target design parameters are acquired according to the preset data synchronization acquisition protocol, the problem of data deviation can be avoided, and the target design parameters are input into the fabricated modeling model, so that the building components and accessories can be safely assembled according to the obtained fabricated building design drawing, and the construction safety of assembly is further ensured.

The invention belongs to the technical field of the electronic circuit, and provides a signal processing circuit and device for a non-contact position sensor. An oscillation signal is provided for anexciting coil through an oscillation module, and a plurality of induction coils output corresponding induction signals after the mutual inductance with the exciting coils occurs when a location of therotor is changed, a digital signal is output to a digital signal processing module so as to compute and reduce the location information of the rotor after a synchronization rectification module performs the rectification, a switch module performs the gating, a filter module performs the filtering, an amplification module performs the signal amplification and an analog-digital conversion module performs the analog-to-digital conversion, so that an electronic control module performs analysis processing and makes the corresponding execution control; and meanwhile, a multi-phase clock productionmodule is used for adjusting phases of multiple collection ports in the synchronization rectification module in real time so as to correct the location information; the effect of precisely acquiring and correcting the rotor location information is realized, thereby precisely detecting a mechanical device connected with the rotor.

The invention discloses a power grid model classification identification method and system, electronic equipment and a storage medium, and the method comprises the following steps: SS1, maintaining stock model classification tags, generating classification tags and wiring diagrams, and storing the classification tags and the wiring diagrams into a training sample model library; step SS2, performing feature extraction on the training sample model library to form a model category feature library; and step SS3, performing model classification identification on the incremental primary equipment wiring diagram to generate a type label of the primary equipment model. According to the invention, training and feature extraction are carried out on a stock power system primary equipment wiring diagram with classification labels, classification features of various power system primary equipment models are identified from the diagram, a power grid model classification feature library is formed, and an effective power grid model classification and identification mechanism is established. Cost waste caused by manual data maintenance can be effectively avoided, data maintenance efficiency and accuracy are improved, and power grid operation safety is improved.

The invention provides a campus health data management method and system, and the method comprises the steps: obtaining the attribute data of a target object through a portable terminal, collecting the health data of the corresponding target object, and outputting the health data to a server side; the server side constructs multiple pieces of hierarchical data according to the health data, and obtains a health data distribution statistical result of the target object in each piece of hierarchical data based on the attribute data of the target object; setting one or more display terminals to be connected with the server side, and displaying the distribution statistical result in real time; creating a health guidance information base, and acquiring corresponding health guidance information from the health guidance information base according to different levels of data and output; the health conditions of students at different stages can be known in real time, and reasonable improvement measures can be given in time.

The invention discloses a biological flora culture monitoring system which structurally comprises an operation block, an operation box, a closing door, an oxygen generation block and a cover body, the operation block is installed on the operation box, the operation box is movably connected with the operation block, the closing door is installed on the operation box, the operation box is in clearance fit with the closing door, and the oxygen generation block is installed on the operation box. The operation box is connected with the oxygen generation block, the cover body is installed on the operation box, and the operation box is embedded with the cover body. After the pH of a culture sample liquid is monitored once, an adjusting and monitoring structure after monitoring is completed is replaced by an unmonitored adjusting and monitoring structure through an assembling and connecting structure of an installer; the adjusting and monitoring structure slides along the sliding rod on the rotating block, the adjusting and monitoring structure is pushed into an incubator for monitoring, culture liquid is not adhered to the surface of the adjusting structure replaced by the system, and therefore the situation that monitoring data deviates due to the fact that the culture liquid is adhered to the adjusting and monitoring structure in the monitoring process is avoided.

The invention provides a miRNA marker of immune thrombocytopenia, a kit and application. The miRNA marker is hsa-miR-302e, the base sequence of the marker is shown as SEQ ID NO:1. The base sequences of the miRNA marker's primers are shown as SEQ ID NO:2 and SEQ ID NO:3. The miRNA marker and the primers thereof can be used for preparation of drugs inhibiting immune thrombocytopenia or preparation of immune thrombocytopenia auxiliary diagnosis kit. The miRNA marker can be used for judging the progress and severity of immune thrombocytopenia. The hsa-miR-302e primers, diagnosis kit and detectionmethod developed based on the miRNA marker have the characteristics of simple operation, little trauma to examinees, high sensitivity and strong specificity, and can be used for diagnosis of immune thrombocytopenia at the early stage and different stages, and have obvious clinical application value.