Patents
Literature
346 results about "Numeric Value" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Information in the form of a numeral contained in a data field.
Numerical data variable-length coding storage method and device, equipment and medium
The invention relates to the technical field of data storage, and discloses a numerical data variable-length coding storage method, device, equipment and medium, which comprises the following steps: converting target numerical data into fixed-length binary representation according to definition precision, executing filling removal interception processing on an integer part, executing digital-level packet compression coding on a decimal part, and performing variable-length coding on the target numerical data; and generating a variable-length binary sequence containing length information and symbol information, and writing the sequence into a storage medium to realize compact storage of numerical data. According to the method, redundant filling bytes are eliminated from an integer part, compressed encoding is performed on a decimal part according to effective digits, and unified processing of sign bits is combined, so that a storage space dynamically changes along with actual numerical digits on the premise of keeping numerical precision and analyzability, space waste of fixed-length numerical types is reduced, and the method is suitable for large-scale popularization and application. Meanwhile, the precision error of the floating-point number is avoided, and the storage efficiency and the data reliability are improved.
Owner:JINZHUAN INFORMATION TECHNOLOGY CO LTD
Floating-point number format conversion device and method, storage medium and program product
ActiveCN121657960ADigital data processing detailsFile system administrationCircuit complexityFloating point
The invention discloses a floating-point number format conversion device and method, a storage medium and a program product, and relates to the technical field of floating-point number format conversion. The device comprises a preprocessing unit and a conversion mapping unit; the preprocessing unit is used for receiving input data in a first floating point format and performing bit width compression processing on at least part of digits influencing rounding operation in the input data to generate a corresponding index value, and the bit width of the index value is smaller than the mantissa bit width of the input data; the conversion mapping unit is coupled to the preprocessing unit, a plurality of mapping relations between index value ranges and numerical value representations in a second floating point format are stored in the conversion mapping unit, and the conversion mapping unit is used for searching the corresponding numerical value representations in the second floating point format according to the index values to serve as output data. The invention aims to reduce the number of logic gates and the circuit complexity required by floating-point number format conversion, so that the area of a related chip is reduced, the power consumption is reduced, and the operation delay is shortened.
Owner:SUZHOU YIZHU INTELLIGENT TECH CO LTD
Data processing method and device
The invention discloses a data processing method and device, which can reduce service time delay and equipment cost. In the method, a data receiving and transmitting end determines N CB groups corresponding to a first TB according to a first numerical value. The sending end executes LCP with the CB group as the granularity, and the receiving end decodes the CB group with the CB group as the granularity. Wherein the first numerical value is the number X of the CBs contained in one CB group, or the number Y of the CB groups contained in one TB. The N CB groups meet one of the following items: the number of CBs included in each CB group in the N CB groups is a second numerical value; or, the number of the CBs contained in the first M CB groups in the N CB groups is a second numerical value, the number of the CBs contained in the last N-M CB groups is a third numerical value, and M is a positive integer smaller than N; or, the number of the CBs contained in the first N-M CB groups in the N CB groups is the third numerical value, and the number of the CBs contained in the last M CB groups is the second numerical value; or, the number of the CBs contained in each CB group in the N CB groups is a fourth numerical value.
Owner:HUAWEI TECH CO LTD
Machine Learning Using Four-Bit Binary Data Formats
PendingUS20260037593A1Digital data processing detailsComplex mathematical operationsAlgorithmTheoretical computer science
A computing system can obtain a machine-learned model comprising one or more parameters having a four-bit binary format. The four-bit binary format can correlate a plurality of sixteen respective binary values to a plurality of sixteen corresponding numerical values represented by the binary values. The sixteen numerical values can be symmetric about a median. A plurality of step sizes between the sixteen numerical values can be non-uniform. The computing system can obtain one or more input values for one or more layers of the machine-learned model. The computing system can process, based at least in part on the one or more parameters having the four-bit binary format, the one or more input values to generate one or more output values.
Owner:GDM HOLDING LLC
Method for changing data frame length, electronic device, and storage medium
ActiveCN118677567Bguaranteed costGuaranteed payloadComputer hardwareElectrical and Electronics engineering
Owner:ZTE CORP
A method, system, and medium for storing the degree-of-freedom numbers of a numerical computation grid.
PendingCN122088190AMemory adressing/allocation/relocationDesign optimisation/simulationData compressionData set
This invention discloses a method, system, and medium for storing degree-of-freedom numbers in a numerical computation grid, relating to the field of numerical computation technology. By improving the storage structure of degree-of-freedom numbers, the storage and access efficiency of degree-of-freedom numbers is optimized, and different numbering methods can be selected according to actual needs, thereby improving the flexibility and overall performance of numerical computation. Two data sets are designed for each variable group to uniformly store the degree-of-freedom numbers of variables within the group, and the storage space occupied is effectively reduced through data compression.
Owner:NUCLEAR POWER INSTITUTE OF CHINA
Fixed-point division calculation method and device, control chip, air conditioner and storage medium
This application relates to a method, apparatus, control chip, air conditioner, and storage medium for calculating fixed-point number division. The method includes: obtaining a fixed-point number to be calculated, wherein the fixed-point number to be calculated includes a first value as the dividend and a second value as the divisor; mapping the first value and the second value to a preset fixed-point format respectively to obtain a third value and a fourth value represented by the preset fixed-point format, wherein the data convergence efficiency of the preset fixed-point format is higher than the data convergence efficiency of the original format of the fixed-point number to be calculated; extracting a value of a preset bit width starting from the least significant bit of the fourth value, and determining the extracted value as a fifth value; performing iterative calculation on the third value and the fifth value until the current value meets a first preset condition, and stopping the iterative calculation; and determining the division result corresponding to the fixed-point number to be calculated based on the third value obtained from the final iterative calculation. This can improve the efficiency of fixed-point number division calculation.
Owner:GREE ELECTRIC APPLIANCE INC OF ZHUHAI +1
Method and apparatus for recording retry in service processing, electronic device and storage medium
ActiveCN115952167BEnergy efficient computingSpecial data processing applicationsComputer hardwareEngineering
The application provides a method and device for recording retries in business processing, an electronic device and a storage medium, wherein the method comprises: obtaining an associated record related to a target task to be executed, wherein the associated record contains a current retry number of a method function corresponding to the target task, a current time, and an interval duration between the current time and a next execution time; storing the associated record to a first preset table and running the target task; and moving the associated record to a second preset table or a third preset table according to a target number obtained after scanning the associated record according to a running result and a target thread, wherein the target number is a value obtained by increasing the preset number to the current retry number after the method function is called the preset number of times based on the interval duration. Through the application, the problem of leaving no trace of retries and being unable to query the retry record in time in the subsequent process in the related art is solved.
Owner:北京自如信息科技有限公司
Floating-point number format conversion apparatus and method, storage medium and program products
ActiveCN121657960Breduce areareduce power consumptionDigital data processing detailsFile system administrationCircuit complexityFloating point
This application discloses a floating-point number format conversion apparatus and method, storage medium, and program product, relating to the field of floating-point number format conversion technology. The apparatus includes a preprocessing unit and a conversion mapping unit. The preprocessing unit receives input data in a first floating-point format and performs bit-width compression processing on at least a portion of the bits in the input data that affect rounding operations to generate corresponding index values, wherein the bit width of the index values is smaller than the mantissa bit width of the input data. The conversion mapping unit is coupled to the preprocessing unit and stores multiple mapping relationships between index value ranges and numerical representations in a second floating-point format, used to look up the corresponding numerical representation in the second floating-point format as output data based on the index values. This application aims to reduce the number of logic gates and circuit complexity required for floating-point number format conversion, thereby reducing the chip area, power consumption, and operation latency.
Owner:SUZHOU YIZHU INTELLIGENT TECH CO LTD
A database replication method and system
ActiveCN121979956BDatabase updatingDatabase distribution/replicationTheoretical computer scienceEngineering
The application provides a database replication method and system, the method comprises the following steps: responding to a binary change log and performing analysis, extracting an initial log source sequence number, a timestamp and an associated partition identifier, and then obtaining a database partition preset sequence number; if the two sequence numbers are not continuous, the timestamp is valid, and the initial log source sequence number satisfies a preset first condition, a new independent numerical range is created and a target missing sequence number interval is identified; a data generation log supplement signal is generated based on the above data and sent to a producer cluster, so that the producer cluster feeds back a missing sequence log; based on the independent numerical range and the database partition preset sequence number, a database partition first sequence number is obtained, and the replication of the database is completed. The database replication method provided by the application avoids the problems of repeated data insertion and missing log omission supplement, realizes efficient consistent replication of the database, and guarantees the stability and reliability of the database replication in a multi-cluster scenario.
Owner:GUANGZHOU SHUANGZHAO ELECTRONIC TECH CO LTD
Coding method and related equipment thereof
Owner:HUAWEI TECH CO LTD
Packet assembling control method of data packet, storage medium and electronic equipment
The embodiment of the invention provides a data packet grouping control method, a storage medium and electronic equipment, and the method comprises the steps: extracting a current serial number of a received current data packet; in response to the fact that the current serial number is different from the first serial number and the difference value between the current serial number and the first serial number is smaller than a preset threshold value, a second serial number is updated based on the current serial number, the first serial number is the minimum serial number of the data packet to be recombined, and the second serial number is the next serial number of the maximum serial number of the received data packet; updating a count value of a preset counter to a difference value between the second serial number and the first serial number; in response to the fact that the counting value of the preset counter is larger than or equal to the counting threshold value, packaging operation is carried out on at least one first data packet, the first serial number is updated to be a second serial number, and the at least one first data packet comprises all data packets which are received and not recombined; the counting threshold is greater than or equal to the number of the data packets scheduled at the single transmission time interval.
Owner:CHINA SATENT NETWORK APPLICATION RESEARCH INSTITUTE CO LTD
Intelligent bill processing method and system based on multi-source data cross validation
The invention discloses an intelligent bill processing method and system based on multi-source data cross validation, and belongs to the technical field of financial data processing. The method comprises the following steps: obtaining and standardizing and aligning heterogeneous bill data in a plurality of data sources associated with the same transaction event to form a to-be-verified data unit; for a preset key field in the to-be-verified data unit, querying and obtaining a dynamic credibility score of each data source on the key field; grouping the data sources based on the numerical values of the key fields, and calculating the sum of weighted credibility of each group; based on the weighted creditworthiness sum, automatically deciding or triggering artificial arbitration to determine a credible value of the key field; and performing feedback updating on the dynamic credibility score of the data source on the key field by adopting a preset excitation and punishment rule according to a judgment or arbitration result. The technical problems that in the prior art, bill processing depends on a complex model, the decision-making process is opaque, and the adaptive learning ability is poor are solved.
Owner:HANGZHOU HANGCHENG TECHNOLOGY CO LTD
Network configuration verification acceleration method and device based on network domain knowledge
The present disclosure provides a network configuration verification acceleration method based on network domain knowledge, an electronic device and a readable storage medium. The method of the present disclosure comprises: screening the first configuration information in the network configuration information irrelevant to the network invariant to be verified, ignoring the first configuration information in the network configuration information, and obtaining the second configuration information; encoding the network invariant to be verified to obtain the first encoded information; encoding the second configuration information to obtain the second encoded information, wherein the numerical attribute information in the second configuration information is abstracted in the encoding process; combining the first encoded information and the second encoded information to obtain the encoded information; determining the search order of the first variable in the encoded information, and determining the assignment order of the first variable in the encoded information; selecting the first variable in the encoded information according to the search order, and then assigning the selected first variable according to the assignment order to verify the encoded information.
Owner:XIAMEN UNIV
AI processor operator overflow optimization method and system
The invention relates to the technical field of operator overflow control, in particular to an AI processor operator overflow optimization method and system.The method comprises the steps that a finite automaton is introduced into numerical values for state combination and sorting, a continuous state sequence covering an operator execution stack is generated, and an intermediate result obtains a state number carrying bit width section information and stack position information at the same time; the method comprises the following steps of: setting a section to which an intermediate value of each operator belongs to a discrete identifier in a fixed-point space, executing weighted truncation reduction transformation by introducing a graph neural network, and deducing a clipping weight under topological connection constraints among the operators, so that the intermediate value of each operator obtains differential correction strength in a longitudinal accumulation direction and a transverse link direction at the same time; an operator correction set is generated by guiding an intermediate numerical value through a clipping weight, so that the effective dynamic range utilization rate after right shift and truncation are matched and reduced is improved, and the stability and repeatability of a reasoning process in a long-chain accumulation scene and the tracking capability of an abnormal calculation behavior are enhanced.
Owner:FANGXIN TECH CO LTD
Chip test data writing method, chip test data reading method and chip test data writing equipment
PendingCN121705096AElectronic circuit testingDetecting faulty computer hardwareComputer hardwareDatasheet
The invention provides a chip test data writing method, a chip test data reading method and chip test data writing equipment. Relates to the technical field of chip testing, and is used for a metadata table and a test data table, the first dimension of the test data table is used for arranging multiple test items, the second dimension is used for setting multiple data steps, and the total number of the test items corresponding to each data step is smaller than or equal to a first numerical value. The metadata table is used for storing each test item included in the test data table and a main key value corresponding to the test item, the main key value is generated according to a to-be-written test item of the chip and a test result of the to-be-written test item, and the test item number of the test item and the main key value are correspondingly stored in the metadata table; according to the primary key value and the first numerical value, determining a target position of a test result in a test data table, the test result being a result obtained by executing the test item by the chip; and writing the test result in the target position of the test data table. Therefore, excessive rows or columns in the test data table are avoided, and the processing efficiency of the data table is improved.
Owner:HANGZHOU XINYI TECH CO LTD
A blueprint life cycle evidence method and system based on a consortium chain
The application discloses a kind of based on alliance chain's drawing life cycle evidence method and system, the method includes: the drawing to be stored in evidence is divided into multiple drawing blocks, the numerical characteristics of each block and are normalized, the numerical value of the whole and each drawing block of the drawing to be stored in evidence is matrixed, and the drawing digital fingerprint of the drawing to be stored in evidence is generated by digital fingerprint generation model;When the drawing to be stored in evidence is modified or updated, obtain the modification matrix difference, and through drawing version evolution model, the evolution value of each drawing version is calculated in conjunction with drawing digital fingerprint;For each access user, according to the identity key of user, authority level, drawing digital fingerprint and the evolution value of each drawing version, the authorized value of user access each drawing version is calculated, and according to authorized value, it is judged whether user can access corresponding drawing version.
Owner:GUANGDONG HIGHWAY CONSTR CO LTD +1
Data migration method and device, equipment, storage medium and program product
PendingCN121858526ASpecial data processing applicationsDatabase design/maintainanceDatasheetAlgorithm
The invention provides a data migration method and device, equipment, a storage medium and a program product. Relates to the technical field of data processing. The method comprises the steps that a plurality of first data tables in a first system and a plurality of second data tables in a second system are obtained, the first data tables comprise a plurality of first fields and numerical values of the first fields, and the second data tables comprise a plurality of second fields; determining a service type of each first field, and determining a mapping relationship between the plurality of first fields and the plurality of second fields according to the plurality of first fields and the plurality of service types; according to the mapping relation and the numerical value of each first field, determining the first numerical value of the second field with the mapping relation and the second field without the mapping relation; according to the first numerical value, determining a second numerical value of the second field without the mapping relationship; and the first numerical value and the second numerical value are migrated into the plurality of second data tables in the second system, so that the adaptation degree of data migration is improved.
Owner:CHINA CONSTRUCTION BANK +1
Metadata structure, data processing method and storage medium
PendingCN121722930AData processing applicationsMultimedia data clustering/classificationReal arithmeticEngineering
The invention provides a metadata structure, a data processing method and a storage medium. The metadata structure is a structured data object on a computer readable medium, can be read and written by a computer program, and adopts a framework of three parts: a basic index part, which is used for uniquely identifying educational resources and supporting retrieval, and at least comprises a unique identification field and a resource classification / retrieval attribute field; the core attainment vector part is used for quantitatively representing an educational resource childbearing target for intelligent calculation and comprises an N-dimensional attainment vector, the dimension key is c01 to cN, and the value of each dimension is a real number in a preset numerical value interval; and the association application part is used for constructing an inter-resource relationship and learning evidence agreement which at least comprises a field for describing a resource dependency relationship and a rule field which is executable by a computer and is used for extracting evidence from learning interaction data. According to the scheme, structuralization and computability of the attestation information are realized.
Owner:BEIJING SANSAN SMART EDUCATION TECHNOLOGY CO LTD
Active real-time firmware debugging method and tested device
The embodiment of the invention provides an active real-time firmware debugging method and a tested device, the active real-time firmware debugging method is used for actively monitoring the tested device in real time for debugging, and the method comprises the following steps: sequentially reading a plurality of monitoring point addresses stored in a plurality of monitoring point address buffers; sequentially reading a plurality of operation numerical values stored in a plurality of test point data buffers corresponding to the plurality of monitoring point addresses; temporarily storing the plurality of read operation numerical values in a plurality of monitoring point data buffers; and combining at least one part of the plurality of operation values in the plurality of monitoring point data buffers to generate a monitoring signal, and enabling the tested device to actively and continuously output the plurality of operation values in the monitoring signal in sequence through a preset output pin of the tested device, and the monitoring point data buffers are electrically connected with the test point data buffers.
Owner:NUVOTON
Data comparison method based on AI
The invention discloses a data comparison method based on AI, and the method comprises the steps: data preprocessing: cleaning and standardizing an original text into a regular text; text vectorization processing: converting the regular text into a numerical form which can be understood and calculated by a machine; calculating similarity and generating candidate pairs; a matching decision is made, whether the two texts point to the same entity or not is judged according to the calculated similarity scores, and a matching result is obtained; and performing post-processing and evaluation, performing optimization processing on the matching result, and scientifically evaluating the performance of the whole system. The tolerance to format difference and noise is improved, and automation and expandability are achieved.
Owner:BEIYIN FINANCIAL TECH CO LTD
Method and apparatus for transmitting indication information, and readable storage medium
Provided in the present disclosure are a method and apparatus for transmitting indication information, and a readable storage medium. The method includes: receiving indication information sent by a network device, where the indication information is used for indicating a physical downlink control channel (PDCCH) skipping duration in the form of a non-numerical value; and performing PDCCH skipping according to the indication information.
Owner:BEIJING XIAOMI MOBILE SOFTWARE CO LTD
Apparatus and method for number-theoretical transformation instructions
PendingCN121765173APublic key for secure communicationInstruction analysisAlgorithmTheoretical computer science
The invention relates to an apparatus and a method for number-theoretical transformation instructions. Instructions are executed for performing a number-theory transform (NTT). For example, one embodiment includes: a decode circuit to decode the instruction; one or more packed data registers to store a first, second, and third plurality of source packed data elements and corresponding result packed data elements; and execution circuitry to execute the NTT using a butterfly operation including Montgomery multiplication of a source packed data element of the second plurality of source packed data elements and a corresponding source packed data element of the third plurality of source packed data elements using a modulus value or an inverse of the modulus value to generate a product. The product is added to a corresponding source packed data element of the first plurality of source packed data elements to obtain a first result, and the product is subtracted from the corresponding source packed data element of the first plurality of source packed data elements to obtain a second result.
Owner:INTEL CORP
An efficient identification method of bit features based on byte anchoring
The application discloses a kind of based on byte anchoring bit feature efficient identification method, comprising the following steps: S1, the length m of bit feature to be identified, determine byte window length L, wherein m≥8, L=⌊(m+7) / 8⌋+1, wherein ⌊·⌋ it indicates the down rounding function;S2, construct second byte fast mapping table;S3, read L byte data in current window in data stream;S4, according to the numerical value of the second byte of current window, obtain the possible bit offset set, based on the possible bit offset set, carry out bit feature to be identified matching;S5, according to the matching result adjustment window sliding step, and return S3 continue to execute.By byte level step, second byte anchoring and fast mapping table, specially solve the technical bottleneck of low bit feature matching efficiency in network traffic, suitable for bit feature of medium length and above.
Owner:北京科来数据分析有限公司
Statistical method, device, electronic equipment, storage medium and program for running exception
The present disclosure provides a statistical method, device, electronic equipment, storage medium and program for operation exception. The method comprises: assigning a corresponding file number to each of a plurality of preset source code files, each line of code in each source code file having a corresponding code line number; in response to monitoring that any source code file runs abnormally, obtaining a corresponding exception type number, file number and abnormal code line number where the exception occurs; querying a target partition directory matching the exception type number from a plurality of preset partition directories, the target partition directory comprising a plurality of type directories, each type directory comprising a plurality of file directories, each file directory representing a source code file and comprising a plurality of line number directories, and each line number directory representing a code line; and adjusting an exception count value of a corresponding directory according to a directory order of the type directory, file directory and line number directory. The embodiment can comprehensively obtain detailed information of an exception event, facilitating subsequent tracing of the exception event.
Owner:UNIONTECH SOFTWARE TECH CO LTD
Artificial intelligence processor operator overflow optimization method and system
The application relates to the technical field of operator overflow control, in particular to an AI processor operator overflow optimization method and system, state combination and sequencing are performed on a numerical value introduction finite automaton, a continuous state sequence covering an operator execution stack is generated, a state number carrying bit width section information and stack position information at the same time is obtained for an intermediate result, a section to which an operator intermediate value belongs is provided with discrete identification in a fixed point space, a graph neural network is introduced to perform a weighted truncation reduction transformation, a clipping weight is derived under a topological connection constraint between operators, each operator intermediate value is simultaneously provided with differentiated correction strength in a longitudinal accumulation direction and a horizontal link direction, the intermediate value is guided to generate an operator correction set through the clipping weight, the utilization rate of an effective dynamic range after right shift and truncation cooperation reduction is improved, and the stability, repeatability and tracking ability for abnormal calculation behavior of an inference process in a long chain accumulation scene are enhanced.
Owner:FANGXIN TECH CO LTD
Data processing method, device and system and related equipment
PendingCN121411732ADigital data processing detailsMachine learningAlgorithmTheoretical computer science
The invention discloses a data processing method, device and system and related equipment, and relates to the technical field of data processing. And obtaining data to be processed, and calculating the data based on a calculation process indicated by the numerical algorithm to obtain a calculation result. Wherein in the process of calculating the data, at least one AI (artificial intelligence) operator is called to participate in calculation for the data, and each AI operator is used for replacing a multiplication or division operation module in a numerical algorithm. And the AI operator runs in a vector calculation unit or a matrix calculation unit. The addition and subtraction operations in the numerical algorithm are performed by the scalar calculation unit. The calculation amount generated by the AI operator is far less than the calculation amount generated by the multiplication or division operation module, so that the AI operator is used for replacing the AI operator, the data calculation overhead can be effectively reduced, and the data calculation time delay can be reduced. Meanwhile, the data calculation logic still meets the theoretical reasoning basis indicated by the numerical algorithm, and it can be guaranteed that the calculation precision reaches the high level.
Owner:HUAWEI TECH CO LTD
A monotonic counter, counting method, electronic device and storage medium
The application belongs to the technical field of monotonic counter, and particularly relates to a monotonic counter, a counting method, an electronic device and a storage medium. The counting method of the monotonic counter of the application, under the condition that no abnormal situation occurs, only counts between two storage blocks in turn, and does not need to perform additional storage blocks to back up the storage block counting value under the abnormal situation, thereby optimizing the operation process of counting. Under the condition that an abnormal situation occurs, firstly, whether the external address is the start address of the data bit of the storage block is judged, and then whether the storage block has an abnormality is judged. For the storage block without an abnormality, the counting value is obtained by counting the data bit, so as to ensure accurate counting. For the storage block with an abnormality during counting, the storage block is switched to another storage block for erasing and then switched back to the current storage block for counting. The counting method optimizes the processing steps of counting and abnormal situations, and reduces the logic complexity.
Owner:SHANGHAI XINCUN TIANXIA ELECTRONIC TECH CO LTD +1
A data processing method and related device for a multi-point shared register
The application discloses a data processing method of multi-point shared register and related devices, and relates to the technical field of software, which comprises the following steps: reading the sixteen-bit integer values of each register in a Modbus slave device, each register being occupied by at least one logical point and each logical point being pre-configured with position occupation information; converting the sixteen-bit integer values of each register into corresponding binary strings, and intercepting the binary string of each logical point from the binary string of the occupied register according to the position occupation information of each logical point; and converting the binary string of each logical point into a corresponding decimal value, and storing the decimal value of each logical point into a local cache. The application can accurately extract the values of different logical points in the scene of multi-point shared register, improve the utilization rate of register bit resources, reduce the cost of equipment, and maximize the number of accessible points.
Owner:青岛海尔暖通空调设备有限公司 +2
Contract extraction text quality evaluation method and device
One or more embodiments of the present specification provide a contract extraction text quality assessment method and device, the method comprising: obtaining a target extraction text of a target contract obtained by natural language processing, the target extraction text comprising a plurality of fields, each field comprising a name portion and a numerical portion; according to a preset evaluation mode rule base, an evaluation mode corresponding to each field of the target extraction text is determined, the preset evaluation mode rule base comprises evaluation modes corresponding to a plurality of historical fields, and the evaluation mode corresponding to each historical field is determined through evaluation effects of the historical field in different evaluation modes; the evaluation mode comprises a zero sample model evaluation mode, or a few sample model evaluation mode, or a zero sample model and few sample model fusion evaluation mode. And after the evaluation mode of each field is determined, determining a corresponding scoring model according to the evaluation mode of each field, and performing quality scoring on each field through the scoring model corresponding to each field.
Owner:QIANTANG CREDIT INFORMATION CO LTD