CN111460250A - Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus - Google Patents
Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus Download PDFInfo
- Publication number
- CN111460250A CN111460250A CN202010136003.3A CN202010136003A CN111460250A CN 111460250 A CN111460250 A CN 111460250A CN 202010136003 A CN202010136003 A CN 202010136003A CN 111460250 A CN111460250 A CN 111460250A
- Authority
- CN
- China
- Prior art keywords
- data
- portrait
- verification
- check
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a method, a device, a medium and electronic equipment for cleaning image data, belonging to the technical field of data processing, wherein the method comprises the following steps: inputting an portrait strategy of the target portrait service into a portrait service standardized model to obtain tag information; extracting key data characteristics according to the label information, and inputting the key data characteristics and the standard data characteristics into a data defect prediction model together to obtain a plurality of check points of the portrait data and check scores corresponding to each check point; respectively acquiring a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point; respectively acquiring a verification template of a series matched with the verification fraction from a target multi-stage verification template corresponding to each verification point; checking the portrait data corresponding to each check point based on the check template to obtain a plurality of cleaning information of the portrait data, and cleaning the portrait data based on the cleaning information. The application effectively promotes the cleaning accuracy of the data for portrait.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, a medium, and an electronic device for cleaning image data.
Background
Data cleansing is a process of cleansing a large amount of data for a certain service according to a service standard, for example, finding and removing useless redundant data in redundant data or repairing error data. The user portrait is a process of clustering and analyzing massive user data information, abstracting and labeling the data, and then constructing a comprehensive and accurate user portrait system by using the labels. Therefore, the user portrait data can be cleaned, and the user portrait accuracy can be guaranteed. At present, data used for user portrait cleaning is rapidly checked and sorted mainly through a data tag, and the problem that the data cleaning is difficult to adjust according to specific user portrait services and is low in cleaning accuracy exists.
It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present application and therefore may include information that does not constitute prior art known to a person of ordinary skill in the art.
Disclosure of Invention
The application aims to provide a cleaning scheme for image data, and further realizes the cleaning accuracy of the image data at least to a certain extent.
According to an aspect of the present application, there is provided a method for cleaning data of an image, including:
when the portrait data used for the target portrait service is crawled, inputting the portrait strategy of the target portrait service into a portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified;
extracting key data characteristics of the image data according to the label information of the data to be verified, and inputting the key data characteristics and the standard data characteristics of the target image service into a data defect prediction model together to obtain a plurality of verification points of the image data and verification scores corresponding to each verification point;
respectively acquiring a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
according to the corresponding check fraction of each check point, acquiring a check template with the number of stages matched with the check fraction from the target multi-stage check template corresponding to each check point;
checking the portrait data corresponding to each checking point sequentially based on the matched series checking template to obtain a plurality of cleaning information of the portrait data, and cleaning the portrait data based on the cleaning information.
In an exemplary embodiment of the present application, inputting the portrait policy of the target portrait service into a portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified, including:
obtaining a portrait policy of the target portrait service meeting a predetermined standard, the portrait policy indicating a policy for portraying based on the portrait data;
normalizing the portrayal policy into a data stream consistent with a logical order of the portrayal policy;
and inputting the data stream into an portrait service standardized model to obtain a multi-level data verification template and label information of data to be verified.
In an exemplary embodiment of the present application, extracting key data features of the image data according to tag information of the data that must be verified, and inputting the key data features and the standard data features of the target portrait service together into a data defect prediction model, includes:
acquiring data attribute information from the label information of the data to be verified;
extracting attribute value data corresponding to the data attribute information from the image data as key data features;
and inputting the key data characteristics and preset standard data characteristics of the target portrait service into a data defect prediction model together.
In an exemplary embodiment of the present application, extracting key data features of the image data according to tag information of the data that must be verified, and inputting the key data features and the standard data features of the target portrait service together into a data defect prediction model, includes:
extracting key data characteristics of the image data according to the label information of the data to be verified;
extracting standard key data features from the standard data features of the target image service according to the label information of the data to be verified;
performing same attribute data association on the key data features and the standard key data features to obtain associated features;
and inputting the associated features into a data defect prediction model.
In an exemplary embodiment of the present application, obtaining, according to a check score corresponding to each check point, a series check template matching the check score from the target multilevel check template corresponding to each check point respectively includes:
acquiring an association table of check fractions and series corresponding to the target check point;
querying the correlation table for the grade number correlated with the check score of the target check point;
and acquiring a verification template corresponding to the progression associated with the verification fraction of the target verification point from the multilevel verification template corresponding to the target verification point.
In an exemplary embodiment of the present application, cleansing the portrait data based on the cleansing information includes:
searching a data cleaning method matched with the cleaning information from a cleaning database according to the cleaning information;
and cleaning the image data according to the matched data cleaning method.
In an exemplary embodiment of the present application, a multi-level data verification template includes:
a data verification template corresponding to a multi-level verification depth of data verification points, the multi-level verification depth indicating different size ranges of data verification.
In an exemplary embodiment of the present application, a method for training a portrait service standardization model includes:
collecting an image strategy sample set of target image service, wherein each sample in the sample set is calibrated with a corresponding multi-level data verification template and label information of data to be verified in advance;
inputting the input data of each sample in the sample set into a machine learning model to obtain a predicted multistage data verification template corresponding to each sample and information of data to be verified;
if the predicted multilevel data verification template and the information of the data to be verified corresponding to the sample obtained after the sample is input into the machine learning model are inconsistent with the label information of the multilevel data verification template and the label information of the data to be verified calibrated in advance for the sample, adjusting the coefficient of the machine learning model to make the data consistent;
and when all the samples are input into the machine learning model, the obtained information of the multi-level data verification template predicted by the samples and the data to be verified is consistent with the label information of the multi-level data verification template calibrated in advance for the samples and the data to be verified, and the training is finished.
In an exemplary embodiment of the present application, a training method of a data defect prediction model includes:
collecting a key data characteristic and standard data characteristic sample set of target portrait service, wherein each sample in the sample set is calibrated with a corresponding check point and a corresponding check fraction in advance;
inputting the input data of each sample in the sample set into a machine learning model to obtain a predicted check point and a corresponding check score corresponding to each sample;
if the predicted check point and the corresponding check score corresponding to the sample obtained after the sample is input into the machine learning model are inconsistent with the check point and the corresponding check score calibrated in advance for the sample, adjusting the coefficient of the machine learning model to make the check points and the corresponding check score consistent;
and after all the samples are input into the machine learning model, the obtained check points and corresponding check scores predicted by the samples are consistent with the check points and corresponding check scores calibrated in advance for the samples, and the training is finished.
According to an aspect of the present application, there is provided a cleaning apparatus for data of an image, comprising:
the strategy analysis module is used for inputting the portrait strategy of the target portrait service into the portrait service standardized model when crawling to the portrait data used for the target portrait service to obtain a multi-level data verification template and tag information of data to be verified;
the defect analysis module is used for extracting key data characteristics of the portrait data according to the label information of the data to be verified, inputting the key data characteristics of the portrait data and the standard data characteristics of the target portrait service into a data defect prediction model together, and obtaining a plurality of check points of the portrait data and check scores corresponding to each check point;
the acquisition module is used for respectively acquiring a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
the matching module is used for respectively acquiring a check template with the series matched with the check fraction from the target multi-stage check template corresponding to each check point according to the check fraction corresponding to each check point;
and the cleaning module is used for verifying the portrait data corresponding to each verification point in sequence based on the matched series verification templates to obtain a plurality of cleaning information of the portrait data so as to clean the portrait data based on the cleaning information.
According to an aspect of the application, there is provided a computer readable storage medium having stored thereon program instructions, characterized in that the program instructions, when executed by a processor, implement the method of any of the above.
According to an aspect of the present application, there is provided an electronic device, comprising:
a processor; and
a memory for storing program instructions for the processor; wherein the processor is configured to perform any of the methods described above via execution of the program instructions.
The application relates to a method and a related device for cleaning image data, which automatically analyze an image strategy of a target image service through an image service standardized model to obtain a multi-level data verification template adaptive to a current image strategy and label information of data to be verified; secondly, through a data defect prediction model, combining and analyzing key data characteristics of image data extracted according to label information of data to be verified and standard data characteristics of target image service together to obtain a plurality of verification points of the current image data and verification scores corresponding to each verification point; then, after a target multilevel data verification template of each verification point in a multilevel data verification template which is suitable for the current portrait strategy is obtained, a multilevel verification template of which the verification fraction of each verification point is matched with that of the target multilevel data verification template is obtained; and finally, respectively checking the portrait data corresponding to each checking point based on the matched series checking template to obtain a plurality of cleaning information of the portrait data, and further cleaning the portrait data based on the cleaning information. By the mode, the image data can be checked efficiently and accurately, and the accuracy of cleaning the image data is effectively improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 schematically shows a flow chart of a cleaning method for data of an image.
Fig. 2 schematically shows an application scenario example of a cleaning method for data of an image.
Fig. 3 schematically shows a flow chart of a method of data entry.
Fig. 4 schematically shows a block diagram of a cleaning device for image data.
Fig. 5 schematically shows an example block diagram of an electronic device for implementing the above-described cleaning method for data of an image.
Fig. 6 schematically shows a computer-readable storage medium for implementing the above-described cleaning method for data of an image.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the subject matter of the present application can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present application.
Furthermore, the drawings are merely schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
In this exemplary embodiment, a method for cleaning image data is first provided, and the method for cleaning image data may be executed on a server, or may also be executed on a server cluster or a cloud server, and of course, a person skilled in the art may also execute the method of the present invention on other platforms as needed, which is not limited in this exemplary embodiment. Referring to fig. 1, the method for cleaning data for an image may include the steps of:
step S110, when the portrait data used for the target portrait service is crawled, the portrait strategy of the target portrait service is input into a portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified;
step S120, extracting key data characteristics of the portrait data according to the label information of the data to be verified, inputting the key data characteristics of the portrait data and the standard data characteristics of the target portrait service into a data defect prediction model together, and obtaining a plurality of check points of the portrait data and check scores corresponding to each check point;
step S130, respectively obtaining a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
step S140, acquiring a series of check templates matched with the check scores from the target multi-stage check templates corresponding to each check point according to the check scores corresponding to the check points;
and S150, checking the portrait data corresponding to each checking point sequentially based on the matched series checking template to obtain a plurality of cleaning information of the portrait data, and cleaning the portrait data based on the cleaning information.
Hereinafter, each step in the above-described cleaning method for image data in the present exemplary embodiment will be explained and explained in detail with reference to the drawings.
In step S110, when the portrait data for the target portrait service is crawled, the portrait policy of the target portrait service is input into the portrait service standardized model, and a multi-level data verification template and tag information of the data to be verified are obtained.
In the present exemplary embodiment, referring to fig. 2, when crawling the server 201 from the server 202 to the portrait data for the target portrait service, the portrait policy of the target portrait service is obtained, and the portrait policy of the target portrait service is input into the pre-trained portrait service standardized model, so as to obtain the multi-level data verification template and the tag information of the data to be verified. Therefore, in the subsequent steps, the server 201 performs data cleaning processing according to the multi-level data verification template and the label information of the data to be verified. It is understood that, the server 201 and the server 202 may be any devices with processing capability, such as computers, microprocessors, etc., and are not limited thereto.
The target representation service is a service for representing a different target for a user, and for example, risk representation or consumption habit representation is performed for a user on a certain platform. The data required for different rendering services is different, while rendering strategies for the same data may also differ for design reasons. The portrait policy of the target portrait service is a policy for analyzing tag data of a user, for example, probability statistics is performed on some tags, and then an analysis method such as pruning is performed to obtain a target result. When a user performs portrait work on a target portrait service, a corresponding portrait policy is determined first, and the portrait policy may be described according to a predetermined description rule, for example, the portrait policy according to the predetermined description rule may be (1) a user tag set: age, sex, etc.; (2) carrying out label probability statistics; (3) a high probability label; (4) carrying out risk label statistics; (5) whether it is a risky user.
The portrait service standardized model is a machine learning model which is trained in advance and used for analyzing the checking condition of portrait data according to portrait strategies and outputting a data checking template and information which needs to check the data.
Input data (such as feature vector data) of the portrait policy is input into the portrait service standardized model, and information (such as A, B1, C and the like) of a predicted data verification template and information (such as identification of information such as income, debt and the like) of data needing to be verified can be obtained.
The multi-level data verification template is a data verification template with different verification ranges and verification templates with different coefficients, for example, the template a is a first level, the template B1 is a second level, the template B2 is a second level, the template C is a third level, the template verification range of the first level is the maximum, and meanwhile, the template a, the template B1, the template B2 and the template C are multi-level templates of the same series; it is understood that there are other series of multi-level templates, corresponding to different data checkpoints. The data verification template is an algorithm detection template for judging whether the data of the user portrait meets the requirements or not. The method for obtaining the multilevel data verification template can be to call from a database according to a template identifier (label information of the multilevel data verification template) output by the portrait service standardized model. The required verification information is information which is required to be verified for the image service in the image data of the current user, for example, whether data corresponding to certain attributes are complete or whether data of sub-attributes of certain attributes are complete or not. Other information is not verified to have no effect on the portrayal strategy.
In one embodiment, the method for inputting the portrait policy of the target portrait service into the portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified includes:
obtaining a portrait policy of the target portrait service meeting a predetermined standard, the portrait policy indicating a policy for portraying based on the portrait data;
normalizing the portrayal policy into a data stream consistent with a logical order of the portrayal policy;
and inputting the data stream into an portrait service standardized model to obtain a multi-level data verification template and label information of data to be verified.
The portrait strategy of the target portrait service meeting the preset standard can be manually filled by a user according to the standard, or the preset portrait strategy of the target portrait service can be directly crawled from a database. The image strategy describes a strategy for performing image analysis on the image data of the target image service at this time, namely processing strategy information of the image data.
The portrait policy is normalized into a data stream consistent with the logical sequence of the portrait policy, which may be obtained by converting portrait action attributes (such as rejection, classification, etc.) and portrait noun attributes (such as low probability, a certain name, etc.) into corresponding identifiers (such as numeric identifiers or character string identifiers), then concatenating the action attribute identifiers and the noun attribute identifiers having an association relationship, and finally concatenating all the identifiers according to the logical sequence to obtain the data stream.
Then, the data stream can be input into the image service standardized model to obtain a multi-level data verification template and label information of the data to be verified.
In one embodiment, the method for training the portrait service standardization model comprises the following steps:
collecting an image strategy sample set of target image service, wherein each sample in the sample set is calibrated with a corresponding multi-level data verification template and label information of data to be verified in advance;
inputting the input data of each sample in the sample set into a machine learning model to obtain a predicted multistage data verification template corresponding to each sample and information of data to be verified;
if the predicted multilevel data verification template and the information of the data to be verified corresponding to the sample obtained after the sample is input into the machine learning model are inconsistent with the label information of the multilevel data verification template and the label information of the data to be verified calibrated in advance for the sample, adjusting the coefficient of the machine learning model to make the data consistent;
and when all the samples are input into the machine learning model, the obtained information of the multi-level data verification template predicted by the samples and the data to be verified is consistent with the label information of the multi-level data verification template calibrated in advance for the samples and the data to be verified, and the training is finished.
In step S120, extracting key data features of the image data according to the tag information of the data to be verified, and inputting the key data features and the standard data features of the target image service together into a data defect prediction model to obtain a plurality of verification points of the image data and a verification score corresponding to each verification point.
In the embodiment of the present example, according to the data attribute tag in the tag information of the data that must be checked, keyword data of a corresponding attribute may be extracted from the portrait data for a user portrait as a key data feature.
The standard data feature of the target portrait service corresponds to the portrait policy, and is a key feature of different attribute data required by a preset portrait, such as the number of certain data or the unit of each data.
The data defect prediction model is a pre-trained machine learning model which contrasts and analyzes key data in the portrait data according to key data features and standard data features and has vulnerability suspicion, and due to the fact that the user portrait data volume is large and the relevance of all data is strong, the machine learning model can be used for efficiently and accurately predicting check points (vulnerability suspicion data) and analyzing the possibility (check fraction) that each check point has the vulnerability. The check score may be, for example, 0-100 points or may be a character representing a score.
In one embodiment, extracting key data features of the image data according to the tag information of the data which needs to be verified, and inputting the key data features and the standard data features of the target image service together into a data defect prediction model, the method includes:
acquiring data attribute information from the label information of the data to be verified;
extracting attribute value data corresponding to the data attribute information from the image data as key data features;
and inputting the key data characteristics and preset standard data characteristics of the target portrait service into a data defect prediction model together.
The data attribute information may be obtained from tag information of the data to be verified, such as image motion attributes (e.g., culling, classification, etc.) and image noun attributes (e.g., low probability, certain name, etc.).
Then, extracting attribute value data corresponding to the data attribute information from the image data as key data characteristics; the attribute value data is an attribute value corresponding to each attribute information, and for example, a low probability may correspond to less than 50%, or the like. And the extracted key data characteristics and preset standard data characteristics of the target portrait service can be input into the data defect prediction model together for analysis.
In one embodiment, extracting key data features of the image data according to the tag information of the data that must be verified, and inputting the key data features and the standard data features of the target image service together into a data defect prediction model, with reference to fig. 3, the method includes:
step S310, extracting key data characteristics of the image data according to the label information of the data to be verified;
step S320, extracting standard key data characteristics from the standard data characteristics of the target image service according to the label information of the data to be verified;
step S330, performing same attribute data association on the key data characteristics and the standard key data characteristics to obtain associated characteristics;
step S340, inputting the relevant features into a data defect prediction model.
The standard key data features are consistent with key data features of image data extracted according to tag information of data to be verified, namely image data features and standard image data features with the same attribute.
And performing same attribute data association on the key data features and the standard key data features to obtain associated features, for example, performing associated storage on the key data features belonging to the attribute A and data in the standard key data features in a serial or parallel storage manner, so as to obtain associated features with a comparison relationship, and inputting the associated features into the data defect prediction model.
In one embodiment, the training method of the data defect prediction model comprises the following steps:
collecting a key data characteristic and standard data characteristic sample set of target portrait service, wherein each sample in the sample set is calibrated with a corresponding check point and a corresponding check fraction in advance;
inputting the input data of each sample in the sample set into a machine learning model to obtain a predicted check point and a corresponding check score corresponding to each sample;
if the predicted check point and the corresponding check score corresponding to the sample obtained after the sample is input into the machine learning model are inconsistent with the check point and the corresponding check score calibrated in advance for the sample, adjusting the coefficient of the machine learning model to make the check points and the corresponding check score consistent;
and after all the samples are input into the machine learning model, the obtained check points and corresponding check scores predicted by the samples are consistent with the check points and corresponding check scores calibrated in advance for the samples, and the training is finished.
In step S130, a target multi-level data verification template in the multi-level data verification templates corresponding to each verification point is respectively obtained according to a preset template and verification point relation comparison table.
In the embodiment of the present example, the preset template and check point relation lookup table stores the multi-level data check template associated with each check point during checking in an associated manner. For example, the check points of the user age data are associated with corresponding multi-level age check templates. Therefore, the multi-stage verification templates corresponding to the predicted verification points can be obtained, and data can be accurately verified.
In step S140, according to the verification score corresponding to each verification point, a verification template of a series matching with the verification score is obtained from the target multi-stage verification templates corresponding to each verification point.
In the embodiment of the present example, the verification score corresponding to each verification point reflects the degree to which the verification point needs to be verified, and the higher the score is, the deeper the degree to which verification is needed is. Meanwhile, the higher the level of the multi-level verification template is, the larger the covered verification range is, the deeper the verification degree is, the larger the verification load is, and the lower the efficiency is. Therefore, the calibration template of the number of stages corresponding to the calibration degree is selected according to the corresponding calibration score of the calibration point, and for example, the calibration template of the target number of stages of each calibration point can be obtained through a calibration score and stage mapping table. In this way, the checking quality and efficiency can be effectively ensured.
In one embodiment, obtaining a calibration template with a series matching with a calibration score from the target multi-stage calibration template corresponding to each calibration point according to the calibration score corresponding to each calibration point includes:
acquiring an association table of check fractions and series corresponding to the target check point;
querying the correlation table for the grade number correlated with the check score of the target check point;
and acquiring a verification template corresponding to the progression associated with the verification fraction of the target verification point from the multilevel verification template corresponding to the target verification point.
The association table of the check score and the progression stores the association relationship of the standard score and progression, such as 10-score association 1 grade, 50-score association 5 grade, etc. Each check point is provided with a corresponding association table of check scores and series, the series associated with the check score of the target check point can be inquired from the association table, and then the check template corresponding to the series associated with the check score of the target check point is obtained from the multi-stage check template corresponding to the target check point.
In step S150, the image data corresponding to each of the verification points is sequentially verified based on the matched series verification templates to obtain a plurality of cleaning information of the image data, so as to clean the image data based on the cleaning information.
By utilizing the target series verification template of each verification point, whether the data corresponding to the verification point has problems and the existing problems (cleaning information) can be judged through scanning verification. For example, the data corresponding to the purchase record label of the user is lacking, or the unit is not in conformity. The user can then be accurately instructed to do the data cleansing.
By the method, the image data can be efficiently and accurately checked, and the accuracy of image data cleaning is effectively improved.
In one embodiment, cleansing the representation data based on the cleansing information includes:
matching a data cleaning method from a cleaning database according to the cleaning information;
and cleaning the image data according to the matched data cleaning method.
And performing similarity matching calculation on the cleaning information and cleaning information samples in the cleaning database to obtain similarity cleaning information samples, wherein the cleaning information samples correspond to one or more data cleaning methods, namely matched data cleaning schemes.
In an embodiment, based on the foregoing scheme, the multi-level data verification template includes:
a data verification template corresponding to a multi-level verification depth of data verification points, the multi-level verification depth indicating different size ranges of data verification.
The data verification templates with multiple levels of verification depths may be verification templates with different parameters or coefficients, and the verification templates with different verification depths have different verified data size ranges for the same verification point, for example, for multi-level attribute data, the verification templates with different levels may correspond to data with attributes of corresponding levels.
The application also provides a cleaning device for the image data. Referring to fig. 4, the cleaning apparatus for the imaged data may include a policy analysis module 410, a defect analysis module 420, an acquisition module 430, a matching module 440, and a cleaning module 450. Wherein:
the strategy analysis module 410 is used for inputting the portrait strategy of the target portrait service into the portrait service standardized model when crawling to the portrait data used for the target portrait service, and obtaining a multi-level data verification template and tag information of data to be verified;
the defect analysis module 420 is configured to extract key data features of the portrait data according to the tag information of the data to be verified, and input the key data features and the standard data features of the target portrait service together into a data defect prediction model to obtain a plurality of verification points of the portrait data and verification scores corresponding to each verification point;
the obtaining module 430 is configured to obtain a target multi-level data verification template in the multi-level data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
the matching module 440 is configured to obtain, according to a check score corresponding to each check point, a series of check templates matching the check score from the target multi-stage check templates corresponding to each check point;
the cleaning module 450 is configured to sequentially check the portrait data corresponding to each of the check points based on the matched series check templates to obtain a plurality of cleaning information of the portrait data, so as to clean the portrait data based on the cleaning information.
The specific details of each module in the above cleaning apparatus for image data have been described in detail in the corresponding cleaning method for image data, and therefore are not described herein again.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the application. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Moreover, although the steps of the methods herein are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.
In an exemplary embodiment of the present application, there is also provided an electronic device capable of implementing the above method.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
An electronic device 500 according to this embodiment of the invention is described below with reference to fig. 5. The electronic device 500 shown in fig. 5 is only an example and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 5, the electronic device 500 is embodied in the form of a general purpose computing device. The components of the electronic device 500 may include, but are not limited to: the at least one processing unit 510, the at least one memory unit 520, and a bus 530 that couples various system components including the memory unit 520 and the processing unit 510.
Wherein the storage unit stores program code that is executable by the processing unit 510 to cause the processing unit 510 to perform steps according to various exemplary embodiments of the present invention as described in the above section "exemplary methods" of the present specification. For example, the processing unit 510 may perform the steps as shown in fig. 1.
The memory unit 520 may include a readable medium in the form of a volatile memory unit, such as a random access memory unit (RAM)5201 and/or a cache memory unit 5202, and may further include a read only memory unit (ROM) 5203.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to make a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) execute the method according to the embodiments of the present application.
In an exemplary embodiment of the present application, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above section "exemplary methods" of the present description, when said program product is run on the terminal device.
Referring to fig. 6, a program product 600 for implementing the above method according to an embodiment of the present invention is described, which may employ a portable compact disc read only memory (CD-ROM) and include program code, and may be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including AN object oriented programming language such as Java, C + +, or the like, as well as conventional procedural programming languages, such as the "C" language or similar programming languages.
Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
Claims (10)
1. A method for cleansing data for an image, comprising:
when the portrait data used for the target portrait service is crawled, inputting the portrait strategy of the target portrait service into a portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified;
extracting key data characteristics of the image data according to the label information of the data to be verified, and inputting the key data characteristics and the standard data characteristics of the target image service into a data defect prediction model together to obtain a plurality of verification points of the image data and verification scores corresponding to each verification point;
respectively acquiring a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
according to the corresponding check fraction of each check point, acquiring a check template with the number of stages matched with the check fraction from the target multi-stage check template corresponding to each check point;
checking the portrait data corresponding to each checking point sequentially based on the matched series checking template to obtain a plurality of cleaning information of the portrait data, and cleaning the portrait data based on the cleaning information.
2. The method of claim 1, wherein the inputting the portrait policy of the target portrait service into a portrait service standardized model to obtain a multi-level data verification template and tag information of data to be verified comprises:
obtaining a portrait policy of the target portrait service meeting a predetermined standard, the portrait policy indicating a policy for portraying based on the portrait data;
normalizing the portrayal policy into a data stream consistent with a logical order of the portrayal policy;
and inputting the data stream into an portrait service standardized model to obtain a multi-level data verification template and label information of data to be verified.
3. The method of claim 1, wherein the extracting key data features of the image data according to the tag information of the data to be verified and inputting the key data features and the standard data features of the target image service together into a data defect prediction model comprises:
acquiring data attribute information from the label information of the data to be verified;
extracting attribute value data corresponding to the data attribute information from the image data as key data features;
and inputting the key data characteristics and preset standard data characteristics of the target portrait service into a data defect prediction model together.
4. The method of claim 1, wherein the extracting key data features of the image data according to the tag information of the data to be verified and inputting the key data features and the standard data features of the target image service together into a data defect prediction model comprises:
extracting key data characteristics of the image data according to the label information of the data to be verified;
extracting standard key data features from the standard data features of the target image service according to the label information of the data to be verified;
performing same attribute data association on the key data features and the standard key data features to obtain associated features;
and inputting the associated features into a data defect prediction model.
5. The method according to claim 1, wherein the obtaining, according to the check score corresponding to each check point, a check template with a series matching with the check score from the target multilevel check template corresponding to each check point respectively comprises:
acquiring an association table of check fractions and series corresponding to the target check point;
querying the correlation table for the grade number correlated with the check score of the target check point;
and acquiring a verification template corresponding to the progression associated with the verification fraction of the target verification point from the multilevel verification template corresponding to the target verification point.
6. The method of claim 1, wherein said cleansing said image data based on said cleansing information comprises:
searching a data cleaning method matched with the cleaning information from a cleaning database according to the cleaning information;
and cleaning the image data according to the matched data cleaning method.
7. The method of any of claims 1 to 6, wherein the multi-level data verification template comprises:
a data verification template corresponding to a multi-level verification depth of data verification points, the multi-level verification depth indicating different size ranges of data verification.
8. A cleaning apparatus for data of an image, comprising:
the strategy analysis module is used for inputting the portrait strategy of the target portrait service into the portrait service standardized model when crawling to the portrait data used for the target portrait service to obtain a multi-level data verification template and tag information of data to be verified;
the defect analysis module is used for extracting key data characteristics of the portrait data according to the label information of the data to be verified, inputting the key data characteristics of the portrait data and the standard data characteristics of the target portrait service into a data defect prediction model together, and obtaining a plurality of check points of the portrait data and check scores corresponding to each check point;
the acquisition module is used for respectively acquiring a target multilevel data verification template in the multilevel data verification templates corresponding to each verification point according to a preset template and verification point relation comparison table;
the matching module is used for respectively acquiring a check template with the series matched with the check fraction from the target multi-stage check template corresponding to each check point according to the check fraction corresponding to each check point;
and the cleaning module is used for verifying the portrait data corresponding to each verification point in sequence based on the matched series verification templates to obtain a plurality of cleaning information of the portrait data so as to clean the portrait data based on the cleaning information.
9. A computer readable storage medium having stored thereon program instructions, characterized in that the program instructions, when executed by a processor, implement the method of any of claims 1-7.
10. An electronic device, comprising:
a processor; and
a memory for storing program instructions for the processor; wherein the processor is configured to perform the method of any of claims 1-7 via execution of the program instructions.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010136003.3A CN111460250B (en) | 2020-03-02 | 2020-03-02 | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus |
| PCT/CN2020/117428 WO2021174812A1 (en) | 2020-03-02 | 2020-09-24 | Data cleaning method and apparatus for profile, and medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010136003.3A CN111460250B (en) | 2020-03-02 | 2020-03-02 | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111460250A true CN111460250A (en) | 2020-07-28 |
| CN111460250B CN111460250B (en) | 2022-07-08 |
Family
ID=71679988
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010136003.3A Active CN111460250B (en) | 2020-03-02 | 2020-03-02 | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111460250B (en) |
| WO (1) | WO2021174812A1 (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111984640A (en) * | 2020-08-04 | 2020-11-24 | 中国科学技术大学智慧城市研究院(芜湖) | Portrait construction method based on multi-element heterogeneous data |
| CN112446744A (en) * | 2020-12-14 | 2021-03-05 | 成都航天科工大数据研究院有限公司 | Method, system and medium for constructing enterprise portrait based on industrial product supply and demand platform |
| CN112651641A (en) * | 2020-12-29 | 2021-04-13 | 车主邦（北京）科技有限公司 | Method and device for processing portrait management data |
| CN112732755A (en) * | 2020-12-30 | 2021-04-30 | 招商局金融科技有限公司 | Client grouping-based label value matching joint verification method and device and computer equipment |
| CN113192622A (en) * | 2021-05-08 | 2021-07-30 | 上海亿为科技有限公司 | Method, device and equipment for checking medical data in AR (augmented reality) inspection based on cloud edge |
| WO2021174812A1 (en) * | 2020-03-02 | 2021-09-10 | 平安科技（深圳）有限公司 | Data cleaning method and apparatus for profile, and medium and electronic device |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113849732A (en) * | 2021-09-23 | 2021-12-28 | 北京思博知网科技有限公司 | Enterprise portrait establishing method and system |
| CN113988070B (en) * | 2021-10-09 | 2023-05-05 | 广州快决测信息科技有限公司 | Investigation problem generation method, investigation problem generation device, computer equipment and storage medium |
| CN115098476A (en) * | 2022-06-23 | 2022-09-23 | 中核核电运行管理有限公司 | Data cleaning method and device for integrating production data of nuclear power station with multiple sources |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120150825A1 (en) * | 2010-12-13 | 2012-06-14 | International Business Machines Corporation | Cleansing a Database System to Improve Data Quality |
| US20170308557A1 (en) * | 2016-04-21 | 2017-10-26 | LeanTaas | Method and system for cleansing and de-duplicating data |
| CN109003027A (en) * | 2018-07-16 | 2018-12-14 | 江苏满运软件科技有限公司 | A kind of management method and system of label of drawing a portrait |
| CN109558396A (en) * | 2018-10-24 | 2019-04-02 | 深圳市万屏时代科技有限公司 | A kind of user demand data cleaning method and system |
| CN109933699A (en) * | 2019-03-05 | 2019-06-25 | 中国科学院文献情报中心 | A kind of construction method and device of academic portrait model |
| CN110070391A (en) * | 2019-04-17 | 2019-07-30 | 同盾控股有限公司 | Data processing method, device, computer-readable medium and electronic equipment |
| CN110263024A (en) * | 2019-05-20 | 2019-09-20 | 平安普惠企业管理有限公司 | Data processing method, terminal device and computer storage medium |
| CN110427358A (en) * | 2019-02-22 | 2019-11-08 | 北京沃东天骏信息技术有限公司 | Data cleaning method and device and information recommendation method and device |
| CN110555019A (en) * | 2019-09-12 | 2019-12-10 | 成都中科大旗软件股份有限公司 | Data cleaning method based on service end |
| CN110781368A (en) * | 2019-10-22 | 2020-02-11 | 北京赛时科技有限公司 | Information crawling system and method for specified experts |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109522746A (en) * | 2018-11-07 | 2019-03-26 | 平安医疗健康管理股份有限公司 | A kind of data processing method, electronic equipment and computer storage medium |
| CN111460250B (en) * | 2020-03-02 | 2022-07-08 | 平安科技（深圳）有限公司 | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus |
-
2020
- 2020-03-02 CN CN202010136003.3A patent/CN111460250B/en active Active
- 2020-09-24 WO PCT/CN2020/117428 patent/WO2021174812A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120150825A1 (en) * | 2010-12-13 | 2012-06-14 | International Business Machines Corporation | Cleansing a Database System to Improve Data Quality |
| US20170308557A1 (en) * | 2016-04-21 | 2017-10-26 | LeanTaas | Method and system for cleansing and de-duplicating data |
| CN109003027A (en) * | 2018-07-16 | 2018-12-14 | 江苏满运软件科技有限公司 | A kind of management method and system of label of drawing a portrait |
| CN109558396A (en) * | 2018-10-24 | 2019-04-02 | 深圳市万屏时代科技有限公司 | A kind of user demand data cleaning method and system |
| CN110427358A (en) * | 2019-02-22 | 2019-11-08 | 北京沃东天骏信息技术有限公司 | Data cleaning method and device and information recommendation method and device |
| CN109933699A (en) * | 2019-03-05 | 2019-06-25 | 中国科学院文献情报中心 | A kind of construction method and device of academic portrait model |
| CN110070391A (en) * | 2019-04-17 | 2019-07-30 | 同盾控股有限公司 | Data processing method, device, computer-readable medium and electronic equipment |
| CN110263024A (en) * | 2019-05-20 | 2019-09-20 | 平安普惠企业管理有限公司 | Data processing method, terminal device and computer storage medium |
| CN110555019A (en) * | 2019-09-12 | 2019-12-10 | 成都中科大旗软件股份有限公司 | Data cleaning method based on service end |
| CN110781368A (en) * | 2019-10-22 | 2020-02-11 | 北京赛时科技有限公司 | Information crawling system and method for specified experts |
Non-Patent Citations (1)
| Title |
|---|
| 雷兵 等: "基于网络店铺订单数据的群体用户画像构建研究", 《河南工业大学学报( 社会科学版)》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021174812A1 (en) * | 2020-03-02 | 2021-09-10 | 平安科技（深圳）有限公司 | Data cleaning method and apparatus for profile, and medium and electronic device |
| CN111984640A (en) * | 2020-08-04 | 2020-11-24 | 中国科学技术大学智慧城市研究院(芜湖) | Portrait construction method based on multi-element heterogeneous data |
| CN112446744A (en) * | 2020-12-14 | 2021-03-05 | 成都航天科工大数据研究院有限公司 | Method, system and medium for constructing enterprise portrait based on industrial product supply and demand platform |
| CN112446744B (en) * | 2020-12-14 | 2023-08-08 | 成都航天科工大数据研究院有限公司 | Method, system and medium for constructing enterprise portraits based on industrial product supply and demand platform |
| CN112651641A (en) * | 2020-12-29 | 2021-04-13 | 车主邦（北京）科技有限公司 | Method and device for processing portrait management data |
| CN112732755A (en) * | 2020-12-30 | 2021-04-30 | 招商局金融科技有限公司 | Client grouping-based label value matching joint verification method and device and computer equipment |
| CN112732755B (en) * | 2020-12-30 | 2024-03-22 | 招商局金融科技有限公司 | Label value matching joint verification method and device based on customer grouping and computer equipment |
| CN113192622A (en) * | 2021-05-08 | 2021-07-30 | 上海亿为科技有限公司 | Method, device and equipment for checking medical data in AR (augmented reality) inspection based on cloud edge |
| CN113192622B (en) * | 2021-05-08 | 2023-06-23 | 上海亿为科技有限公司 | Method, device and equipment for checking medical data through AR (augmented reality) inspection based on cloud edge |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111460250B (en) | 2022-07-08 |
| WO2021174812A1 (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111460250B (en) | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus | |
| CN110781276B (en) | Text extraction method, device, equipment and storage medium | |
| CN110555372A (en) | Data entry method, device, equipment and storage medium | |
| CN109872162B (en) | Wind control classification and identification method and system for processing user complaint information | |
| CN112270196B (en) | Entity relationship identification method and device and electronic equipment | |
| CN109189767B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN109783631B (en) | Community question-answer data verification method and device, computer equipment and storage medium | |
| CN112347244A (en) | Method for detecting website involved in yellow and gambling based on mixed feature analysis | |
| CN108550054B (en) | Content quality evaluation method, device, equipment and medium | |
| CN113450147B (en) | Product matching method, device, equipment and storage medium based on decision tree | |
| CN113486664A (en) | Text data visualization analysis method, device, equipment and storage medium | |
| CN114647713A (en) | Knowledge graph question-answering method, device and storage medium based on virtual confrontation | |
| CN112579781B (en) | Text classification method, device, electronic equipment and medium | |
| CN112989050A (en) | Table classification method, device, equipment and storage medium | |
| US20200279148A1 (en) | Material structure analysis method and material structure analyzer | |
| CN113836297B (en) | Training method and device for text emotion analysis model | |
| CN109657710B (en) | Data screening method and device, server and storage medium | |
| CN114842982B (en) | Knowledge expression method, device and system for medical information system | |
| CN112732908B (en) | Test question novelty evaluation method and device, electronic equipment and storage medium | |
| CN115392787A (en) | Enterprise risk assessment method, device, equipment, storage medium and program product | |
| CN111460206B (en) | Image processing method, apparatus, electronic device, and computer-readable storage medium | |
| CN114528908A (en) | Network request data classification model training method, classification method and storage medium | |
| CN110083807B (en) | Contract modification influence automatic prediction method, device, medium and electronic equipment | |
| CN113033170A (en) | Table standardization processing method, device, equipment and storage medium | |
| CN116341531B (en) | Knowledge-driven character information extraction and inspection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |