CN113761850A - Form filling method and device - Google Patents
Form filling method and device Download PDFInfo
- Publication number
- CN113761850A CN113761850A CN202011274813.1A CN202011274813A CN113761850A CN 113761850 A CN113761850 A CN 113761850A CN 202011274813 A CN202011274813 A CN 202011274813A CN 113761850 A CN113761850 A CN 113761850A
- Authority
- CN
- China
- Prior art keywords
- filled
- form item
- item
- current value
- determining
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000012549 training Methods 0.000 claims description 37
- 238000006243 chemical reaction Methods 0.000 claims description 24
- 238000004590 computer program Methods 0.000 claims description 10
- 238000010586 diagram Methods 0.000 description 21
- 238000012545 processing Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 13
- 230000006870 function Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 8
- 230000009466 transformation Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 238000013145 classification model Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000012821 model calculation Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005429 filling process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/174—Form filling; Merging
-
- 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)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a form filling method and device, and relates to the technical field of computers. One embodiment of the method comprises: determining form items to be filled of a current form; acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated; determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled; and filling the current value of the to-be-filled form item into an input box of the to-be-filled form item. According to the embodiment, manual form filling operation of the user can be reduced, and the form filling time of the user is saved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a form filling method and device.
Background
With the development of technology, users often need to fill in various forms on the network. Such as shopping or submitting questions online, the user is typically required to fill in many forms. In another example, in a company, cross-department communication or transaction processing has become online, and a user also needs to fill in various forms such as business examination and approval documents and cross-department work orders when processing company business. In the process of filling in the form, the user needs to continuously perform manual input operation within a long time, the operation is complicated, and the precious time of the user is wasted.
Disclosure of Invention
In view of this, embodiments of the present invention provide a form filling method and apparatus, which can reduce operations of manually filling a form by a user and save time for filling the form by the user.
In a first aspect, an embodiment of the present invention provides a form filling method, including:
determining form items to be filled of a current form;
acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated;
determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled;
and filling the current value of the to-be-filled form item into an input box of the to-be-filled form item.
Alternatively,
the determining the form item to be filled of the current form includes:
and determining whether a filling switch corresponding to the form item of the current form is turned on, and if so, determining that the form item is the form item to be filled.
Alternatively,
further comprising:
acquiring historical values of a plurality of form items; wherein the form items are derived from history forms;
integrating the historical values of the plurality of form items into a characteristic width table;
carrying out format conversion on the data in the feature width table according to a pre-stored feature conversion rule;
training the prediction model according to the pre-specified form item to be filled and the converted feature width table to obtain a training result; wherein the training result comprises: and (5) training a well-trained prediction model.
Alternatively,
the training result further comprises: the influence value of a plurality of other form items on the form item to be filled; wherein the influence value is used for representing the influence degree of the other form items on the form item to be filled;
further comprising: and determining a target form item associated with the form item to be filled in the plurality of other form items according to the influence value.
Alternatively,
determining the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item, wherein the determining comprises the following steps:
carrying out format conversion on the current value of the target form item according to a pre-stored characteristic conversion rule;
determining a link address of the prediction model in an online service;
and taking the converted current value of the target form item as an input parameter, and requesting a prediction service corresponding to the link address to obtain the current value of the form item to be filled.
Alternatively,
the input box of the form item to be filled is a drop-down box;
determining the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item, wherein the determining comprises the following steps:
generating a plurality of predicted values and probabilities of the to-be-filled form items according to the current values of the target form items and the prediction models of the to-be-filled form items;
and determining the predicted value corresponding to the maximum probability as the current value of the form to be filled.
Alternatively,
further comprising:
for each of the predicted values: determining the position of the predicted value in a drop-down list of the drop-down box according to the probability of the predicted value;
and displaying the drop-down box according to the plurality of predicted values and the positions of the predicted values in the drop-down list.
In a second aspect, an embodiment of the present invention provides a form filling apparatus, including:
the form item determining module is configured to determine a form item to be filled of the current form;
a current value acquisition module configured to acquire a current value of the target form item; wherein the target form item is associated with the form item to be populated;
the prediction module is configured to determine the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item;
and the input box filling module is configured to fill the current value of the to-be-filled form item into the input box of the to-be-filled form item.
In a third aspect, an embodiment of the present invention provides an electronic device, including:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any of the embodiments described above.
In a fourth aspect, an embodiment of the present invention provides a computer-readable medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method of any one of the above embodiments.
One embodiment of the above invention has the following advantages or benefits: and determining the current value of the form item to be filled according to the current value of the target form item and the prediction model of the form item to be filled, and filling the current value of the form item to be filled into the input box of the form item to be filled. And automatically filling the input box of the form item to be filled according to the predicted current value of the form item to be filled, so that the manual form filling operation of a user can be reduced, and the form filling time of the user is saved.
Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a schematic diagram illustrating a flow of a form filling method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a method for determining a target form item according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the flow of another form filling method provided by an embodiment of the invention;
FIG. 4 is a block diagram of a framework of a smart form filling system according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a flow of another form filling method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of an offline training process provided by an embodiment of the present invention;
FIG. 7 is a schematic diagram of a feature engineering system of the present invention;
FIG. 8 is a block diagram of a framework for a further system for filling smart forms, according to an embodiment of the invention;
FIG. 9 is a schematic diagram of a form filling apparatus according to an embodiment of the present invention;
FIG. 10 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;
fig. 11 is a schematic structural diagram of a computer system suitable for implementing a terminal device or a server according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
An embodiment of the present invention provides a form filling method, as shown in fig. 1, including:
step 101: and determining the form item to be filled of the current form.
The current form is the form that the user needs to fill in. Several to-be-filled form items may be included in the current form. The input box to be populated with form items may include: text boxes, check boxes, radio boxes, drop down boxes, and the like. The form item to be filled in the current form can be predicted and filled by the method of the embodiment of the invention.
Step 102: acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated.
The target form entry is the form entry associated with the to-be-populated. And predicting the value of the to-be-filled form item according to the value of the target form item. The target form item can be one or more. The target form item may originate from the same form as the form item to be populated or may originate from a different form than the form item to be populated.
For example, the current form is a user's shopping wish list, which is used to store items that the user likes and wants to purchase. The wish text box is a to-be-filled form item in the shopping wish list, and a user can fill a plurality of favorite items into the wish text boxes. The user identification can be used as a key word, a plurality of form items corresponding to the behavior information of the user are associated from a browsing article information table of the user, an order information table of the user and a collection article information table of the user, and the form items corresponding to the behavior information are used as target form items.
Step 103: and determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled.
The prediction model is a model in which the current value of the target form item is used as input and the predicted value of the form item to be filled is used as output. If a plurality of to-be-filled form items exist in the current form, a plurality of prediction models can be set for each to-be-filled form item, and a multi-target combined model can also be set for the to-be-filled form items. The predictive model may include: regression prediction models, temporal prediction models, and the like.
Step 104: and filling the current value of the to-be-filled form item into the input box of the to-be-filled form item.
The method of the embodiment of the invention determines the current value of the form item to be filled according to the current value of the target form item and the prediction model of the form item to be filled, and fills the current value of the form item to be filled into the input box of the form item to be filled. The method and the device for filling the form items can automatically fill the input boxes of the form items to be filled, and a user does not need to manually fill the form items to be filled. Therefore, the method of the embodiment of the invention can solve the problems of complicated operation and time waste of the user to fill the form.
In one embodiment of the present invention, determining the to-be-filled form item of the current form comprises:
and determining whether a filling switch corresponding to the form item of the current form is opened, and if so, determining the form item as a form item to be filled.
When the filling switch is turned on, the system calls the method of the embodiment of the invention to automatically fill the form item to be filled of the current form. When the filling switch is closed, the system does not call the method of the embodiment of the invention to fill the form item to be filled of the current form, but the user fills the form item by himself.
Acquiring an actual value filled by a user aiming at the form item to be filled; and if the deviation between the actual value of the form item to be filled and the current value of the form item to be filled exceeds a first deviation threshold, setting a filling switch corresponding to the form item of the current form to be in a closed state.
The first deviation threshold may be set according to specific needs. If the deviation between the actual value filled by the user and the current value predicted by the system is large, the prediction result of the current prediction model is possibly not accurate enough, and the prediction model needs to be iterated further to enhance the accuracy of the prediction result of the system. At this time, the filling switch of the form item to be filled can be set to be in a closed state, so as to avoid the improper influence on the subsequent form filling process of the user due to the fact that the prediction result is not accurate enough.
Acquiring an actual value filled by a user aiming at the form item to be filled; if the deviation between the actual value of the form item to be filled and the current value of the form item to be filled exceeds a second deviation threshold value, saving form data; wherein the form data includes: the current value of the target form item and the actual value of the form item to be filled; the form data is used to train the predictive model.
The second deviation threshold may be set according to specific needs. If the deviation between the actual value filled by the user and the current value predicted by the system is large, the prediction result of the current prediction model is possibly not accurate enough, and the prediction model needs to be further iterated. And the form data with larger deviation is recorded and stored for subsequent prediction model training, so that the prediction result of the trained prediction model is more accurate.
In one embodiment of the invention, the method further comprises:
acquiring historical values of a plurality of form items; wherein, the plurality of form items are derived from a plurality of historical forms;
integrating the historical values of a plurality of form items into a characteristic width table;
carrying out format conversion on the data in the characteristic width table according to a pre-stored characteristic conversion rule;
training the prediction model according to the pre-specified form item to be filled and the converted feature width table to obtain a training result; wherein, the training result includes: and (5) training a well-trained prediction model.
The embodiment of the invention provides a training method of a prediction model. First, the correlation tables are integrated into a feature width table. Because the characteristic width table comprises different data from a plurality of tables, the efficiency of iterative computation in the process of predicting the model training can be improved by utilizing the characteristic width table to train the model.
Second, each field in the feature width table is feature transformed so that the transformed field can be identified by the predictive model. For example, if the predictive model can only recognize input parameters in numerical form, the fields in the feature width table can be converted as follows:
for discrete fields: the original values are encoded as 1, 2, 3, etc. in turn, and the original values should correspond to the encoded values one-to-one. The encoded value may serve as a unique identification of the original value. Examples are: possible values for gender characteristics: for male and female, the male may be coded as 1 and the female may be coded as 2.
For text fields: the word segmentation is carried out firstly, then the word vector corresponding to each word is obtained respectively, and finally the vector average value of all the segmented words is calculated and used as the characteristic value of the text field. Examples are: "i want to return goods", after this word segmentation { "i", "want", "return goods" }, "i" corresponds to a word vector [0.5, 0.3. ], "want" corresponds to [0.2, 0.3. ], "return goods" corresponds to [0.2, 0.9. ], and the vector of all the segmentation words is calculated to be averaged, and the characteristic value of the text field is [0.3, 0.5. ].
And finally, training the prediction model according to the converted feature width table to obtain the trained prediction model.
In addition, the feature transformation rules can be stored in the feature schema, instead of being hard-coded and written in codes, the feature transformation rules can be directly obtained through the feature schema in the later iteration of the model and the processing of the original data in the online service process, and the feature schema can play a role in decoupling the online service engineering and the model service.
In an embodiment of the present invention, determining the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item includes:
carrying out format conversion on the current value of the target form item according to a pre-stored characteristic conversion rule;
determining a link address of a prediction model in an online service;
and taking the converted current value of the target form item as an input parameter, and requesting a prediction service corresponding to the link address to obtain the current value of the form item to be filled.
And taking the current value of the converted target form item as an input parameter, and requesting the prediction service based on the prediction model by using a link address in the online service of the prediction model. The embodiment of the invention provides a method for filling a form by using a prediction model conveniently and effectively.
In addition, the link addresses of the feature conversion rules and the prediction models in online services can be stored in the model outline without being written in codes, so that the effect of decoupling each module can be achieved, and the construction and later maintenance of the whole system are facilitated.
An embodiment of the present invention provides a method for determining a target form item, as shown in fig. 2, including:
step 201: acquiring historical values of a plurality of form items; wherein the plurality of form items are derived from the plurality of history forms.
Step 202: and integrating the historical values of a plurality of table entries into a characteristic wide table.
Step 203: and carrying out format conversion on the data in the characteristic width table according to a pre-stored characteristic conversion rule.
Step 204: training the prediction model according to the pre-specified form item to be filled and the converted feature width table to obtain a training result; wherein, the training result includes: the influence values of the table items to be filled by a plurality of other table items; wherein the influence value is used for representing the influence degree of other form items on the form item to be filled.
Step 205: the target form entry associated with the form entry to be populated is determined among several other form entries according to the impact value.
The target form item associated with the form item to be populated may be determined in a number of ways depending on the impact value. An influence threshold may be set, and all the form items corresponding to influence values higher than the influence threshold may be determined as target form items. The number of target form items, such as 5, 8, etc., can also be set. And sequencing all other form items according to the influence values, and selecting other form items with the number of the front target form items from the sequenced form items as target form items.
According to the embodiment of the invention, the influence value of other form items on the form item to be filled is obtained according to the training result, and then other form items with larger influence on the form item to be filled are selected as the target form item according to the influence value. In the subsequent prediction process, only the values of the target list items are input into the prediction model, and the values of all the list items in the characteristic width table do not need to be obtained. Therefore, the method of the embodiment of the invention can not only accurately predict the list items to be filled, but also accelerate the prediction process of the prediction model.
FIG. 3 is a flow diagram of another form filling method provided by another embodiment of the invention. As shown in fig. 3, the method includes:
step 301: determining form items to be filled of a current form; wherein, the input box of the form item to be filled is a drop-down box.
Step 302: acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated.
Step 303: and generating a plurality of predicted values and probabilities of the to-be-filled list items according to the current values of the target list items and the prediction models of the to-be-filled list items.
Step 304: and determining the predicted value corresponding to the maximum probability as the current value of the form to be filled.
Step 305: and filling the current value of the to-be-filled form item into the input box of the to-be-filled form item.
According to the method provided by the embodiment of the invention, when the input frame of the form item to be filled is the drop-down frame, the predicted value with the maximum probability is taken as the current value of the form to be filled, and is filled into the following frame to be filled, so that the predicted value filled into the drop-down frame has higher accuracy.
In one embodiment of the invention, the method further comprises:
for each predictor: determining the position of the predicted value in a drop-down list of a drop-down frame according to the probability of the predicted value;
and displaying the drop-down box according to the plurality of predicted values and the positions of the predicted values in the drop-down list.
The greater the probability of a predicted value, the greater the probability that the user will select the predicted value as the input value for the drop-down box. Therefore, predicted values having a greater probability may be arranged at a position earlier in the drop-down box for convenient selection by the user.
The scheme of the embodiment of the invention aims to provide a learning mechanism, which uses a classification algorithm in machine learning to learn the filling condition of a past form and fill the form item to be filled. The following explanation takes the form item to be filled as an example of the following box.
In the method of the embodiment of the present invention, the refinement history may affect the target form item filled in by the form item to be filled, and the target form item may include: form type, form writer, form entry, etc. Among them, the form types may include: question consultation, approval documents, question documents, and the like; the form entry may include: a telephone customer service workbench, an after-sale workbench, a spare part warehouse workbench and the like. If the form to be filled is related to the order, then the order-related form item, the orderer-related form item, etc. may also be taken. And performing characterization processing on the related list items, and predicting the optimal default value of a certain drop-down box. The scheme of the embodiment of the invention can improve the intelligent filling proportion, reduce the cost, improve the form filling efficiency, reduce the manual configuration and provide the continuous automatic learning capability.
FIG. 4 is a block diagram of a framework of a system for filling smart forms according to an embodiment of the present invention. As shown in FIG. 4, the smart form filling system includes: the system comprises a model calculation layer, a data service layer and a system application layer.
Model calculation layer: focusing on off-line learning, and counting characteristic values according to analysis; collecting off-line data, cleaning the data and establishing a characteristic project; continuously training the models by using the classification models; and outputting a final model result.
A data service layer: focusing on real-time application, and acquiring real-time data of various characteristic values required by the model; processing real-time data and storing and managing; and calling an algorithm API (Application Programming Interface) to receive the model result.
A system application layer: a call switch is configured in the system. When the call switch is turned on, the intelligent form filling system is applied to fill the form; calling a model result in an interface form; when the monitoring finds that the prediction deviation is large, service degradation processing can be carried out at any time, and stable and available application of the system is realized. The service downgrade process is used to stop the method of applying the form fill system, enabling the traditional list scheme.
The system abstracts three layers of single responsibility, comprising: offline learning, real-time application and business rule control, and a whole set of system for filling the unmanned form is realized. Fig. 5 is a schematic diagram of a flow of another form filling method according to an embodiment of the present invention, and as shown in fig. 5, a specific technical solution includes a main flow and a big data model calculation flow. The main process comprises the following steps:
s1: and (3) creating a form application: and acquiring the form to be filled by the user and the form item to be filled.
S2: and determining whether a filling switch corresponding to the form item of the current form is opened, and if so, determining the form item as a form item to be filled.
And according to a preset switch, determining whether each form item to be filled needs to call the form filling method provided by the embodiment of the invention.
S3: and determining the form item to be filled of the current form.
S4: acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated.
S5: and determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled.
This step determines the current value of the form item to be filled by model prediction. The step of model prediction is the core of the embodiments of the present invention. The model prediction comprises two parts of off-line training and on-line service.
Fig. 6 is a schematic diagram of an offline training process according to an embodiment of the present invention. As shown in fig. 6, the offline training stream includes: the model calculation comprises two parts of off-line training and on-line service.
S51: and (3) offline data processing: on the market of a large data set, relevant data tables are collected, offline analysis is carried out, required features are integrated into a feature wide table and are provided for the subsequent links for use, and the offline data processing task is updated regularly.
S52: characteristic engineering: cleaning and processing related characteristics, such as form type, form filling person and form entry, if the related characteristics are related to the order, then taking order related characteristics, such as form, normalization and missing value, and the like; in addition, the feature transformation rules are written to the feature schema cache.
FIG. 7 is a schematic diagram of a feature engineering system of the present invention. As shown in fig. 7, discrete features and text features are separately transformed to generate a digital code that the prediction model can recognize.
S53: model training: the algorithm is to complete the classification task of the default value of the drop-down box, and train the multi-classification model according to the drop-down option. And if a plurality of drop-down frames exist, respectively generating a plurality of multi-classification models or multi-target combined models to generate model files.
S54: model iteration: the model iteration task can be executed regularly or based on the command of the user or the system. And in the model iteration process, the model effect is improved by continuously iterating and updating the model.
Online service deployment involves the following 4 steps:
s55: model deployment: and deploying the model file generated by offline training into online service through engineering packaging, and writing information such as a model service url (Uniform Resource Locator), a model version and the like into a model outline cache.
S56: acquiring real-time data: and calling a multi-party interface or data storage to obtain all data required by the model.
S57: feature conversion: and converting the original characteristics of the real-time request into a format required by the model through a conversion rule specified by the characteristic outline.
S58: model prediction: and analyzing the configuration and parameters generated by the model training result, requesting the model service url specified by the feature outline according to the prediction function and the real-time conversion feature, and requesting the prediction service to obtain a predicted value.
S6: and acquiring input information confirmed by a user. The generic model does not provide a completely correct fill and the user needs to confirm that the underlying fill is correct and can manually re-fill if it is wrong.
S7: and filling the current value of the to-be-filled form item into the input box of the to-be-filled form item. A complete filling form is created and submitted.
S8: model iteration: and for data with errors, the system records the data, and automatically feeds the data back to the model training link at regular time, so that the model is iterated and corrected, and the effect of the model is improved.
FIG. 8 is a block diagram of a framework of another system for filling smart forms according to an embodiment of the present invention. As shown in fig. 8, the system stores information such as url of the model service applied in the process of performing the feature engineering, the model version number, and the feature transformation rule dictionary into the feature schema. And the online service performs characteristic conversion on the original characteristics according to characteristic conversion rules in the characteristic outline, acquires the url of the model service through the characteristic outline, calls the prediction service and fills the field to be filled.
The prediction function is a prediction method corresponding to the prediction model. And taking the converted features as the input parameters of the prediction function, and calling the prediction function to predict the category. The xdeepFM model can be adopted in the embodiment of the invention.
The feature schema is a set of model specific metadata, which can be generated by the feature engineering and model deployment links and stored in the cache server. The feature transformation module transforms the original features according to the feature outline, instead of writing the transformation rules in the codes in a hard coding mode, and directly updates the metadata to the feature outline cache after the model is iterated in the later period and the features are added, deleted and changed, thereby avoiding the modification and the restart of the feature transformation module. The schema functions to decouple online service engineering from model services.
As shown in fig. 9, an embodiment of the present invention provides a form filling apparatus, including:
a form item determining module 901 configured to determine a form item to be filled in a current form;
a current value obtaining module 902 configured to obtain a current value of the target form item; wherein the target form item is associated with a form item to be populated;
a prediction module 903, configured to determine a current value of the form item to be filled according to the current value of the target form item and a trained prediction model of the form item to be filled;
an input box populating module 904 configured to populate an input box of a to-be-populated form item with a current value of the to-be-populated form item.
In an embodiment of the present invention, the form item determining module 901 is specifically configured to:
and determining whether a filling switch corresponding to the form item of the current form is opened, and if so, determining the form item as a form item to be filled.
In one embodiment of the invention, the prediction module 903 is configured to obtain historical values of a number of form items; wherein, the plurality of form items are derived from a plurality of historical forms;
integrating the historical values of a plurality of form items into a characteristic width table;
carrying out format conversion on the data in the characteristic width table according to a pre-stored characteristic conversion rule;
training the prediction model according to the pre-specified form item to be filled and the converted feature width table to obtain a training result; wherein, the training result includes: and (5) training a well-trained prediction model.
In an embodiment of the present invention, the training result further includes: the influence values of the table items to be filled by a plurality of other table items; wherein, the influence value is used for representing the influence degree of other form items on the form items to be filled;
further comprising: a prediction module 903 configured to determine a target form item associated with the form item to be populated among a number of other form items according to the impact value.
In one embodiment of the present invention, the prediction module 903 is specifically configured to:
carrying out format conversion on the current value of the target form item according to a pre-stored characteristic conversion rule;
determining a link address of a prediction model in an online service;
and taking the converted current value of the target form item as an input parameter, and requesting a prediction service corresponding to the link address to obtain the current value of the form item to be filled.
In one embodiment of the invention, the input box of the form item to be filled is a drop-down box;
the prediction module 903 is specifically configured to: generating a plurality of predicted values and probabilities of the form items to be filled according to the current value of the target form item and the prediction model of the form items to be filled;
and determining the predicted value corresponding to the maximum probability as the current value of the form to be filled.
In one embodiment of the invention, the prediction module 903 is configured to, for each predictor: determining the position of the predicted value in a drop-down list of a drop-down frame according to the probability of the predicted value;
and displaying the drop-down box according to the plurality of predicted values and the positions of the predicted values in the drop-down list.
An embodiment of the present invention provides an electronic device, including:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method of any of the embodiments described above.
Fig. 10 shows an exemplary system architecture 1000 to which the abnormal behavior recognition method or the abnormal behavior recognition apparatus of the embodiment of the present invention can be applied.
As shown in fig. 10, the system architecture 1000 may include terminal devices 1001, 1002, 1003, a network 1004, and a server 1005. The network 1004 is used to provide a medium for communication links between the terminal devices 1001, 1002, 1003 and the server 1005. Network 1004 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
A user may use the terminal devices 1001, 1002, 1003 to interact with a server 1005 via a network 1004 to receive or transmit messages or the like. The terminal devices 1001, 1002, 1003 may have installed thereon various communication client applications, such as shopping applications, web browser applications, search applications, instant messaging tools, mailbox clients, social platform software, office applications, etc. (by way of example only).
The terminal devices 1001, 1002, 1003 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 1005 may be a server that provides various services, such as a back office management server (just an example) that supports a shopping website or a company department internal management system browsed by a user using the terminal devices 1001, 1002, 1003. The background management server can acquire forms which need to be filled by a user; determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled; and filling the current value of the form item to be filled into the input box of the form item to be filled, and feeding back the filled form to the terminal equipment.
It should be noted that the method for processing the notification trigger message provided by the embodiment of the present invention is generally executed by the server 1005, and accordingly, the form filling apparatus for the article is generally disposed in the server 1005.
It should be understood that the number of terminal devices, networks, and servers in fig. 10 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring now to FIG. 11, shown is a block diagram of a computer system 1100 suitable for use with a terminal device implementing an embodiment of the present invention. The terminal device shown in fig. 11 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. 11, the computer system 1100 includes a Central Processing Unit (CPU)1101, which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)1102 or a program loaded from a storage section 1108 into a Random Access Memory (RAM) 1103. In the RAM 1103, various programs and data necessary for the operation of the system 1100 are also stored. The CPU 1101, ROM 1102, and RAM 1103 are connected to each other by a bus 1104. An input/output (I/O) interface 1105 is also connected to bus 1104.
The following components are connected to the I/O interface 1105: an input portion 1109 including a keyboard, a mouse, and the like; an output portion 1107 including a signal output unit such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage section 1108 including a hard disk and the like; and a communication section 1109 including a network interface card such as a LAN card, a modem, or the like. The communication section 1109 performs communication processing via a network such as the internet. A driver 1110 is also connected to the I/O interface 1105 as necessary. A removable medium 1111 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1110 as necessary, so that a computer program read out therefrom is mounted into the storage section 1108 as necessary.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 1109 and/or installed from the removable medium 1111. The above-described functions defined in the system of the present invention are executed when the computer program is executed by a Central Processing Unit (CPU) 1101.
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer 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 of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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. In the present invention, a computer 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. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer 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 computer readable signal medium may also be any computer readable medium that is not a computer 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 computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a sending module, an obtaining module, a determining module, and a first processing module. The names of these modules do not form a limitation on the modules themselves in some cases, and for example, the sending module may also be described as a "module sending a picture acquisition request to a connected server".
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise:
determining form items to be filled of a current form;
acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated;
determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled;
and filling the current value of the to-be-filled form item into an input box of the to-be-filled form item.
According to the technical scheme of the embodiment of the invention, the current value of the to-be-filled form item is determined according to the current value of the target form item and the prediction model of the to-be-filled form item, and the current value of the to-be-filled form item is filled into the input box of the to-be-filled form item. According to the method and the device for filling the form items, the input box of the form items to be filled is automatically filled according to the predicted current value of the form items to be filled, so that the input operation of filling the form by a user can be reduced, and the form filling time of the user is saved.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A form filling method, comprising:
determining form items to be filled of a current form;
acquiring a current value of a target form item; wherein the target form item is associated with the form item to be populated;
determining the current value of the form item to be filled according to the current value of the target form item and the trained prediction model of the form item to be filled;
and filling the current value of the to-be-filled form item into an input box of the to-be-filled form item.
2. The method of claim 1,
the determining the form item to be filled of the current form includes:
and determining whether a filling switch corresponding to the form item of the current form is turned on, and if so, determining that the form item is the form item to be filled.
3. The method of claim 1, further comprising:
acquiring historical values of a plurality of form items; wherein the form items are derived from history forms;
integrating the historical values of the plurality of form items into a characteristic width table;
carrying out format conversion on the data in the feature width table according to a pre-stored feature conversion rule;
training the prediction model according to the pre-specified form item to be filled and the converted feature width table to obtain a training result; wherein the training result comprises: and (5) training a well-trained prediction model.
4. The method of claim 3,
the training result further comprises: the influence value of a plurality of other form items on the form item to be filled; wherein the influence value is used for representing the influence degree of the other form items on the form item to be filled;
further comprising: and determining a target form item associated with the form item to be filled in the plurality of other form items according to the influence value.
5. The method of claim 1,
determining the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item, wherein the determining comprises the following steps:
carrying out format conversion on the current value of the target form item according to a pre-stored characteristic conversion rule;
determining a link address of the prediction model in an online service;
and taking the converted current value of the target form item as an input parameter, and requesting a prediction service corresponding to the link address to obtain the current value of the form item to be filled.
6. The method of claim 1,
the input box of the form item to be filled is a drop-down box;
determining the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item, wherein the determining comprises the following steps:
generating a plurality of predicted values and probabilities of the to-be-filled form items according to the current values of the target form items and the prediction models of the to-be-filled form items;
and determining the predicted value corresponding to the maximum probability as the current value of the form to be filled.
7. The method of claim 6, further comprising:
for each of the predicted values: determining the position of the predicted value in a drop-down list of the drop-down box according to the probability of the predicted value;
and displaying the drop-down box according to the plurality of predicted values and the positions of the predicted values in the drop-down list.
8. A form filling apparatus, comprising:
the form item determining module is configured to determine a form item to be filled of the current form;
a current value acquisition module configured to acquire a current value of the target form item; wherein the target form item is associated with the form item to be populated;
the prediction module is configured to determine the current value of the to-be-filled form item according to the current value of the target form item and the trained prediction model of the to-be-filled form item;
and the input box filling module is configured to fill the current value of the to-be-filled form item into the input box of the to-be-filled form item.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.
10. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011274813.1A CN113761850A (en) | 2020-11-13 | 2020-11-13 | Form filling method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011274813.1A CN113761850A (en) | 2020-11-13 | 2020-11-13 | Form filling method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113761850A true CN113761850A (en) | 2021-12-07 |
Family
ID=78786013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011274813.1A Pending CN113761850A (en) | 2020-11-13 | 2020-11-13 | Form filling method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113761850A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050257134A1 (en) * | 2004-05-12 | 2005-11-17 | Microsoft Corporation | Intelligent autofill |
| WO2012174680A1 (en) * | 2011-06-24 | 2012-12-27 | Hewlett-Packard Development Company, L.P. | Identity based automated form filling |
| CN104834996A (en) * | 2015-04-22 | 2015-08-12 | 中国建设银行股份有限公司 | Form-filling method and device |
| CN109308350A (en) * | 2018-09-26 | 2019-02-05 | 平安普惠企业管理有限公司 | Format Auto Filling Forms method, apparatus, computer equipment and storage medium |
| CN109814779A (en) * | 2019-01-04 | 2019-05-28 | 平安科技（深圳）有限公司 | Menu display method, device, computer equipment and storage medium |
| CN111046663A (en) * | 2019-11-26 | 2020-04-21 | 华南理工大学 | Intelligent correction method for Chinese form |
| CN111128330A (en) * | 2019-12-30 | 2020-05-08 | 天津开心生活科技有限公司 | Automatic entry method and device for electronic case report table and related equipment |
| CN111461397A (en) * | 2020-02-26 | 2020-07-28 | 山东浪潮通软信息科技有限公司 | Budget prediction method, equipment and medium based on improved support vector regression |
-
2020
- 2020-11-13 CN CN202011274813.1A patent/CN113761850A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050257134A1 (en) * | 2004-05-12 | 2005-11-17 | Microsoft Corporation | Intelligent autofill |
| WO2012174680A1 (en) * | 2011-06-24 | 2012-12-27 | Hewlett-Packard Development Company, L.P. | Identity based automated form filling |
| CN104834996A (en) * | 2015-04-22 | 2015-08-12 | 中国建设银行股份有限公司 | Form-filling method and device |
| CN109308350A (en) * | 2018-09-26 | 2019-02-05 | 平安普惠企业管理有限公司 | Format Auto Filling Forms method, apparatus, computer equipment and storage medium |
| CN109814779A (en) * | 2019-01-04 | 2019-05-28 | 平安科技（深圳）有限公司 | Menu display method, device, computer equipment and storage medium |
| CN111046663A (en) * | 2019-11-26 | 2020-04-21 | 华南理工大学 | Intelligent correction method for Chinese form |
| CN111128330A (en) * | 2019-12-30 | 2020-05-08 | 天津开心生活科技有限公司 | Automatic entry method and device for electronic case report table and related equipment |
| CN111461397A (en) * | 2020-02-26 | 2020-07-28 | 山东浪潮通软信息科技有限公司 | Budget prediction method, equipment and medium based on improved support vector regression |
Non-Patent Citations (1)
| Title |
|---|
| 张洋;高岭;高全力;杨建锋;: "一种基于协同过滤的表单推荐录入模型", 计算机工程, no. 01, 15 January 2016 (2016-01-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460513B (en) | Method and apparatus for generating click rate prediction model | |
| CN110390408B (en) | Transaction object prediction method and device | |
| CN110020162B (en) | User identification method and device | |
| CN107644036B (en) | Method, device and system for pushing data object | |
| CN109993627B (en) | Recommendation method, recommendation model training device and storage medium | |
| CN113763093A (en) | User portrait-based item recommendation method and device | |
| US11741094B2 (en) | Method and system for identifying core product terms | |
| CN110109901B (en) | Method and device for screening target object | |
| CN109961328B (en) | Method and device for determining order cool-still period | |
| CN113743971A (en) | Data processing method and device | |
| CN112825182A (en) | Method and device for determining recommended commodities | |
| CN112257868A (en) | Method and device for constructing and training integrated prediction model for predicting passenger flow | |
| CN115238710A (en) | Intelligent document generation and management method and device | |
| CN112231299B (en) | Method and device for dynamically adjusting feature library | |
| WO2024051707A1 (en) | Recommendation model training method and apparatus, and resource recommendation method and apparatus | |
| CN117114618A (en) | To-be-handled item follow-up method, system, computer equipment and storage medium | |
| CN113761850A (en) | Form filling method and device | |
| CN115860872A (en) | Target object determination method and device, electronic equipment and storage medium | |
| US20220100768A1 (en) | Systems and methods for selectively prefetching data | |
| CN114677174A (en) | Method and device for calculating sales volume of unladen articles | |
| CN113360693A (en) | Method and device for determining image label, electronic equipment and storage medium | |
| CN110334177B (en) | Semantic similarity model training and semantic similarity recognition methods and devices and electronic equipment | |
| CN113743973A (en) | Method and device for analyzing market hotspot trend | |
| CN113762992A (en) | Method and device for processing data | |
| CN112528103A (en) | Method and device for recommending objects |
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 |