CN111782960A - File pushing method and device and electronic equipment - Google Patents
File pushing method and device and electronic equipment Download PDFInfo
- Publication number
- CN111782960A CN111782960A CN202010733172.5A CN202010733172A CN111782960A CN 111782960 A CN111782960 A CN 111782960A CN 202010733172 A CN202010733172 A CN 202010733172A CN 111782960 A CN111782960 A CN 111782960A
- Authority
- CN
- China
- Prior art keywords
- pushing
- user
- result
- client
- file
- 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 105
- 238000012549 training Methods 0.000 claims description 28
- 238000012545 processing Methods 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 21
- 238000002372 labelling Methods 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 17
- 230000001133 acceleration Effects 0.000 claims description 14
- 238000005286 illumination Methods 0.000 claims description 14
- 230000035945 sensitivity Effects 0.000 claims description 14
- 230000006399 behavior Effects 0.000 claims description 10
- 238000012795 verification Methods 0.000 description 20
- 230000001680 brushing effect Effects 0.000 description 19
- 238000010586 diagram Methods 0.000 description 16
- 238000007477 logistic regression Methods 0.000 description 11
- 230000006872 improvement Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 238000007493 shaping process Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000003062 neural network model Methods 0.000 description 6
- 230000001815 facial effect Effects 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
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
- G06F16/953—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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Security & Cryptography (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Collating Specific Patterns (AREA)
Abstract
The specification discloses a method and a device for pushing a document and electronic equipment, wherein the method for pushing the document comprises the following steps: receiving a first pushing result sent by a client; acquiring server side characteristics, and inputting the first pushing result and the server side characteristics into a second file pushing model local to the server side to obtain a second pushing result; and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a document pushing method and apparatus, and an electronic device.
Background
The biological core body mode is a safe and convenient core body mode, can realize rapidly carrying out identity recognition, promotes user's core body experience.
At present, the server analyzes historical data of a user to determine a biometric way to push to the user. However, the above recommendation method is poor in real-time performance, and the recommended biometric authentication method may not be suitable for the user.
Therefore, a new method for pushing the bio-core is required.
Disclosure of Invention
In view of this, an embodiment of the present disclosure provides a document pushing method, which is used to solve the problem that in the prior art, the pushing real-time performance of the biometric authentication method is poor, and the recommended biometric authentication method may not be suitable for the user.
The embodiment of the specification adopts the following technical scheme:
the embodiment of the present specification provides a document pushing method, which is applied to a server and includes:
receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
obtaining server-side characteristics, and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
An embodiment of the present specification further provides another document pushing method, which is applied to a client, and includes:
responding to the target operation and acquiring the characteristics of the client;
inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating a biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
sending the first pushing result to a server, wherein the first pushing result is used for the server to input the first pushing result and server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
receiving a second pushing result returned by the server;
and judging whether to push the documents to the user according to the second pushing result, and if the documents are judged to be required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
An embodiment of the present specification further provides a document pushing apparatus, including:
the first receiving module is used for receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
the first processing module is used for acquiring server-side characteristics and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and the first sending module is used for sending the second pushing result to the client, the second pushing result is used for judging whether to push the file to the user according to the second pushing result by the client, and if the file needs to be pushed to the user, the second pushing result is also used for selecting a target file from the candidate files to push the target file to the user according to the second pushing result by the client.
An embodiment of the present specification further provides a document pushing apparatus, including:
the acquisition module is used for responding to the target operation and acquiring the characteristics of the client;
the second processing module is used for inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating the biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
the second sending module is used for sending the first pushing result to a server, the first pushing result is used for the server to input the first pushing result and the server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
the second receiving module is used for receiving a second pushing result returned by the server;
and the second processing module judges whether to push the documents to the user according to the second pushing result, and if the documents are required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
The embodiment of the present specification further provides an electronic device, which includes a memory and a processor, where the memory stores a program and is configured to execute the above-mentioned document pushing method by the processor.
The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects:
whether the user is recommended to open a new biological core mode is judged by combining the client and the server, whether the user has the service requirement for opening the biological core mode at present can be accurately identified, the biological core mode suitable for the user is recommended to the user, and the pushing real-time performance and the pushing accuracy of the biological core mode are better.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification and are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and together with the description serve to explain the specification and not to limit the specification in a non-limiting sense. In the drawings:
fig. 1 is a schematic system structure diagram of a document pushing method provided in an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a document pushing method provided in an embodiment of the present disclosure;
FIG. 3 is a flow chart of another document pushing method provided in the embodiments of the present disclosure;
FIG. 4 is a schematic structural diagram of a document pushing device provided in an embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of another document pushing device provided in the embodiments of the present disclosure;
fig. 6 is a schematic structural diagram of an electronic device provided in an embodiment of the present specification.
Detailed Description
In order to make the objects, technical solutions and advantages of the present disclosure more clear, the technical solutions of the present disclosure will be clearly and completely described below with reference to the specific embodiments of the present disclosure and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without any creative effort belong to the protection scope of the present specification.
In practical applications, the core-body mode is divided into a biological core-body mode and a non-biological core-body mode.
It is understood that the biometric authentication method is a method for authenticating a user by using a biometric recognition technology. Examples of the biometric authentication method include, but are not limited to, a fingerprint authentication method, a face-brushing authentication method, and a face-beautifying authentication method.
The fingerprint verification mode is that whether the fingerprint characteristics input by the user are consistent with the prestored fingerprint characteristics or not, if so, the user is determined to pass the identity verification, and if not, the user is determined not to pass the identity verification.
The face brushing and body checking mode is that whether the facial features input by the user are consistent with the facial features of the authority database or not, if yes, the user is determined to pass the identity authentication, and if not, the user is determined not to pass the identity authentication. The authoritative database is a database with high credibility provided by an authority, and is set according to actual business requirements, and the authoritative database is a population identity information database of a related department.
The face appearance body checking mode is that whether the facial features input by the user are consistent with the reserved facial features or not, if yes, the user is determined to pass the identity authentication, and if not, the user is determined not to pass the identity authentication. It is noted that the reserved facial features are biometric features extracted from a user-specified picture including his face.
It is understood that the non-biometric authentication approach is an approach for authenticating a user without biometric identification techniques. The non-biological body verification method is, for example, a password body verification method, a short message verification code body verification method, and the like.
The password authentication mode requires that a user inputs a password first, then judges whether the password input by the user is consistent with a pre-stored registered password, if so, determines that the user passes authentication, and if not, determines that the user does not pass authentication.
The short message verification code verification method comprises the steps that a user inputs a short message verification code first, then whether the short message verification code input by the user is consistent with a short message verification code sent to the user is judged, if yes, the user is determined to pass identity verification, and if not, the user is determined not to pass identity verification.
Because the biological core mode is a safe and convenient core mode, in order to realize rapid identity recognition and improve the core experience of the user, more and more business scenes recommend the user to adopt the biological core mode for identity recognition. At present, the server analyzes historical data of a user to determine a biometric way to push to the user. However, the above recommendation method is poor in real-time performance, and the recommended biometric authentication method may not be suitable for the user.
In order to solve the technical problem, the method for pushing the document provided in the embodiment of the present specification can accurately identify whether the user has a current business requirement for opening the bio-core mode, recommend the bio-core mode suitable for the user to the user, and has better real-time performance and higher accuracy in pushing the bio-core mode.
It should be noted that the document pushing method provided in the embodiment of the present specification is applicable to an application scenario that requires verification. The application scenario is a payment scenario or a login application scenario, but is not limited thereto.
It should be noted that the document pushing method provided in the embodiment of the present specification is suitable for recommending a biological core mode that has not been opened to a user. It is understood that when a user has opened a certain biometric authentication method, the user is not recommended the opened biometric authentication method.
It should be noted that, in the document pushing method provided in the embodiment of the present specification, it needs to be ensured that the user opens at least one core mode, and the core mode may be a non-biological core mode or a biological core mode.
Fig. 1 is a schematic system structure diagram of a document pushing method provided in an embodiment of the present specification. As shown in fig. 1, the system includes a client and a server. The client may be configured in an electronic device of the user, and the server may be any type of electronic device, such as a mobile phone, a tablet computer, an in-vehicle device, a wearable device, and the like, but not limited thereto.
The document pushing method provided by the embodiment of the specification is a document pushing method combining a client and a server. Specifically, a first pushing result is determined by the client, the first pushing result comprises a first pushing probability of each candidate case, and each candidate case is a case in a biological core pushing mode. Then, the client sends the first pushing result to the server, at this time, the server acquires the characteristics of the server, and inputs the acquired characteristics of the server and the first pushing result into a local second file pushing model of the server to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability of each candidate file; and then, the server side sends the second pushing result to the client side, the client side makes a decision according to the second pushing result to determine whether the file needs to be pushed to the user, and when the file needs to be pushed to the user, a target file is selected from all candidate files and pushed to the user.
Since the client is the link closest to the user, the first push result determined by the client is likely to be close to the biometric way expected by the user, and the first push result determined by the client has better real-time performance. Therefore, the server determines the second pushing result by combining the first pushing result sent by the client and the server characteristics, so that whether the user has a service requirement for opening the biological core mode at present can be accurately identified, the time-efficient biological core mode recommendation can be performed for the user, and the accuracy of the biological core mode recommendation is improved.
The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.
Fig. 2 is a schematic flow chart of a document pushing method provided in an embodiment of the present disclosure. The method is applied to a server, and the server can be any type of electronic equipment, including but not limited to: the system comprises a mobile phone, a tablet personal computer, intelligent wearable equipment, a vehicle machine, a personal computer, a large and medium-sized computer, a computer cluster and the like.
As shown in fig. 2, the document pushing method may include the following steps:
step 202: receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user.
In embodiments of the present description, the bio-nuclear approach includes any of: the fingerprint body-checking mode, the face-brushing body-checking mode, and the face-beautifying body-checking mode, but are not limited thereto.
In the embodiment of the specification, the candidate file is a candidate file for pushing a biological core mode. The number of the documents of each biological nuclear mode can be one or more, and the number of the documents is determined according to the actual business requirement.
It can be understood that the diversified paperwork for pushing the biological core mode can meet the recommendation requirement of opening the biological core mode of the user, so that the user can upgrade the biological core mode into the biological core mode using different scenes by using only one biological core mode, and the core experience of the user is improved.
For example, some mobile phones are provided with fingerprint modules of older models, the success rate of fingerprint body verification is reduced due to aging of sensors, failure of keys and the like, and at this time, a face-brushing body verification mode or a face-beautifying body verification mode can be recommended to a user. For another example, the screen for brushing face at night is too bright, and the mask must be taken off when the face is brushed off online in an epidemic situation, and at this time, the fingerprint body-checking mode can be recommended to the user.
Taking a payment scenario as an example, the recommended copy types for face-brushing payment may include the following: a sense of security type, a convenience type, a public type, a praise type. Wherein, the security type case can be 'safe payment, you are worth owning'. The convenient type of file can be 'your ware of paying, swish face and pay the happy one step'. The most popular types of documents can be "do you use them by everyone, do not drive up to try? ". A quart-type document may be "let the password hide in your color value".
In practical application, the file of each biological core mode can be determined by comprehensively considering the characteristics of the user, the biological core mode, the scene characteristic and the like. For example, a sense of security type may be recommended for older users; a convenient type of paperwork can be recommended for an offline payment scene; a popular type of literature may be recommended for young users; a praise type of document may be recommended for users who like to be praise.
Optionally, the first pushing result is obtained by processing the client characteristics by a first document pushing model local to the client. The first document pushing model can process a multi-classification problem and determine the probability of each candidate document being pushed to the user according to the input client characteristics.
Specifically, a first document push model may be deployed locally at the client, the first document push model being capable of outputting a first push result based on the input client characteristics.
It can be understood that, compared with the process of executing the client features by the server, the process of executing the client features by the client can reduce the consumption of the memory and the computing resources of the server, and reduce the system pressure.
The client characteristics can reflect the current core experience of the user most, and the current requirements of the user can be accurately understood by recommending based on the client characteristics. When the client-side characteristics comprise more user characteristics, the understanding of the user is more comprehensive, and the accuracy of the biological core mode recommendation can be improved. In addition, when the client characteristics comprise the user characteristics, the client characteristics are not convenient to upload to the server, so that the client executes the processing flow of the client characteristics instead of the processing flow of the client characteristics, and the safety risk of revealing the privacy information of the user caused by uploading the client characteristics to the server can be avoided. And then on the premise of protecting the privacy of the user, more real-time client characteristics including the user characteristics can be obtained, and the accuracy of recommending the biological core mode is improved.
Optionally, the first pattern pushing model may be an LR (Logistic Regression) model, an XGBoost (eXtreme Gradient Boosting) model, or a neural network model, but is not limited thereto. The LR model, the XGboost model and the neural network model can all process multi-classification problems.
Preferably, the first document pushing model is a lightweight LR model, so as to reduce consumption of computing resources of the client by the first document pushing model as much as possible and avoid too fast power consumption of the electronic device where the client is located as much as possible.
In the embodiment of the specification, the client characteristics can best reflect the current service requirements of the user, and the first recommendation result obtained by processing the client characteristics by the client has good real-time performance, so that the subsequent server is facilitated to improve the accuracy of the core-body mode recommendation.
In the embodiment of the present specification, the client characteristics are related to an actual application scenario, different application scenarios may have different client characteristics, and corresponding client characteristics are obtained according to specific core service requirements.
As an example, the client-side features include one or more of the following: the client-side user characteristics, the device characteristics of the electronic device where the client is located, and the environment characteristics of the environment where the electronic device is located.
The client-side user characteristics, which are mainly characteristics that have an influence on the biometric way, can be understood as user characteristics obtained by the client through the analysis of the user data.
The device characteristics of the electronic device where the client is located are mainly characteristics affecting the biological core mode, and the device characteristics are, for example, acceleration of the electronic device, sensitivity of the fingerprint module, and the like.
The environmental characteristics of the environment in which the electronic device is located are mainly characteristics that affect the biological nuclear manner, and the environmental characteristics are, for example, the illumination intensity of the environment in which the electronic device is located.
Taking the application scenario as an example of a payment scenario, the client characteristics include one or more of the following characteristics: the method comprises the following steps of setting the page staying time of a kernel page, setting the page behavior track of the kernel page, configuring the illumination intensity of the environment where the electronic equipment of the client is located, setting the acceleration of the electronic equipment and setting the sensitivity of a fingerprint module in the electronic equipment. The page dwell time of the core page and the page behavior trajectory of the core page may be understood as the user characteristics of the client side.
The core page is a page of a core mode which is opened by a user. Analyzing the page dwell time of the core page can determine whether the user will be strong enough for the current core mode. For example, if the current core mode is the password core mode, if the stay time of the user on the page of the password core mode is long, the user prefers the password core mode, and other core modes may not be recommended to the user.
The page behavior track of the core page refers to the click sequence of the user on the core page. Analyzing the page behavior trace of the kernel page can judge whether the user is hesitant to the kernel mode corresponding to the kernel page, and if so, the result shows that other kernel modes can be pushed to the user.
The illumination intensity of the environment where the electronic device configured with the client is located can be used for judging whether the face brushing and body checking mode or the face appearance and body checking mode is suitable. For example, the illumination intensity is less than a preset illumination intensity threshold, and the illumination intensity is considered to be less, which indicates that the electronic device is in a dark environment and is not very suitable for recommending a face brushing and body shaping manner or a face appearance and body shaping manner.
The acceleration of the electronic device can be used for judging whether the face brushing and body checking mode or the face appearance and body checking mode is suitable. For example, if the acceleration of the electronic device is greater than a preset acceleration threshold, the acceleration of the electronic device is considered to be greater, which indicates that the face brushing and body shaping mode or the face appearance and body shaping mode is not suitable for recommendation.
Wherein, the sensitivity of fingerprint module can be used for judging whether be fit for carrying out the fingerprint and check the body mode among the electronic equipment. For example, the sensitivity of fingerprint module is greater than preset sensitivity threshold, and the sensitivity of fingerprint module is considered poor this moment, and it is unsuitable to recommend the fingerprint to check the body mode.
Step 204: the method comprises the steps of obtaining server-side characteristics, inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises second pushing probabilities corresponding to the candidate files, and the second pushing probabilities represent probabilities of the candidate files pushed to a user.
In the embodiment of the present specification, a second document pushing model is deployed at the server, where the second document pushing model may be an LR (logical Regression) model, an XGBoost (eXtreme Gradient Boosting) model, or a neural network model, but is not limited thereto. The LR model, the XGboost model and the neural network model can all process multi-classification problems.
Preferably, the second document pushing model is a lightweight LR model, so as to reduce consumption of computing resources of the server by the second document pushing model as much as possible.
In the embodiment of the description, the server determines the second push result by combining the first push result sent by the client and the server characteristics, so that whether the user has a service requirement for opening the biometric authentication method can be accurately identified, and the time-efficient biometric authentication method recommendation can be performed for the user.
In embodiments of the present specification, the server-side feature may include a real-time feature and an offline feature. The real-time characteristics can help the server to improve the accuracy of the core-body mode recommendation.
In the embodiment of the present specification, the server-side characteristics are related to an actual application scenario, and different application scenarios may have different server-side characteristics, and corresponding client-side characteristics are acquired according to specific core service requirements.
As an example, the server-side features include: user features on the server side. Of course, the server-side characteristics are not limited to the above-mentioned characteristics, and the specific server-side characteristics may be determined according to specific core business requirements.
The client-side user characteristics can be understood as user characteristics obtained by the server through analysis of user data, and the server-side user characteristics are mainly characteristics having an influence on the biometric way.
Taking the application scenario as an example of a payment scenario, the real-time features include one or more of the following features: the current body checking mode of the user, whether the current body checking mode is switched to the password body checking mode by the user or not and the current payment scene of the user. The offline features include one or more of the following features: the method comprises the steps of historical core-body success rate of a user, preference information of the user to a core-body mode, historical payment information of the user and brand information of electronic equipment of the user. Of course, the above real-time and offline features are not limited to the above illustration.
The current core body mode of the user indicates that the user opens the corresponding core body mode, and the core body mode does not need to be opened, at the moment, the probability that the server side determines that the opened core body mode is pushed to the user is low, so that a subsequent client side does not recommend the opened core body mode to the user, and disturbance to the user is reduced.
Whether the current core mode is switched to the password core mode by the user or not indicates that the user probably does not like the biological core mode, and at the moment, the server gives smaller probability to each biological core mode, so that a subsequent client does not recommend the biological core mode to the user, and the disturbance to the user is reduced.
The current payment scene of the user is, for example, an offline payment scene and an online payment scene, the offline payment scene is suitable for a face brushing and body checking mode and a face appearance and body checking mode and is not suitable for a fingerprint body checking mode, and the online payment scene is suitable for a face brushing and body checking mode, a face appearance and body checking mode and a fingerprint body checking mode.
The historical nuclear body success rate of the user can be used for judging whether the experience of the nuclear body mode which is opened by the user is good or not, and if not, the unopened biological nuclear body mode can be recommended to the user.
The preference information of the user for the core body mode can be used for judging whether the user has a preference for the specific core body mode, and if the user has the preference for the specific core body mode, the unopened biological core body mode can not be recommended to the user.
The historical payment information of the user can be used for judging whether the user has more online payments or more offline payments; if the offline payment is more, a face brushing and body checking mode and a face appearance and body checking mode can be recommended; if the online payment is more, a face brushing and body checking mode, a face appearance and body checking mode and a fingerprint body checking mode can be recommended.
The brand information of the electronic device can be used for judging whether the biological nuclear mode is supported.
Step 206: and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
In the embodiment of the specification, whether the user is recommended to open a new biological core mode is judged by combining the client and the server, and when the user needs to open the new biological core mode, which type of document is adopted and the conversion effect of pushing the document to the user is good, so that the requirement of opening various biological core modes by the user is met, resource waste caused by frequently recommending the biological core mode to the user can be avoided, frequent disturbance to the user is effectively avoided, and good core experience is ensured.
For example, the client analyzes the second push result and determines that the second push probability of none of the candidate documents is greater than the push probability threshold, at which point the client may determine that the documents need not be pushed to the user.
For another example, the client analyzes the second pushing result, and determines that the second pushing probability of at least one candidate document is greater than the pushing probability threshold, at this time, the client may determine that the document needs to be pushed to the user, and may push the candidate document with the highest second pushing probability as the target document to the user. Of course, if it is detected that the user refuses to push the target document, the remaining candidate documents with higher second push probability may be sequentially pushed to the user as new target documents according to the sequence from the second push probability to the second push probability, until the maximum push frequency is reached or it is detected that the user accepts the pushed target documents. The maximum pushing times are set according to actual service requirements, for example, the maximum pushing times is 2 times. The probability of the document being pushed to the user is greater than a push probability threshold, which depends on the actual situation.
According to the file pushing method provided by the embodiment of the specification, whether the user is recommended to open the new biological core mode is judged in a mode of combining the client and the server, whether the user has the service requirement of opening the biological core mode at present can be accurately identified, the biological core mode suitable for the user is recommended to the user, and the pushing of the biological core mode is better in real time and higher in accuracy.
On the basis of the foregoing embodiment, optionally, the first pushing result is obtained by processing client characteristics with a first document pushing model local to the client, and the way for the server to train the first document pushing model is as follows: obtaining a first sample set, wherein a first sample in the first sample set comprises one or more client features; determining a first labeling result of the first sample, wherein the first labeling result is used for indicating a desired push file corresponding to the first sample; training an initial first pattern pushing model according to the first sample set and each first labeling result; and if the training round number reaches the maximum training round number or the loss function of the trained first pattern pushing model is converged, stopping training to obtain the first pattern pushing model, wherein the output result of the first pattern pushing model is the actual pushed pattern of the first sample and the probability thereof.
In this embodiment, the server may train the first document pushing model and push the first document pushing model to the client. The server side trains the first document pushing model, so that the first document pushing model can be conveniently pushed to each client side, and the first document pushing model on each client side can be conveniently updated.
On the basis of the above embodiment, optionally, the way for the server to train the second pattern push model is as follows: obtaining a second sample set, wherein a second sample in the second sample set comprises at least one server-side feature and a model output result of a first sample combined with the at least one server-side feature, and the model output result of the first sample is obtained through the first file push model; determining a second labeling result of the second sample, wherein the second labeling result is used for indicating a desired push file corresponding to the second sample; training an initial second pattern pushing model according to the second sample set and each second labeling result; and if the training round number reaches the maximum training round number or the loss function of the trained second pattern pushing model is converged, stopping training, and determining the trained second pattern pushing model as the second pattern pushing model, wherein the output result of the second pattern pushing model is the actual pushed pattern of the second sample and the probability thereof.
It should be noted that the maximum number of training rounds and the loss function involved in the model training can be set according to actual situations. In this case, a gradient descent method may be used in calculating the loss function. An exemplary loss function is as follows:
where m is the total number of samples, x(i)For the ith sample, y(i)Is the actual output result of the ith sample, hθ(x(i)) For the expected output result of the ith sample, J(θ)Solving the optimal theta for the loss function with theta as a parameter to obtain a loss function J(θ)Taking the minimum value.
It should be noted that if there are N types of biological nuclear systems, the following situations need to be considered when training the model: the user does not turn on any one of the bio-core mode, the user turns on two of the bio-core mode … … the user turns on N-1 bio-core modes. And preparing a first sample set and a second sample set according to the situations, labeling the first sample in the first sample set, labeling the second sample in the second sample set, wherein the labeled first sample set and the labeled second sample set can cover the situations needing to be considered, so that the trained first and second case recommendation models can perform case recommendation according to the situations.
Taking 3 body-checking modes such as a fingerprint body-checking mode, a face-brushing body-checking mode, a face-beautifying body-checking mode and the like as examples, the following conditions need to be considered:
under the condition that any biological body verification mode is not opened, recommending a fingerprint body verification mode;
under the condition that any biological core body mode is not opened, a face brushing core body mode is recommended;
under the condition that any biological nuclear mode is not opened, a face-to-face nuclear mode is recommended;
under the condition that the fingerprint body checking mode is opened, a face brushing body checking mode is recommended;
recommending a face-to-face body-checking mode under the condition that the fingerprint body-checking mode is opened;
under the condition that a face brushing and body checking mode is opened, a fingerprint body checking mode is recommended;
recommending a face-beautifying and body-checking mode under the condition that a face-brushing and body-checking mode is opened;
under the condition that the face-beautifying and body-checking mode is opened, a face-brushing and body-checking mode is recommended;
under the condition that a face-to-face body-checking mode is opened, a fingerprint body-checking mode is recommended;
recommending a face-face body checking mode under the condition that a fingerprint body checking mode and a face-brushing body checking mode are opened;
under the condition that a fingerprint body checking mode and a face-to-face body checking mode are opened, a face brushing body checking mode is recommended;
and recommending a fingerprint body checking mode under the condition that a face brushing body checking mode and a face appearance body checking mode are opened.
Specifically, the first and second document recommendation models obtained by training may recommend documents according to the above situations, and recommend a biological core manner that has not been activated by the user, but not recommend a biological core manner that has been activated by the user.
Fig. 3 is a schematic flow chart of another document pushing method provided in the embodiments of the present disclosure. The method is applied to a client, and the client can be configured on any type of electronic device, including but not limited to: the system comprises a mobile phone, a tablet personal computer, intelligent wearable equipment, a vehicle machine, a personal computer, a large and medium-sized computer, a computer cluster and the like.
As shown in fig. 3, the document pushing method may include the following steps:
step 302: and responding to the target operation to acquire the client characteristics.
In the embodiment of the present specification, the target operation may be regarded as a trigger condition for making a case recommendation. In different application scenarios, the target operation is different. For example, in a payment scenario, the target operation may be a payment completion operation. In the scenario of the login application, the target operation is a confirm login operation.
In the embodiment of the specification, the client characteristics can best reflect the current service requirements of the user, and the first recommendation result obtained by processing the client characteristics by the client has good real-time performance, so that the subsequent server is facilitated to improve the accuracy of the core-body mode recommendation.
In the embodiment of the present specification, the client characteristics are related to an actual application scenario, different application scenarios may have different client characteristics, and corresponding client characteristics are obtained according to specific core service requirements.
As an example, the client-side features include one or more of the following: the client-side user characteristics, the device characteristics of the electronic device where the client is located, and the environment characteristics of the environment where the electronic device is located.
The client-side user characteristics, which are mainly characteristics that have an influence on the biometric way, can be understood as user characteristics obtained by the client through the analysis of the user data.
The device characteristics of the electronic device where the client is located are mainly characteristics affecting the biological core mode, and the device characteristics are, for example, acceleration of the electronic device, sensitivity of the fingerprint module, and the like.
The environmental characteristics of the environment in which the electronic device is located are mainly characteristics that affect the biological nuclear manner, and the environmental characteristics are, for example, the illumination intensity of the environment in which the electronic device is located.
Taking the application scenario as an example of a payment scenario, the client characteristics include one or more of the following characteristics: the method comprises the following steps of setting the page staying time of a kernel page, setting the page behavior track of the kernel page, configuring the illumination intensity of the environment where the electronic equipment of the client is located, setting the acceleration of the electronic equipment and setting the sensitivity of a fingerprint module in the electronic equipment. The page dwell time of the core page and the page behavior trajectory of the core page may be understood as the user characteristics of the client side.
The core page is a page of a core mode which is opened by a user. Analyzing the page dwell time of the core page can determine whether the user will be strong enough for the current core mode. For example, if the current core mode is the password core mode, if the stay time of the user on the page of the password core mode is long, the user prefers the password core mode, and other core modes may not be recommended to the user.
The page behavior track of the core page refers to the click sequence of the user on the core page. Analyzing the page behavior trace of the kernel page can judge whether the user is hesitant to the kernel mode corresponding to the kernel page, and if so, the result shows that other kernel modes can be pushed to the user.
The illumination intensity of the environment where the electronic device configured with the client is located can be used for judging whether the face brushing and body checking mode or the face appearance and body checking mode is suitable. For example, the illumination intensity is less than a preset illumination intensity threshold, and the illumination intensity is considered to be less, which indicates that the electronic device is in a dark environment and is not very suitable for recommending a face brushing and body shaping manner or a face appearance and body shaping manner.
The acceleration of the electronic device can be used for judging whether the face brushing and body checking mode or the face appearance and body checking mode is suitable. For example, if the acceleration of the electronic device is greater than a preset acceleration threshold, the acceleration of the electronic device is considered to be greater, which indicates that the face brushing and body shaping mode or the face appearance and body shaping mode is not suitable for recommendation.
Wherein, the sensitivity of fingerprint module can be used for judging whether be fit for carrying out the fingerprint and check the body mode among the electronic equipment. For example, the sensitivity of fingerprint module is greater than preset sensitivity threshold, and the sensitivity of fingerprint module is considered poor this moment, and it is unsuitable to recommend the fingerprint to check the body mode.
Step 304: inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating the biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to the user.
In an embodiment of the present specification, a first document push model may be deployed locally at the client, and the first document push model may be capable of outputting a first push result based on the input client characteristics. The first document pushing model can process a multi-classification problem and determine the probability of each candidate document being pushed to the user according to the input client characteristics.
It can be understood that, compared with the process of executing the client features by the server, the process of executing the client features by the client can reduce the consumption of the memory and the computing resources of the server, and reduce the system pressure.
The client characteristics can reflect the current core experience of the user most, and the current requirements of the user can be accurately understood by recommending based on the client characteristics. When the client-side characteristics comprise more user characteristics, the understanding of the user is more comprehensive, and the accuracy of the biological core mode recommendation can be improved. In addition, when the client characteristics comprise the user characteristics, the client characteristics are not convenient to upload to the server, so that the client executes the processing flow of the client characteristics instead of the processing flow of the client characteristics, and the safety risk of revealing the privacy information of the user caused by uploading the client characteristics to the server can be avoided. And then on the premise of protecting the privacy of the user, more real-time client characteristics including the user characteristics can be obtained, and the accuracy of recommending the biological core mode is improved.
Optionally, the first pattern pushing model may be an LR (Logistic Regression) model, an XGBoost (eXtreme Gradient Boosting) model, or a neural network model, but is not limited thereto. The LR model, the XGboost model and the neural network model can all process multi-classification problems.
Preferably, the first document pushing model is a lightweight LR model, so as to reduce consumption of computing resources of the client by the first document pushing model as much as possible and avoid too fast power consumption of the electronic device where the client is located as much as possible.
In the embodiment of the specification, the client characteristics can best reflect the current service requirements of the user, and the first recommendation result obtained by processing the client characteristics by the client has good real-time performance, so that the subsequent server is facilitated to improve the accuracy of the core-body mode recommendation.
Step 306: and sending the first pushing result to a server, wherein the first pushing result is used for the server to input the first pushing result and the server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user.
It should be noted that, for the content of the service end, reference may be made to the description of the foregoing embodiment, and details of this embodiment are not described again.
Step 308: and receiving a second pushing result returned by the server.
Step 310: and judging whether to push the documents to the user according to the second pushing result, and if the documents are judged to be required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
In the embodiment of the specification, whether the user is recommended to open a new biological core mode is judged by combining the client and the server, and when the user needs to open the new biological core mode, which type of document is adopted and the conversion effect of pushing the document to the user is good, so that the requirement of opening various biological core modes by the user is met, resource waste caused by frequently recommending the biological core mode to the user can be avoided, frequent disturbance to the user is effectively avoided, and good core experience is ensured.
For example, the client analyzes the second push result and determines that the second push probability of none of the candidate documents is greater than the push probability threshold, at which point the client may determine that the documents need not be pushed to the user.
For another example, the client analyzes the second pushing result, and determines that the second pushing probability of at least one candidate document is greater than the pushing probability threshold, at this time, the client may determine that the document needs to be pushed to the user, and may push the candidate document with the highest second pushing probability as the target document to the user. Of course, if it is detected that the user refuses to push the target document, the remaining candidate documents with higher second push probability may be sequentially pushed to the user as new target documents according to the sequence from the second push probability to the second push probability, until the maximum push frequency is reached or it is detected that the user accepts the pushed target documents. The maximum pushing times are set according to actual service requirements, for example, the maximum pushing times is 2 times. The probability of the document being pushed to the user is greater than a push probability threshold, which depends on the actual situation.
According to the file pushing method provided by the embodiment of the specification, whether the user is recommended to open the new biological core mode is judged in a mode of combining the client and the server, whether the user has the service requirement of opening the biological core mode at present can be accurately identified, the biological core mode suitable for the user is recommended to the user, and the pushing of the biological core mode is better in real time and higher in accuracy.
The embodiment of the specification also provides a document pushing device corresponding to the document pushing device shown in fig. 2. Fig. 4 is a schematic structural diagram of a document pushing device provided in an embodiment of the present disclosure. As shown in fig. 4, the document pushing apparatus includes:
the first receiving module 10 is configured to receive a first pushing result sent by a client, where the first pushing result includes a first pushing probability corresponding to each candidate document indicating a biometric way, and the first pushing probability represents a probability that the corresponding candidate document is pushed to a user;
the first processing module 20 is configured to obtain a server-side feature, and input the first push result and the server-side feature into a local second document push model of the server to obtain a second push result, where the second push result includes a second push probability corresponding to each candidate document, and the second push probability represents a probability that the corresponding candidate document is pushed to a user;
the first sending module 30 sends the second pushing result to the client, where the second pushing result is used by the client to determine whether to push the documents to the user according to the second pushing result, and if it is determined that the documents need to be pushed to the user, the second pushing result is also used by the client to select a target document from the candidate documents according to the second pushing result and push the target document to the user.
Further, the first pushing result is obtained by processing client characteristics with a first document pushing model local to the client, and the way for the server to train the first document pushing model is as follows:
obtaining a first sample set, wherein a first sample in the first sample set comprises one or more client features;
determining a first labeling result of the first sample, wherein the first labeling result is used for indicating a desired push file corresponding to the first sample;
training an initial first pattern pushing model according to the first sample set and each first labeling result;
and if the training round number reaches the maximum training round number or the loss function of the trained first pattern pushing model is converged, stopping training to obtain the first pattern pushing model, wherein the output result of the first pattern pushing model is the actual pushed pattern of the first sample and the probability thereof.
Further, the way for the server to train the second pattern pushing model is as follows:
obtaining a second sample set, wherein a second sample in the second sample set comprises at least one server-side feature and a model output result of a first sample combined with the at least one server-side feature, and the model output result of the first sample is obtained through the first file push model;
determining a second labeling result of the second sample, wherein the second labeling result is used for indicating a desired push file corresponding to the second sample;
training an initial second pattern pushing model according to the second sample set and each second labeling result;
and if the training round number reaches the maximum training round number or the loss function of the trained second pattern pushing model is converged, stopping training, and determining the trained second pattern pushing model as the second pattern pushing model, wherein the output result of the second pattern pushing model is the actual pushed pattern of the second sample and the probability thereof.
Further, the server-side feature includes: user features on the server side.
Further, the server-side feature includes: real-time features and offline features;
wherein the real-time features include one or more of the following features: the current body checking mode of the user, whether the current body checking mode is switched to a password body checking mode by the user or not, and the current payment scene of the user;
wherein the offline features include one or more of the following features: the historical core success rate of the user, the preference information of the user to the core mode, the historical payment information of the user and the brand information of the electronic equipment of the user.
Further, the biological nuclear mode comprises any one of the following modes:
fingerprint body-checking mode, face-brushing body-checking mode and face-beautifying body-checking mode.
The apparatuses provided in this specification correspond to the methods provided in this application one to one, and therefore, the apparatuses also have advantageous technical effects similar to the methods, and since the advantageous technical effects of the methods have been described in detail above, the advantageous technical effects of the apparatuses are not described herein again.
The embodiment of the specification also provides a document pushing device corresponding to the document pushing device in figure 3. Fig. 5 is a schematic structural diagram of another document pushing device provided in the embodiments of the present disclosure. As shown in fig. 5, the document pushing device includes:
the acquisition module 40 responds to the target operation and acquires the client characteristics;
the second processing module 50 is configured to input the client characteristics into a first file pushing model local to the client to obtain a first pushing result, where the first pushing result includes first pushing probabilities corresponding to the candidate files for indicating the biometric authentication mode, and the first pushing probabilities represent probabilities of the corresponding candidate files being pushed to the user;
a second sending module 60, configured to send the first pushing result to a server, where the first pushing result is used for the server to input the first pushing result and server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result includes a second pushing probability corresponding to each candidate file, and the second pushing probability represents a probability that the corresponding candidate file is pushed to a user;
a second receiving module 70, configured to receive a second pushing result returned by the server;
the second processing module 50 determines whether to push the documents to the user according to the second pushing result, and if it is determined that the documents need to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
Further, the client-side features include one or more of the following features:
the user characteristics of the client side, the equipment characteristics of the electronic equipment where the client is located, and the environment characteristics of the environment where the electronic equipment is located.
Further, the client-side features include one or more of the following features: the method comprises the following steps of setting the page staying time of a kernel page, setting the page behavior track of the kernel page, configuring the illumination intensity of the environment where the electronic equipment of the client is located, setting the acceleration of the electronic equipment and setting the sensitivity of a fingerprint module in the electronic equipment.
The apparatuses provided in this specification correspond to the methods provided in this application one to one, and therefore, the apparatuses also have advantageous technical effects similar to the methods, and since the advantageous technical effects of the methods have been described in detail above, the advantageous technical effects of the apparatuses are not described herein again.
The embodiment of the specification also provides electronic equipment. Fig. 6 is a schematic structural diagram of an electronic device provided in an embodiment of the present specification. As shown in fig. 6, the electronic apparatus includes: a memory 11 and a processor 12, the memory 11 storing a program and configured to perform the following steps by the processor 12:
receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
obtaining server-side characteristics, and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
And/or, configured to perform the following steps by the processor 12:
responding to the target operation and acquiring the characteristics of the client;
inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating a biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
sending the first pushing result to a server, wherein the first pushing result is used for the server to input the first pushing result and server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
receiving a second pushing result returned by the server;
and judging whether to push the documents to the user according to the second pushing result, and if the documents are judged to be required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
The embodiment of the specification also provides a computer readable medium corresponding to the method. The computer readable medium has computer readable instructions stored thereon that are executable by a processor to implement the method of:
receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
obtaining server-side characteristics, and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
And/or, the computer readable instructions are executable by a processor to implement a method of:
responding to the target operation and acquiring the characteristics of the client;
inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating a biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
sending the first pushing result to a server, wherein the first pushing result is used for the server to input the first pushing result and server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
receiving a second pushing result returned by the server;
and judging whether to push the documents to the user according to the second pushing result, and if the documents are judged to be required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Language Description Language), traffic, pl (core unified Programming Language), HDCal, JHDL (Java Hardware Description Language), langue, Lola, HDL, laspam, hardsradware (Hardware Description Language), vhjhd (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the various elements may be implemented in the same one or more software and/or hardware implementations of the present description.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.
Claims (12)
1. A file pushing method is applied to a server and comprises the following steps:
receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
obtaining server-side characteristics, and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and sending the second pushing result to the client, wherein the second pushing result is used for the client to judge whether to push the file to the user according to the second pushing result, and if the file needs to be pushed to the user, the second pushing result is also used for the client to select a target file from the candidate files according to the second pushing result and push the target file to the user.
2. The method as claimed in claim 1, wherein the first pushing result is obtained by processing client characteristics with a first document pushing model local to the client, and the way for the server to train the first document pushing model is as follows:
obtaining a first sample set, wherein a first sample in the first sample set comprises one or more client features;
determining a first labeling result of the first sample, wherein the first labeling result is used for indicating a desired push file corresponding to the first sample;
training an initial first pattern pushing model according to the first sample set and each first labeling result;
and if the training round number reaches the maximum training round number or the loss function of the trained first pattern pushing model is converged, stopping training to obtain the first pattern pushing model, wherein the output result of the first pattern pushing model is the actual pushed pattern of the first sample and the probability thereof.
3. The method of claim 2, wherein the way for the server to train the second pattern pushing model is as follows:
obtaining a second sample set, wherein a second sample in the second sample set comprises at least one server-side feature and a model output result of a first sample combined with the at least one server-side feature, and the model output result of the first sample is obtained through the first file push model;
determining a second labeling result of the second sample, wherein the second labeling result is used for indicating a desired push file corresponding to the second sample;
training an initial second pattern pushing model according to the second sample set and each second labeling result;
and if the training round number reaches the maximum training round number or the loss function of the trained second pattern pushing model is converged, stopping training, and determining the trained second pattern pushing model as the second pattern pushing model, wherein the output result of the second pattern pushing model is the actual pushed pattern of the second sample and the probability thereof.
4. The method of any of claims 1 to 3, the server-side features comprising: user features on the server side.
5. The method of any of claims 1 to 3, the server-side features comprising: real-time features and offline features;
wherein the real-time features include one or more of the following features: the current body checking mode of the user, whether the current body checking mode is switched to a password body checking mode by the user or not, and the current payment scene of the user;
wherein the offline features include one or more of the following features: the historical core success rate of the user, the preference information of the user to the core mode, the historical payment information of the user and the brand information of the electronic equipment of the user.
6. The method of any one of claims 1 to 3, wherein the biosuclear process comprises any one of:
fingerprint body-checking mode, face-brushing body-checking mode and face-beautifying body-checking mode.
7. A file pushing method is applied to a client and comprises the following steps:
responding to the target operation and acquiring the characteristics of the client;
inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating a biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
sending the first pushing result to a server, wherein the first pushing result is used for the server to input the first pushing result and server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
receiving a second pushing result returned by the server;
and judging whether to push the documents to the user according to the second pushing result, and if the documents are judged to be required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
8. The method of claim 7, the client-side features comprising one or more of the following:
the user characteristics of the client side, the equipment characteristics of the electronic equipment where the client is located, and the environment characteristics of the environment where the electronic equipment is located.
9. The method of claim 7, the client-side features comprising one or more of the following: the method comprises the following steps of setting the page staying time of a kernel page, setting the page behavior track of the kernel page, configuring the illumination intensity of the environment where the electronic equipment of the client is located, setting the acceleration of the electronic equipment and setting the sensitivity of a fingerprint module in the electronic equipment.
10. A document pushing device, comprising:
the first receiving module is used for receiving a first pushing result sent by a client, wherein the first pushing result comprises a first pushing probability corresponding to each candidate file for indicating a biological core mode, and the first pushing probability represents the probability that the corresponding candidate file is pushed to a user;
the first processing module is used for acquiring server-side characteristics and inputting the first pushing result and the server-side characteristics into a local second file pushing model of the server side to obtain a second pushing result, wherein the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
and the first sending module is used for sending the second pushing result to the client, the second pushing result is used for judging whether to push the file to the user according to the second pushing result by the client, and if the file needs to be pushed to the user, the second pushing result is also used for selecting a target file from the candidate files to push the target file to the user according to the second pushing result by the client.
11. A document pushing device, comprising:
the acquisition module is used for responding to the target operation and acquiring the characteristics of the client;
the second processing module is used for inputting the client characteristics into a first file pushing model local to the client to obtain a first pushing result, wherein the first pushing result comprises first pushing probabilities corresponding to all candidate files for indicating the biological core mode, and the first pushing probabilities represent the probabilities of the corresponding candidate files being pushed to a user;
the second sending module is used for sending the first pushing result to a server, the first pushing result is used for the server to input the first pushing result and the server characteristics into a local second file pushing model of the server to obtain a second pushing result, the second pushing result comprises a second pushing probability corresponding to each candidate file, and the second pushing probability represents the probability that the corresponding candidate file is pushed to a user;
the second receiving module is used for receiving a second pushing result returned by the server;
and the second processing module judges whether to push the documents to the user according to the second pushing result, and if the documents are required to be pushed to the user, the client selects a target document from the candidate documents according to the second pushing result and pushes the target document to the user.
12. An electronic device comprising a memory and a processor, the memory storing a program and configured to perform the pattern push method of any one of claims 1-6 or the pattern push method of any one of claims 7-9 by the processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010733172.5A CN111782960A (en) | 2020-07-27 | 2020-07-27 | File pushing method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010733172.5A CN111782960A (en) | 2020-07-27 | 2020-07-27 | File pushing method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111782960A true CN111782960A (en) | 2020-10-16 |
Family
ID=72765049
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010733172.5A Pending CN111782960A (en) | 2020-07-27 | 2020-07-27 | File pushing method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111782960A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109558199A (en) * | 2018-11-29 | 2019-04-02 | 广州市百果园信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
WO2019174395A1 (en) * | 2018-03-13 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for information recommendation, and device |
WO2020039241A1 (en) * | 2018-08-24 | 2020-02-27 | 优视科技新加坡有限公司 | Message pushing method and apparatus, device/terminal/server, and computer-readable medium |
CN110852793A (en) * | 2019-10-28 | 2020-02-28 | 北京深演智能科技股份有限公司 | Document recommendation method and device and electronic equipment |
-
2020
- 2020-07-27 CN CN202010733172.5A patent/CN111782960A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019174395A1 (en) * | 2018-03-13 | 2019-09-19 | 阿里巴巴集团控股有限公司 | Method and apparatus for information recommendation, and device |
WO2020039241A1 (en) * | 2018-08-24 | 2020-02-27 | 优视科技新加坡有限公司 | Message pushing method and apparatus, device/terminal/server, and computer-readable medium |
CN109558199A (en) * | 2018-11-29 | 2019-04-02 | 广州市百果园信息技术有限公司 | A kind of information-pushing method, device, equipment and storage medium |
CN110852793A (en) * | 2019-10-28 | 2020-02-28 | 北京深演智能科技股份有限公司 | Document recommendation method and device and electronic equipment |
Non-Patent Citations (1)
Title |
---|
宋闻达;: "基于深度检测的用户行为分析以及Web推送的设计与实现", 软件, no. 12 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107358157B (en) | Face living body detection method and device and electronic equipment | |
CN107679686B (en) | Service execution method and device | |
US20220058524A1 (en) | Distributed training of machine learning models for personalization | |
US10692089B2 (en) | User classification using a deep forest network | |
US11093774B2 (en) | Optical character recognition error correction model | |
US20200202068A1 (en) | Computing apparatus and information input method of the computing apparatus | |
CN111325444B (en) | Risk prevention and control decision method, device, system and equipment | |
CN110674188A (en) | Feature extraction method, device and equipment | |
US10831870B2 (en) | Intelligent user identification | |
CN113826135B (en) | System, method and computer system for contactless authentication using voice recognition | |
CN110008394B (en) | Public opinion information identification method, device and equipment | |
CN111292734A (en) | Voice interaction method and device | |
CN113111340A (en) | Method, device, equipment and readable medium for displaying login mode options | |
CN115146601A (en) | Method and device for executing language processing task, readable storage medium and equipment | |
CN110532755B (en) | Computer-implemented risk identification method and device | |
CN114819614A (en) | Data processing method, device, system and equipment | |
CN116909889A (en) | Model risk evaluation method, device and equipment | |
CN110705622A (en) | Decision-making method and system and electronic equipment | |
CN115545720B (en) | Model training method, business wind control method and business wind control device | |
CN113408254A (en) | Page form information filling method, device, equipment and readable medium | |
CN113221717A (en) | Model construction method, device and equipment based on privacy protection | |
JP2021149965A (en) | Report evaluation device and operation method therefor | |
CN112597459A (en) | Identity verification method and device | |
CN111400443A (en) | Information processing method, device and storage medium | |
CN111782960A (en) | File pushing method and device and electronic equipment |
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 |