CN114491246A - Message template recommendation method, device and equipment - Google Patents
Message template recommendation method, device and equipment Download PDFInfo
- Publication number
- CN114491246A CN114491246A CN202210043871.6A CN202210043871A CN114491246A CN 114491246 A CN114491246 A CN 114491246A CN 202210043871 A CN202210043871 A CN 202210043871A CN 114491246 A CN114491246 A CN 114491246A
- Authority
- CN
- China
- Prior art keywords
- merchant
- message template
- template
- type
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 230000000694 effects Effects 0.000 claims abstract description 15
- 238000013507 mapping Methods 0.000 claims description 18
- 238000003062 neural network model Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 230000000875 corresponding effect Effects 0.000 description 24
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 12
- 230000006872 improvement Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000004927 fusion Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 238000004519 manufacturing process 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
- 229920002799 BoPET Polymers 0.000 description 1
- 239000005041 Mylar™ Substances 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit 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
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003442 weekly 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The embodiment of the specification discloses a method, a device and equipment for recommending a message template. Determining the type of the merchant according to the historical frequency by determining the historical frequency of the merchant using the message template, wherein the type of the merchant comprises a first type with the historical frequency of the message template being zero; determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different; acquiring a message template to be recommended meeting the template recommendation index, pushing the message template to be recommended to the merchant, and evaluating the fitness of the pushed message template to the merchant of the first type according to the activity degree of a user in the interface of the merchant of the applet. Therefore, fitness evaluation of the message template is carried out on the merchant newly entering the applet.
Description
Technical Field
The present disclosure relates to the field of internet technologies, and in particular, to a method, an apparatus, and a device for recommending a message template.
Background
Related information is pushed to a user, which is one of important means for merchant digital operation, many current applications provide a small program function, and merchants can enter and reside the applications through the small program and push information to the user by using a message template provided in the small program, so as to inform the user of information such as the result and the state of the current behavior.
Different merchants have different experience and capability of pushing messages to users, so that the application side needs to perform hierarchical operation on the merchants to meet the actual needs of different types of merchants in self-operation.
Based on this, there is a need for a recommendation scheme for message templates that is more efficient for merchant self-operations.
Disclosure of Invention
The embodiment of the specification provides a recommendation method, a recommendation device, recommendation equipment and a storage medium for a message template, and aims to solve the following technical problems: there is a need for a recommendation scheme for message templates that is more efficient for merchant self-operations.
To solve the above technical problem, one or more embodiments of the present specification are implemented as follows:
in a first aspect, an embodiment of the present specification provides a method for recommending a message template, which is applied to an applet, and includes: determining the historical frequency of a merchant using a message template, and determining the merchant type of the merchant according to the historical frequency, wherein the merchant type comprises a first type with the historical frequency of using the message template being zero; determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different; acquiring a message template to be recommended meeting the template recommendation index, and pushing the message template to be recommended to the merchant; and evaluating the fitness of the pushed message template to the first type of merchants according to the activity degree of the user in the interfaces of the merchants of the applets.
In a second aspect, an embodiment of the present specification provides a recommendation apparatus for a message template, which is applied in an applet, and includes: the merchant type determining module is used for determining the historical frequency of the merchant using the message template and determining the merchant type of the merchant according to the historical frequency, wherein the merchant type comprises a first type with zero historical frequency of using the message template; the recommendation index determining module is used for determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different; the acquisition module acquires a message template to be recommended, which meets the template recommendation index; the pushing module is used for pushing the message template to be recommended to the merchant; and the evaluation module is used for evaluating the fitness of the pushed message template to the first type of merchants according to the activity degree of the user in the interfaces of the merchants of the applet.
In a third aspect, embodiments of the present specification provide an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect.
In a fourth aspect, embodiments of the present specification provide a non-volatile computer storage medium having stored thereon computer-executable instructions that, when read by a computer, cause the one or more processors to perform the method of the first aspect.
At least one technical scheme adopted by one or more embodiments of the specification can achieve the following beneficial effects: determining the type of the merchant according to the historical frequency by determining the historical frequency of the merchant using the message template, wherein the type of the merchant comprises a first type with the historical frequency of the message template being zero; determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different; acquiring a message template to be recommended meeting the template recommendation index, pushing the message template to be recommended to the merchant, and evaluating the fitness of the pushed message template to the merchant of the first type according to the activity degree of a user in the interface of the merchant of the applet. Therefore, fitness evaluation of the message template is carried out on the merchant newly entering the applet, so that the message template more suitable for operation of the merchant is recommended, the merchant newly entering the applet is assisted to obtain a better message template, and self-operation efficiency of the newly entering merchant is improved.
Drawings
In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.
Fig. 1 is a flowchart illustrating a method for recommending a message template according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of feature fusion provided in an embodiment of the present disclosure;
fig. 3 is a schematic diagram of a message template display performed at a merchant end according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a recommendation apparatus for a message template according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an electronic device provided in an embodiment of the present specification.
Detailed Description
The embodiment of the specification provides a message template recommendation method, a message template recommendation device, message template recommendation equipment and a storage medium.
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any inventive step based on the embodiments of the present disclosure, shall fall within the scope of protection of the present application.
As shown in fig. 1, fig. 1 is a flowchart illustrating a method for recommending a message template according to an embodiment of the present specification, including the following steps:
s101: determining the historical frequency of the message template used by the merchant, and determining the merchant type of the merchant according to the historical frequency, wherein the merchant type comprises a first type with zero historical frequency of the message template.
Current merchants typically operate digitally through resident applets and push messages to their historical users (i.e., users who have been consumed) or potential users (i.e., users who are more likely to consume at their store) through the applets. At this time, the application server providing the applet can provide various different message templates for the merchant to select.
For example, the message template may be "xxx holiday welfare, fast neck [ holiday gift bag ], N" s per day for the coupon, and provide a relevant link for the user to click into the store. The specific content in the festival gift bag and the specific number of copies of the N can be filled by the merchant based on the needs of the merchant. That is, the message template is a preset template having a fixed format or content. The message template may contain various preset contents such as text, pictures, animations, etc.
When a merchant enters the merchant interface of the merchant through the applet, the application server can obtain the historical frequency of the merchant using the message template based on the historical data query of the merchant. The frequency may include a frequency (e.g., number of monthly or weekly uses) or a number (e.g., number of uses of the merchant in the past three months, or number of uses of the merchant after the application is resident through the applet), etc.
Further, the merchants can be classified according to the use conditions of the merchants for the message templates. For example, merchants can be classified into three types according to their historical frequency of use of the template:
a. when the historical frequency of the merchant using the message template is zero, determining that the merchant type is a first type, wherein the first type is a merchant who never uses the template, for example, a merchant who just resides in the application;
b. when the historical frequency of the merchant using the message template is greater than M, determining that the merchant type is a second type, namely, the merchant with more abundant message template use experience;
c. when the historical frequency of the merchant using the message template is greater than zero and does not exceed M, the merchant type is determined to be a third type, namely, a merchant between the first type and the second type, which has little experience with the use of the message template but is not a highly skilled merchant.
The value of M may be preset at the application server according to needs, for example, M is 5. By classifying the merchants, the message templates can be recommended according to different template recommendation indexes in different directions for different types of merchants.
S103, determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different.
The template recommendation indicator may include, for example, an attribute of the message template being higher than a preset value. The attributes of the message template include an actual click rate of the message template, an estimated click rate of the message template, a number of times of using the message template within a certain time range, a heat of the message template as a whole (the heat is positively correlated with a frequency of using the message template among the total number of templates, for example, the frequency of using the message template 1 is 1%, and the frequency of using the message template 2 is 2%, the heat of the message template 1 should be less than the heat of the message template 2), and the like.
The information template using experience of each type of merchant is different, the respective requirements are also different, and the application server can give different types of template recommendation indexes in advance based on different types of merchant types.
For example, for a first type of merchant of merchants who never use the template, the knowledge of the message template is less, and the application server does not have the usage data of the template message for the merchant, so that the merchant needs to be familiar with the usage of the message template and is simple and applicable as much as possible, and therefore the application server can guide the merchant to use the template message through widely applicable indexes. That is, the popularity may be determined as the template recommendation index corresponding to the first type, and based on the recommendation policy of popularity, a message template with a high frequency of use in a business in the same industry may be recommended, and at the same time, a suitable document may be attracted to the business for use, for example, document information is added to the message template with the popularity: "message templates that are used by 90% of the peers" for the merchant to make a more intuitive selection.
For another example, for a second type of merchant who uses a message template with a history frequency greater than M, the merchant is familiar with the template message and always uses the message to operate its own user, at this time, the estimated click rate may be a template recommendation indicator corresponding to the second type, where the estimated click rate is used to represent the total number of possible clicks of the message template by users (including potential users) of the merchant.
For another example, for a third type of merchant who uses a message template with a history frequency greater than zero and not greater than M, such a user has sent a message to its own user using the message template, but the related operations are not many, and there is no more experience, and at this time, the popularity and the estimated click rate may be used comprehensively as the template recommendation index corresponding to the third type. And simultaneously pushing the template to the third type of merchants according to the popularity and the estimated click rate so as to match the type of user to transit from the unfamiliar message template to the familiar message template.
And S105, obtaining the message template to be recommended meeting the template recommendation index, and pushing the message template to be recommended to the merchant. The application server side can conduct calling inquiry or real-time evaluation on the relevant indexes of each message template in real time. For example, the attributes such as the popularity or the actual click rate of the message template can be determined and stored in real time in advance, and can be called by the application server to query at any time.
For attributes such as the estimated click rate, the application server can obtain the merchant, the message template and the relevant characteristics of the user for estimation based on the pre-trained estimation model to obtain the estimated click rate of each message template. Therefore, once the merchant type is determined, the application server side can query from the storage device based on the template recommendation index to obtain the message template to be recommended, which meets the template recommendation index. The number of the acquired message templates to be recommended meeting the template recommendation index may be multiple.
The application server can push all the obtained message templates to be recommended to the merchant terminal, or select the top-ranked N message templates to recommend to the merchant terminal. For example, the number of the acquired message templates to be recommended meeting the template recommendation index may be 10, and the application server only selects the message template to be recommended from the top 3 for pushing. Or determining the message template to be recommended, which has more use intention of the merchant, from the message template to be recommended based on the merchant characteristics and the template characteristics, so as to improve the user experience of the merchant.
S107, evaluating the fitness of the pushed message template to the first type of merchants according to the activity degree of the user in the interfaces of the merchants of the applet.
It is emphasized that applets have an inherent advantage over conventional applications in that they are "available without downloading the corresponding application". For example, a merchant only needs to perform docking through a server, the server can provide a corresponding interface in an applet mode in an application, the merchant does not need to provide any program, and a user can enter the interface of the merchant through the applet in the application without downloading.
Thus, a first type of merchant in this application (i.e., a merchant who has not used a message template, typically a merchant who has a new resident applet) when promoted with a message template can expect that it pushes a new user (i.e., the user first enters the merchant's interface through the applet) that must also be the merchant. In other words, it is therefore possible to accurately perform user tracing, that is, consider that a new user always enters an interface of a new merchant from the message template (whereas if the new user is a regular application, the new user always needs to download the regular application before entering the interface, and it is difficult to accurately trace the source by downloading the regular application), and further, it is possible to evaluate the fitness of the pushed message template for the first type of merchant based on the activity level of the user in the interface of the merchant of the applet.
The level of user activity in the merchant's interface of the applet may include multidimensional activity parameters such as: the number of users entering the merchant's interface, the click-through rate in the merchant's interface, the number of transactions initiated in the merchant's interface, the proportion of transactions initiated, the time the user resides in the merchant's interface, and the like.
For the pushed message template, the fitness of the merchant of the first type may be evaluated by combining the foregoing multidimensional activity parameters, for example, after the foregoing multidimensional activity parameters are normalized in each dimension, weighting and summing up are performed; or, a pre-trained evaluation model is adopted, the multi-dimensional activity parameters and the characteristics of the message template are used as input characteristics, and the evaluation score and the like of the message template are output through the evaluation model.
The resulting fitness evaluated may be used to characterize whether the message template is suitable for a new merchant for which to promote a message. For example, some message templates may be more suitable for old users (i.e., users who have interacted with the merchant interface) and may be less attractive for new users, by which the message templates may be further evaluated hierarchically, and in total better assisting a new applet merchant in choosing a more suitable message template for promotion.
Determining the type of the merchant according to the historical frequency by determining the historical frequency of the merchant using the message template, wherein the type of the merchant comprises a first type with the historical frequency of the message template being zero; determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different; acquiring a message template to be recommended meeting the template recommendation index, pushing the message template to be recommended to the merchant, and evaluating the fitness of the pushed message template to the merchant of the first type according to the activity degree of a user in the interface of the merchant of the applet. Therefore, fitness evaluation of the message template is carried out on the merchant newly entering the applet, so that the message template more suitable for operation of the merchant is recommended, the merchant newly entering the applet is assisted to obtain a better message template, and self-operation efficiency of the newly entering merchant is improved.
In an embodiment, when recommending a message template to a merchant of a first type, pushing the message template by industry, specifically, determining an industry type to which the merchant belongs, and acquiring an industry message template corresponding to the industry type; and determining the use frequency of each industry message template in the industry type, and determining the industry message template with the front use frequency sequence as a message template to be recommended.
For example, the ranking of the number of times of use of the message template 1 in the whole industry is 1 st, but the ranking of the number of times of use in the "catering" industry is 5 th, at this time, when the merchant is the catering industry, if the message template 3 before the number of times of use ranking needs to be recommended to the merchant, the message template 1 may not be recommended any more, but the number of times of use of the message template in the catering industry is rearranged, so as to determine to obtain a more accurate industry ranking. Through the recommendation of the message templates in different industries, the message template to be recommended can better accord with the use rule of each industry, and more accurate template recommendation is realized.
In one embodiment, when recommending a message template to a first type of merchant, in order to obtain the message template to be recommended more accurately, the estimated click rate of each template may be determined as follows: determining a user set of a merchant to push a message; aiming at any message template, predicting the click probability of any user in the user set on the message template; determining the estimated click rate of the message template according to the click probability of the total number of users in the user set on the message template; and determining the message templates in the front row of the estimated click rate as the message templates to be recommended.
The set of users about to push messages may include users that the merchant has interacted with and potential users of the merchant (e.g., the potential users may be users that have interacted with other merchants in the merchant's co-business). At an application server, firstly, available message templates are obtained to be used as a set for traversing, for each message template, the click probability (the click probability is a numerical value which is larger than 0 and smaller than 1) of each user of the user set to the message template is determined, and the click probabilities of all users in the user set are summed to obtain the estimated click rate of the message template, so that the click probability of each user to each template is accurately estimated, and more accurate template recommendation is realized.
For the click probability of a single user to a single template, statistics can be performed based on the historical data of the user to obtain a statistical probability. For example, the message template pushes 10 messages to the user, but the user clicks only 3 times, and the click probability is 0.3. Or a pre-estimation model can be trained in advance, and the click probability of a single user for a single template can be obtained based on the user characteristics and the pre-estimation model.
In one embodiment, if the prediction model is used to predict the click probability of any user in the user set on a single message template, the following method may be used: acquiring application characteristics of an application adopted by the merchant in message pushing; aiming at any message template, determining the message template characteristics of the message template; determining user characteristics and cross characteristics of any user, wherein the cross characteristics at least comprise the usage characteristics of the user for the application and the usage characteristics of the user for the message template; fusing the application characteristic, the message template characteristic, the user characteristic and the cross characteristic to generate a mapping characteristic; and determining the click probability of the user on the message template by adopting a pre-trained pre-estimation model according to the mapping characteristics.
Specifically, it is first required to obtain features of various aspects, including:
(1) the application features of the application used in the message pushing process include: application name, name of the applet employed by the merchant hosting the application, description of the applet, category of the applet, industry involved, number of users, etc.
(2) Message template features of the message template, comprising: the industry to which the template belongs, the content of the template, the number of exposures to the template, the number of clicks, the click rate, etc.
(3) And user characteristics: including user attribute characteristics and behavioral characteristics such as city, age, gender, preferences, liveness, etc., as well as the number of times the user accesses the applet, the number of transactions, the number of times the user clicks on the message, etc.
(4) Cross-feature, which includes the usage feature of the application by the user, such as the usage frequency, usage duration, usage function, etc. of the application or applet by the user); and the cross feature also includes the use feature of the message template by the user, such as the number of clicks, the frequency of clicks, the browsing duration of the message template by the user, the frequency and duration of interaction with the merchant through the message template, and the like.
And further fusing the application characteristic, the message template characteristic, the user characteristic and the cross characteristic to generate a mapping characteristic, and determining the click probability of the user on the message template by adopting a pre-trained pre-estimation model according to the mapping characteristic. The specific fusion mode can be determined based on the training mode of the pre-estimated model, and only the fusion mode is consistent with the mode adopted in the training of the model.
In one embodiment, for the fusion of multiple features, a multi-model combination mode can be adopted. For example, the model is divided into textual features and non-textual features, wherein the textual features include template content in the message template, industry information, and the like, and the non-textual features include user features, applet features, cross-over features, and the like. And then extracting to obtain a first feature vector of the message template features by adopting a Text Convolutional network (TextCNN) model, and extracting to obtain a second feature vector of the application features, the user features and the cross features by adopting a Deep Neural network (Deep Neural network, DNN) model.
As shown in fig. 2, fig. 2 is a schematic diagram of feature fusion provided in the embodiment of the present disclosure. In this approach, the text content that is the basis of each feature may be obtained and input to the text convolution network model. In Textcnn, the original text content of the input is mapped to obtain a first feature vector using a pre-trained word vector as a mapping layer (i.e., embedding layer). In Textcnn, for all words in the data set, an embedding matrix is obtained because each word can be characterized as a vector, and each row in the embedding matrix is a word vector. This embedded matrix may be static, i.e. fixed. May be non-static, i.e. may be derived from back-propagation update training.
For the non-text features, encoding the non-text features (for example, one-hot encoding may be adopted), inputting the text features to an input layer of the DNN model, and convolving the output non-text features with a pre-trained hidden layer and an output layer to obtain a second feature vector.
Furthermore, as shown in fig. 2, the first feature vector and the second feature vector are merged in a splicing manner to generate a mapping feature, and the mapping feature comprehensively represents a non-text feature and a text feature, that is, the application feature, the message template feature, the user feature and the cross feature are integrally included, so that the pre-trained predictive model can be directly calculated based on the mapping feature to obtain the click probability of the user on the message template. That is, the click rate of a single user output by the model is estimated for a single message template, and for the single message template, assuming that the click rate of the ith single user is ctr _ i and the number of the user sets is count (user _ i), the average estimated click rate of the message template is ctr sum (ctr _ i)/count (user _ i)
In one embodiment, for some merchants, the merchant has had a relatively large experience with the use of message templates, which has been able to accurately understand the specific meaning of various types of template recommendation indicators for message templates. When the commercial tenant recommends the message template, an index evaluation value of the template recommendation index of the message template to be recommended can be obtained; and pushing the message template to be recommended and the index evaluation value to the merchant so as to display the index evaluation value at the merchant end. For example, the application server estimates that for a user group of the merchant a, the estimated click frequency of the message template 3 is 3 ten thousands of times based on the estimation model, and the average single estimated click rate is 0.2, so that the estimated click rate and the single estimated click rate can be pushed to the merchant together when the model is pushed, and the evaluation values of the indexes and the message template currently used by the merchant are displayed at the merchant simultaneously, so that the merchant can directly compare various message templates, thereby making a selection more in line with the needs of the merchant, and improving user experience. Fig. 3 is a schematic diagram of a message template display at a merchant end according to an embodiment of the present disclosure, as shown in fig. 3.
In an embodiment, for the application server, when the message template is pushed to the third type of merchant, the message template to be recommended may be further filtered based on the model evaluation index. Namely, determining the actual click rate of the message template currently used by the merchant; and determining the message template with the estimated click rate which is listed in the front row and the estimated click rate which is greater than the actual click rate as the message template to be recommended.
For example, when the application server detects that the actual click rate of the message template currently used by the merchant is 5 ten thousand times, and the estimated click rates of the 3 message templates arranged in the front row are respectively 6 ten thousand times, 4 ten thousand times and 3 ten thousand times, the application server can cancel the push of the two next message templates and only recommend the first 6 ten thousand message templates (i.e., the message templates exceeding the actual click rate by 5 ten thousand times) to adapt to the actual needs of the merchant.
In a second aspect, the present specification further provides a recommendation device for a message template. As shown in fig. 4, fig. 4 is a schematic structural diagram of a recommendation device for a message template provided in an embodiment of this specification, where the recommendation device for a message template includes:
the merchant type determining module 401 determines a historical frequency of the merchant using the message template, and determines the merchant type of the merchant according to the historical frequency, wherein the merchant type includes a first type in which the historical frequency of the merchant using the message template is zero;
a recommendation index determining module 403, configured to determine template recommendation indexes corresponding to the merchant types, where the template recommendation indexes corresponding to different merchant types are different;
an obtaining module 405, obtaining a message template to be recommended, which meets the template recommendation index;
the pushing module 407 is configured to push the message template to be recommended to the merchant;
the evaluation module 409 evaluates the fitness of the pushed message template to the first type of merchant according to the activity of the user in the interface of the merchant of the applet.
Optionally, wherein, when the merchant type determining module 401 determines that the merchant type is the first type; correspondingly, the recommendation index determining module 403 determines that the heat is the template recommendation index corresponding to the first type, where the heat is used to characterize the number of times of using the message template;
when the historical frequency of the merchant using the message template is greater than M, the merchant type determining module 401 determines that the merchant type is a second type; correspondingly, the recommendation index determining module 403 determines an estimated click rate as the template recommendation index corresponding to the second type, where the estimated click rate is used to represent the total number of times that the message template may be clicked by the user of the merchant;
when the historical frequency of the merchant using the message template is greater than zero and does not exceed M, the merchant type determining module 401 determines that the merchant type is a third type; correspondingly, the recommendation index determining module 403 determines that the popularity and the estimated click rate are the template recommendation indexes corresponding to the third type.
Optionally, when the type of the merchant is the first type, the obtaining module 405 determines the industry type to which the merchant belongs, and obtains an industry message template corresponding to the industry type; and determining the use frequency of each industry message template in the industry type, and determining the industry message template with the front use frequency sequence as a message template to be recommended.
Optionally, when the merchant type is the second type, the obtaining module 405 determines a user set for the merchant to push the message; aiming at any message template, predicting the click probability of any user in the user set on the message template; determining the estimated click rate of the message template according to the click probability of the total number of users in the user set on the message template; and determining the message templates in the front row of the estimated click rate as the message templates to be recommended.
Optionally, the obtaining module 405 determines an actual click rate of a message template currently used by the merchant; and determining the message template with the estimated click rate which is listed in the front row and the estimated click rate which is greater than the actual click rate as the message template to be recommended.
Optionally, the obtaining module 405 obtains application features of an application used by the merchant when pushing the message; aiming at any message template, determining the message template characteristics of the message template; determining user characteristics and cross characteristics of any user, wherein the cross characteristics at least comprise the usage characteristics of the user for the application and the usage characteristics of the user for the message template; fusing the application characteristic, the message template characteristic, the user characteristic and the cross characteristic to generate a mapping characteristic; and determining the click probability of the user on the message template by adopting a pre-trained pre-estimation model according to the mapping characteristics.
Optionally, the obtaining module 405 extracts a first feature vector of the message template features by using a text convolution network model; extracting by adopting a deep neural network model to obtain a second feature vector of the application feature, the user feature and the cross feature; and combining the first feature vector and the second feature vector to generate mapping features.
Optionally, the pushing module 407 obtains an index evaluation value of a template recommendation index of the message template to be recommended; and pushing the message template to be recommended and the index evaluation value to the merchant so as to display the index evaluation value at the merchant end.
In a third aspect, an embodiment of the present specification further provides an electronic device, as shown in fig. 5, fig. 5 is a schematic structural diagram of the electronic device provided in the embodiment of the present specification, where the electronic device includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect.
In a fourth aspect, based on the same idea, the present specification further provides a non-volatile computer storage medium corresponding to the method described above, and storing computer-executable instructions, which, when read by a computer, cause one or more processors to execute the method according to the first aspect.
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 manufacturing an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to the software compiler used in program development, but the original code before compiling is also written in a specific Programming Language, which is called Hardware Description Language (HDL), and the HDL is not only one kind but many kinds, such as abel (advanced boot Expression Language), ahdl (alternate Language Description Language), communication, CUPL (computer universal Programming Language), HDCal (Java Hardware Description Language), langa, Lola, mylar, HDL, PALASM, rhydl (runtime Description Language), vhjhdul (Hardware Description Language), and vhygl-Language, which are currently used commonly. 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: the ARC625D, 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, the present specification embodiments may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description 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 so forth) having computer-usable program code embodied therein.
The description has been presented with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. 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 magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that 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 embodiments of the apparatus, the device, and the nonvolatile computer storage medium, since they are substantially similar to the embodiments of the method, the description is simple, and for the relevant points, reference may be made to the partial description of the embodiments of the method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The above description is merely one or more embodiments of the present disclosure and is not intended to limit the present disclosure. Various modifications and alterations to one or more embodiments of the present description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of one or more embodiments of the present specification should be included in the scope of the claims of the present specification.
Claims (17)
1. A recommendation method of a message template is applied to an applet and comprises the following steps:
determining the historical frequency of a merchant using a message template, and determining the merchant type of the merchant according to the historical frequency, wherein the merchant type comprises a first type with the historical frequency of using the message template being zero;
determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different;
acquiring a message template to be recommended meeting the template recommendation index, and pushing the message template to be recommended to the merchant;
and evaluating the fitness of the pushed message template to the first type of merchants according to the activity degree of the user in the interfaces of the merchants of the applets.
2. The method of claim 1, wherein determining the merchant type of the merchant from historical frequency of merchant usage of message templates comprises:
when the merchant type is a first type; correspondingly, the determining the template recommendation index corresponding to the merchant type includes: determining the heat as a template recommendation index corresponding to the first type, wherein the heat is used for representing the using times of the message template;
when the historical frequency of the merchant using the message template is greater than M, determining that the merchant type is a second type; correspondingly, the determining the template recommendation index corresponding to the merchant type includes: determining an estimated click rate as a template recommendation index corresponding to the second type, wherein the estimated click rate is used for representing the total times of possible clicks of the message template by the user of the merchant;
when the historical frequency of the merchant using the message template is greater than zero and does not exceed M, determining that the merchant type is a third type; correspondingly, the determining the template recommendation index corresponding to the merchant type includes: and determining the popularity and the estimated click rate as template recommendation indexes corresponding to the third type.
3. The method of claim 2, wherein when the merchant type is a first type, obtaining a message template to be recommended that meets the template recommendation indicator comprises:
determining the industry type of the merchant, and acquiring an industry message template corresponding to the industry type;
and determining the use frequency of each industry message template in the industry type, and determining the industry message template with the front use frequency sequence as a message template to be recommended.
4. The method of claim 2, wherein when the merchant type is a second type, acquiring the message template to be recommended that meets the template recommendation index comprises:
determining a user set of a merchant to push a message;
aiming at any message template, predicting the click probability of any user in the user set on the message template;
determining the estimated click rate of the message template according to the click probability of the total number of users in the user set on the message template;
and determining the message templates in the front row of the estimated click rate as the message templates to be recommended.
5. The method of claim 4, wherein determining the message template with the pre-estimated click rate listed in the front row as the message template to be recommended comprises:
determining the actual click rate of the message template currently used by the merchant;
and determining the message template with the estimated click rate which is listed in the front row and the estimated click rate which is greater than the actual click rate as the message template to be recommended.
6. The method of claim 4, wherein predicting, for any message template, a click probability of any user in the set of users on the message template comprises:
acquiring application characteristics of an application adopted by the merchant in message pushing;
aiming at any message template, determining the message template characteristics of the message template;
determining user characteristics and cross characteristics of any user, wherein the cross characteristics at least comprise the usage characteristics of the user for the application and the usage characteristics of the user for the message template;
fusing the application characteristic, the message template characteristic, the user characteristic and the cross characteristic to generate a mapping characteristic;
and determining the click probability of the user on the message template by adopting a pre-trained pre-estimation model according to the mapping characteristics.
7. The method of claim 4, wherein fusing the application feature, the message template feature, the user feature, and the cross feature to generate a mapping feature comprises:
extracting a first feature vector of the message template features by adopting a text convolution network model;
extracting by adopting a deep neural network model to obtain a second feature vector of the application feature, the user feature and the cross feature;
and combining the first feature vector and the second feature vector to generate mapping features.
8. The method of claim 1, wherein pushing the message template to be recommended to the merchant comprises:
acquiring an index evaluation value of a template recommendation index of the message template to be recommended;
and pushing the message template to be recommended and the index evaluation value to the merchant so as to display the index evaluation value at the merchant end.
9. A recommendation device of a message template is applied to an applet and comprises the following components:
the merchant type determining module is used for determining the historical frequency of the merchant using the message template and determining the merchant type of the merchant according to the historical frequency, wherein the merchant type comprises a first type with zero historical frequency of using the message template;
the recommendation index determining module is used for determining template recommendation indexes corresponding to the merchant types, wherein the template recommendation indexes corresponding to different merchant types are different;
the acquisition module acquires a message template to be recommended, which meets the template recommendation index;
the pushing module is used for pushing the message template to be recommended to the merchant;
and the evaluation module is used for evaluating the fitness of the pushed message template to the first type of merchants according to the activity degree of the user in the interfaces of the merchants of the applet.
10. The apparatus according to claim 9, wherein when the merchant type determining module determines that the merchant type is a first type, correspondingly, the recommendation indicator determining module determines that the popularity is a template recommendation indicator corresponding to the first type, where the popularity is used to characterize the number of times of using the message template;
when the historical frequency of the commercial tenant using the message template is greater than M, the commercial tenant type determining module determines that the commercial tenant type is a second type; correspondingly, the recommendation index determining module determines that an estimated click rate is a template recommendation index corresponding to the second type, wherein the estimated click rate is used for representing the total number of times that the message template is possibly clicked by the user of the merchant;
when the historical frequency of the commercial tenant using the message template is greater than zero and does not exceed M, the commercial tenant type determining module determines that the commercial tenant type is a third type; correspondingly, the recommendation index determining module determines that the popularity and the estimated click rate are the template recommendation indexes corresponding to the third type.
11. The apparatus according to claim 10, wherein when the merchant type is a first type, the obtaining module determines an industry type to which the merchant belongs, and obtains an industry message template corresponding to the industry type; and determining the use frequency of each industry message template in the industry type, and determining the industry message template with the front use frequency sequence as a message template to be recommended.
12. The apparatus of claim 10, wherein when the merchant type is a second type, the obtaining module determines a set of users for which the merchant is to push messages; aiming at any message template, predicting the click probability of any user in the user set on the message template; determining the estimated click rate of the message template according to the click probability of the total number of users in the user set on the message template; and determining the message templates in the front row of the estimated click rate as the message templates to be recommended.
13. The apparatus of claim 12, wherein the obtaining module determines an actual click rate of a message template currently used by the merchant; and determining the message template with the estimated click rate which is listed in the front row and the estimated click rate which is greater than the actual click rate as the message template to be recommended.
14. The apparatus of claim 12, wherein the obtaining module obtains application features of an application employed by the merchant in message pushing; aiming at any message template, determining the message template characteristics of the message template; determining user characteristics and cross characteristics of any user, wherein the cross characteristics at least comprise the usage characteristics of the user for the application and the usage characteristics of the user for the message template; fusing the application characteristic, the message template characteristic, the user characteristic and the cross characteristic to generate a mapping characteristic; and determining the click probability of the user on the message template by adopting a pre-trained pre-estimation model according to the mapping characteristics.
15. The apparatus of claim 12, wherein the obtaining module extracts a first feature vector of the message template features by using a text convolution network model; extracting by adopting a deep neural network model to obtain a second feature vector of the application feature, the user feature and the cross feature; and combining the first feature vector and the second feature vector to generate mapping features.
16. The apparatus of claim 9, wherein the pushing module obtains an index evaluation value of a template recommendation index of the message template to be recommended; and pushing the message template to be recommended and the index evaluation value to the merchant so as to display the index evaluation value at the merchant end.
17. An electronic device, comprising:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210043871.6A CN114491246A (en) | 2022-01-14 | 2022-01-14 | Message template recommendation method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210043871.6A CN114491246A (en) | 2022-01-14 | 2022-01-14 | Message template recommendation method, device and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114491246A true CN114491246A (en) | 2022-05-13 |
Family
ID=81511924
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210043871.6A Pending CN114491246A (en) | 2022-01-14 | 2022-01-14 | Message template recommendation method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114491246A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110035272A1 (en) * | 2009-08-05 | 2011-02-10 | Yahoo! Inc. | Feature-value recommendations for advertisement campaign performance improvement |
| US20140344035A1 (en) * | 2013-05-17 | 2014-11-20 | Datasphere Technologies, Inc. | Managing content recommendations for customers |
| US20150154588A1 (en) * | 2011-08-18 | 2015-06-04 | Visa International Service Association | Reversed User Account Generation Apparatuses, Methods and Systems |
| US20150269606A1 (en) * | 2014-03-24 | 2015-09-24 | Datasphere Technologies, Inc. | Multi-source performance and exposure for analytics |
| CN111064655A (en) * | 2019-12-17 | 2020-04-24 | 北京每日优鲜电子商务有限公司 | Template message pushing method, device, equipment and storage medium |
| US10664876B1 (en) * | 2013-06-20 | 2020-05-26 | Groupon, Inc. | Method and apparatus for promotion template generation |
-
2022
- 2022-01-14 CN CN202210043871.6A patent/CN114491246A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110035272A1 (en) * | 2009-08-05 | 2011-02-10 | Yahoo! Inc. | Feature-value recommendations for advertisement campaign performance improvement |
| US20150154588A1 (en) * | 2011-08-18 | 2015-06-04 | Visa International Service Association | Reversed User Account Generation Apparatuses, Methods and Systems |
| US20140344035A1 (en) * | 2013-05-17 | 2014-11-20 | Datasphere Technologies, Inc. | Managing content recommendations for customers |
| US10664876B1 (en) * | 2013-06-20 | 2020-05-26 | Groupon, Inc. | Method and apparatus for promotion template generation |
| US20200302488A1 (en) * | 2013-06-20 | 2020-09-24 | Groupon, Inc. | Method and apparatus for promotion template generation |
| US20150269606A1 (en) * | 2014-03-24 | 2015-09-24 | Datasphere Technologies, Inc. | Multi-source performance and exposure for analytics |
| CN111064655A (en) * | 2019-12-17 | 2020-04-24 | 北京每日优鲜电子商务有限公司 | Template message pushing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070399B (en) | Coupon pushing method and device | |
| CN108733825B (en) | Object trigger event prediction method and device | |
| CN106295832B (en) | Product information pushing method and device | |
| CN113688313A (en) | Training method of prediction model, information pushing method and device | |
| CN110503206A (en) | A kind of prediction model update method, device, equipment and readable medium | |
| US11074634B2 (en) | Probabilistic item matching and searching | |
| US20220270121A1 (en) | Probabilistic search biasing and recommendations | |
| CN114202370A (en) | Information recommendation method and device | |
| CN112733024A (en) | Information recommendation method and device | |
| CN115455166A (en) | Method, device, medium and equipment for detecting abnormality of intelligent dialogue system | |
| CN114331602A (en) | Model training method based on transfer learning, information recommendation method and device | |
| CN110134860B (en) | User portrait generation method, device and equipment | |
| CN113343085B (en) | Information recommendation method and device, storage medium and electronic equipment | |
| CN117909840A (en) | Model training method and device, storage medium and electronic equipment | |
| CN113010562B (en) | Information recommendation method and device | |
| CN111046304A (en) | Data searching method and device | |
| JP2022514156A (en) | Probabilistic item matching and search | |
| CN114116816B (en) | Recommendation method and device | |
| CN113343132B (en) | Model training method, information display method and device | |
| CN114792256A (en) | Population expansion method and device based on model selection | |
| CN114491246A (en) | Message template recommendation method, device and equipment | |
| CN114861043A (en) | Model training and recommended position determining method and device | |
| CN113344590A (en) | Method and device for model training and complaint rate estimation | |
| CN118535715B (en) | Automatic reply method, equipment and storage medium based on tree structure knowledge base | |
| US12073457B1 (en) | Parameterized method, device, and user interface for enabling filtering |
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 |