CN109241455B - Recommended object display method and device - Google Patents
Recommended object display method and device Download PDFInfo
- Publication number
- CN109241455B CN109241455B CN201810989073.6A CN201810989073A CN109241455B CN 109241455 B CN109241455 B CN 109241455B CN 201810989073 A CN201810989073 A CN 201810989073A CN 109241455 B CN109241455 B CN 109241455B
- Authority
- CN
- China
- Prior art keywords
- type
- display
- class
- user
- keywords
- 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.)
- Active
Links
Images
Landscapes
- User Interface Of Digital Computer (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the disclosure provides a display method and a device of a recommended object, wherein the method comprises the following steps: acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords; generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; displaying the first type of object at a display position with the display type being a first type; and displaying the second type of object at a display position with the display type being a second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of networks, in particular to a method and a device for displaying a recommended object.
Background
For the technical field of networks, a user can input keywords in a specified area so as to retrieve related information. Wherein the related information comprises: object information and word information related to the keyword. The user can access the detail page of the object information to directly place a bill or perform other effective operations, and can also access a plurality of object information corresponding to the word information and determine a bill below the target object or perform other effective operations.
In the prior art, object information and word information are presented to a user in a separate manner. The object information is displayed at the front position, and the word information is displayed behind the object information. As shown in fig. 1, when the user inputs the letter "h", the acquisition of the related object information includes three: pirate shrimp meal (Tanjin shop), Coilia (Tanjin Baolio shop), Korea (Tanjin shop), Zhou black duck (Beijing Jia Fule Tanjin shop), word information includes six: hamburger, chafing dish, royal tea, Haolilai, Heli house, hamburger.
However, since the click rate of the front position is high, the fixed display sequence is not favorable for recommending words at the rear position, and is not favorable for ordering operation or other effective operation performed by the user through comparison of multiple objects, so that the ordering rate and the effective operation efficiency are low.
Disclosure of Invention
The embodiment of the disclosure provides a method and a device for displaying recommended objects, which are beneficial to improving the accuracy of sequencing.
According to a first aspect of the embodiments of the present disclosure, a method for presenting a recommended object is provided, where the method includes:
acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords;
generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user;
displaying the first type of object at a display position with the display type being a first type;
and displaying the second type of object at a display position with the display type being a second type.
According to a second aspect of the embodiments of the present disclosure, there is provided a presentation apparatus of a recommended object, the apparatus including:
the recall module is used for acquiring keywords input by a user and recalling the first class object and the second class object respectively according to the keywords;
the display type sequence generation module is used for generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user;
the first display module is used for displaying the first type of object at a display position with the display type being a first type;
and the second display module is used for displaying the second type of object at the display position with the display type being the second type.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:
the recommendation system comprises a processor, a memory and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the recommendation object presentation method.
According to a fourth aspect of the embodiments of the present disclosure, there is provided a readable storage medium, wherein instructions of the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the aforementioned presentation method of a recommended object.
The embodiment of the disclosure provides a display method and a device of a recommended object, wherein the method comprises the following steps: acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords; generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; displaying the first type of object at a display position with the display type being a first type; and displaying the second type of object at a display position with the display type being a second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed to be used in the description of the embodiments of the present disclosure will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.
FIG. 1 is a schematic diagram of a prior art display of recommended words and recommended objects;
fig. 2 is a flowchart illustrating specific steps of a method for displaying a recommended object according to an embodiment of the present disclosure;
fig. 3 is a flowchart illustrating specific steps of a method for presenting a recommended object according to a second embodiment of the present disclosure;
fig. 4 is a structural diagram of a display device of a recommended object according to a third embodiment of the disclosure;
fig. 5 is a structural diagram of a display apparatus for recommending an object according to a fourth embodiment of the disclosure;
fig. 6 is a block diagram of an electronic device provided in an embodiment of the present disclosure.
Detailed Description
The technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some, but not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Example one
Referring to fig. 2, a flowchart illustrating specific steps of a presentation method for a recommended object according to an embodiment of the present disclosure is shown.
The method and the device for searching the keywords are suitable for being recommended to the user when the user inputs the keywords in the designated area for searching and associated information is associated according to part of the keywords input by the user. For example, when the user inputs "electricity", the user may associate "electricity business peak meeting", "computer host", "movie", etc. Thereby saving user input time.
Among these, keywords include, but are not limited to: characters, letters, etc.
Specifically, the first class object and the second class object may be recalled based on relevance to the keyword. The first class of objects are effective objects which can be directly used for login and order placement, the second class of objects are words associated with keywords, and a plurality of effective objects which are associated with the words and can be used for login and order placement are obtained after a user clicks the words. Namely: the lower nodes of the second class of objects are the first class of objects.
Optionally, in another embodiment of the present disclosure, the first type object is a merchant, and the second type object is a good or service provided by the merchant.
It can be understood that, in the embodiment of the invention disclosed in the present disclosure, for the takeout platform, the first type object is a takeout merchant, the second type object is a takeout and takeout delivery service, and when a user clicks the takeout commodity, a plurality of merchants providing the takeout commodity can be obtained, and an order is placed at the merchants.
And 102, generating a preset display type corresponding to at least one display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user.
The feature information of the current user includes, but is not limited to: the current user's click-through rate for the first type and the user's click-through rate for the second type represent the current user's preferences. The feature information of the current user can be periodically counted and stored, so that the latest feature information can be obtained according to the current user identification.
The characteristic information of the keyword includes but is not limited to: the click-through rates for the first type and the second type of all users searching for the keyword represent the preferences of all users. The feature information of all users can be periodically counted and stored, so that the latest feature information can be obtained according to the keywords.
The characteristic information of the keywords reflects the preference of all users of the platform on the display types, and the characteristic information of the current user reflects the preference of the user on the display types. In practical applications, the feature information may be described in terms of clicks, orders, etc. for various presentation types.
The display positions are display positions for recommending the first and second types of objects to the user, and can be determined by setting the number of the display positions. For example, for a take-away platform, the number of display locations is typically 10. It can be understood that, in practical applications, the number of the display positions may be set according to practical application scenarios, and the embodiment of the present disclosure does not limit the number.
The display types comprise a first type and a second type, the first type is used for displaying the first type of objects, and the second type is used for displaying the second type of objects. In practical application, the first type and the second type may be presented alternately or successively. The display type sequence generated by the display type sequence generation model enables the display effect to be the best, namely the total click rate or the conversion rate of the displayed first type object and the displayed second type object is the highest.
The type sequence generation model can predict and obtain the display type sequence according to the preference of the current user, the preference of all users and other information. It will be appreciated that the length of the presentation type sequence is related to the preset number of presentation positions. The user's preference may be expressed by the user's click-through rate for objects of the first type or objects of the second type, i.e.: if the click rate of the current user to the first type of object is higher, and the click rate of all the users to the first type of object is higher, displaying the object at the front display position; and if the click rate of the current user to the first type of object is lower and the click rate of all the users to the first type of object is lower, displaying the object at the later display position.
It is understood that in practical applications, other information affecting the click rate may also be used as input information of the type sequence generation model to determine the presentation type of the presentation position.
Specifically, first, the first class objects are sorted; and then displaying a plurality of search results ranked at the top in the first type of display positions until the last display position.
It can be understood that when the number of the first type objects is larger than the number of the first type display positions, a part of the first type objects are intercepted and displayed, and the rest of the first type objects are not displayed; when the number of the first type objects is less than the number of the first type display positions, displaying all the first type objects, and remaining free display positions.
And 104, displaying the second type of object at a display position with the display type being the second type.
Specifically, the second class objects are sorted first; and then displaying a plurality of search results ranked at the top in the display positions of the second type until the last display position.
It can be understood that when the number of the second class objects is greater than the number of the second type of display positions, a part of the second class objects are intercepted for display, and the rest of the second class objects are not displayed; and when the number of the second class objects is less than that of the second type display positions, displaying all the second class objects and leaving free display positions.
To sum up, the embodiment of the present disclosure provides a method for displaying a recommended object, where the method includes: acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords; generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; displaying the first type of object at a display position with the display type being a first type; and displaying the second type of object at a display position with the display type being a second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
Example two
Referring to fig. 3, a flowchart illustrating specific steps of a method for presenting a recommended object according to a second embodiment of the present disclosure is shown.
The user behavior record comprises behavior records of all users accessing the application platform, and can be stored on an internal or external storage device of the application platform.
In practical applications, the user behavior record includes, but is not limited to, a record of user access to the first class of objects, and a record of user access to the second class of objects. It can be understood that when a user logs in or otherwise accesses the application platform, the platform records the access behavior of the user; when a user searches related objects by inputting keywords, the keywords input by the user and clicked objects can be recorded; when the user performs order placing operation or other effective operation on one of the objects, the order placing operation and the effective operation can be recorded.
In the embodiment of the disclosure, a clicked recommendation object may be identified, and a ordering operation or other validation operation for the object may be acquired.
In practical application, when the effective identification is recorded, the object information is recorded at the same time. For example, the validated flag may be 1, and thus the non-validated flag may be 0. It is understood that the valid identifier and the invalid identifier may be represented by a character string, a character, or other manners, which are not limited in the embodiments of the present disclosure.
The feature information corresponding to the first class object may be all information affecting the generating efficiency of the first class object, including the intrinsic feature of the configuration and the real-time feature. The characteristic information may be stored on an external or internal storage device of the application platform. When the first type of object is registered on the application platform, the inherent characteristics can be configured, the real-time characteristics are changed along with the access of the user, and the latest real-time characteristics can be obtained through periodic statistics. For example, for a take-away platform, intrinsic characteristics include, but are not limited to, store level, store location, type of sale, etc., and real-time characteristics may include, but are not limited to: a score of the store, sales of the store, rate of placing orders, etc.
Step 203, training a sample set composed of the feature information and the effective identifier to obtain a first ranking model, wherein the effective identifier corresponds to a positive sample, and the ineffective identifier corresponds to a negative sample.
In the embodiment of the present disclosure, the XGBOOST model may be used to train to obtain the first ordering model, that is: and fitting the relationship between the characteristic information and the effectiveness rate of the first type of objects, so that the first type of objects can be sorted according to the production efficiency.
Step 204, obtaining the user behavior record, and extracting the access record aiming at the second class object from the user behavior record.
The user behavior record may refer to the detailed description of step 201, which is not described herein again.
Step 205, extracting a second class object and an effective identifier from the access record, and counting feature information corresponding to the second class object, where the effective identifier includes an effective identifier and an ineffective identifier.
In practical application, when the effective identifier is recorded, the second class object is recorded at the same time. If the second kind of object is entered by the keyword, the keyword is recorded at the same time.
The feature information corresponding to the second class object may be all information affecting the generation efficiency of the second class object, and the feature information of the second class object is a real-time feature and may be stored on an external or internal storage device of the application platform. The real-time characteristics of the second class of objects are changed along with the access of the user, and the latest real-time characteristics can be periodically counted. For example, for a take-away platform, the real-time characteristics of the second class of objects may include, but are not limited to: click rate, conversion rate and the like of the user on all the first-class objects associated with the second-class object.
And 206, training a sample set consisting of the feature information and the effective identification to obtain a second ranking model, wherein the effective identification corresponds to a positive sample, and the ineffective identification corresponds to a negative sample.
In the embodiment of the present disclosure, an XGBOOST model may be used to train to obtain a second ranking model, that is: and fitting the relationship between the characteristic information and the effectiveness rate of the second class of objects, so that the second class of objects can be sorted according to the production efficiency.
Step 207, obtaining a user behavior record, and extracting an access record of the first class object or the second class object from the user behavior record, where the access record includes: the display method comprises user identification information, a display type sequence where the first type object or the second type object is located, and keywords corresponding to the display type sequence.
The user behavior record can refer to the detailed description of step 201, and is not described in detail here.
In practical applications, when a user clicks a recommended first-class object or a recommended second-class object, a recommended sequence in which the first-class object or the second-class object is located may be recorded. For example, for one-time recommendation sequences (D1, D2, C1, D3, C2, C3, D4, C4, D5, and C5), wherein D1, D2, D3, D4, and D5 are objects of a first class, and C1, C2, C3, C4, and C5 are objects of a second class, if a user clicks D2, a recommendation type sequence (1, 1, 0, 1, 0, 0, 1, 0, 1, 0) is recorded while recording D2, wherein 1 represents the first type, and 0 represents the second type.
The user identification information may be a login account or other identification information when the user logs in the application platform.
And 208, acquiring user characteristic information corresponding to the user identification information and characteristic information corresponding to the keyword.
The user characteristic information may be all user characteristic information affecting the sequence click rate or the conversion rate. In practical application, the setting can be set according to practical application scenes. For example, for a take-away platform, the user characteristic information includes a user's click-through rate for a first class of objects and a click-through rate for a second class of objects. It can be understood that the user characteristic information is a real-time characteristic, and can be periodically counted and stored.
And 209, training through a sample set consisting of the user characteristic information, the characteristic information corresponding to the keywords and the display type sequence to obtain a type sequence generation model.
The disclosed embodiments may be trained through reinforcement learning. Reinforcement learning is unsupervised learning and does not require sample labeling.
In reinforcement learning, the return function can be determined according to the following rules, for non-click, the return is-1, the number of click is reported when clicking, and the amount of transaction is reported when ordering.
Step 210, obtaining keywords input by a user, and recalling the first class object and the second class object respectively according to the keywords.
This step can refer to the detailed description of step 101, and is not described herein again.
And step 211, generating a preset display type corresponding to at least one display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user.
This step can refer to the detailed description of step 102, and is not described herein again.
It can be understood that the input of the type sequence generation model is the feature information of the current user and the feature information of the keyword, and the output is a display type sequence with a preset length.
And 212, sequencing the first class of objects by adopting a first sequencing model obtained by pre-training to obtain a first sequence.
The first-class objects can be sorted by adopting a first sorting model, the input of the model is the characteristic information of each first-class object, and the output of the model is the sorting order of the first-class objects. Wherein, the characteristic information is all characteristics which can influence the first class object to be clicked. For example, for a take-away platform, if the object is a store, the characteristic information may be a store sales amount, a store score, and the like. It can be understood that the higher the sales of stores, the higher the score of stores, and the higher the ranking; the lower the store sales, the lower the store score, and the further back in the ranking.
In practical application, the first ranking model can be obtained through training of a large number of samples, so that various characteristic information and ranking knowledge information of the first class of objects are learned to guide object ranking.
And step 213, sequentially and one by one, displaying the first type of objects in the first sequence at the display positions of the first type.
In practical applications, if the first type of presentation location is still free, the location is deleted.
And 214, sequencing the second class of objects by adopting a second sequencing model obtained by pre-training to obtain a second sequence.
The second sort model may be used to sort the second class objects, where the input of the model is the feature information of each second class object, and the output is the sort order of the second class objects. Wherein, the characteristic information is all characteristics which can influence the second class object to be clicked. For example, the click rate of the second type object, the number of stores corresponding to the second type object, etc. It can be understood that the higher the click rate is, the more the number of corresponding stores is, and the higher the ranking is; the lower the click rate, the fewer the number of stores, and the later the ranking.
In practical application, the second ranking model can be obtained through training of a large number of samples, so that various feature information and ranking knowledge information of the second class of objects are learned to guide the ranking of the second class of objects.
It is to be understood that the feature information of the first class object and the feature information of the second class object may be set according to practical applications, and the embodiment of the present disclosure does not limit them.
And step 215, sequentially and one by one, displaying the second type objects in the second sequence at the display positions of the second type.
In practical applications, if the second type of presentation location is still free, the location is deleted.
To sum up, the embodiment of the present disclosure provides a method for displaying a recommended object, where the method includes: acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords; generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; displaying the first type of object at a display position with the display type being a first type; and displaying the second type of object at a display position with the display type being a second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
EXAMPLE III
Referring to fig. 4, a structural diagram of a display apparatus of a recommended object according to a third embodiment of the present disclosure is shown, which is as follows.
The recall module 301 is configured to obtain a keyword input by a user, and recall the first class object and the second class object according to the keyword.
Optionally, in another embodiment of the present disclosure, the first type object is a merchant, and the second type object is a good or service provided by the merchant.
And a display type sequence generating module 302, configured to generate a display type corresponding to at least one preset display position through a type sequence generating model obtained through pre-training according to the keyword and the feature information of the current user.
A first displaying module 303, configured to display the first type object at a display position where the display type is a first type;
a second showing module 304, configured to show the second type object in a showing position where the showing type is the second type.
To sum up, the embodiment of the present disclosure provides a display device for a recommended object, the device including: the recall module is used for acquiring keywords input by a user and recalling the first class object and the second class object respectively according to the keywords; the display type sequence generation module is used for generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; the first display module is used for displaying the first type of object at a display position with the display type being a first type; and the second display module is used for displaying the second type of object at the display position with the display type being the second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
The third embodiment is a corresponding apparatus embodiment to the first embodiment, and details thereof are not repeated herein.
Example four
Referring to fig. 5, a structural diagram of a display apparatus of a recommended object according to a fourth embodiment of the disclosure is shown, which is as follows.
The first access record obtaining module 401 is configured to obtain a user behavior record, and extract an access record for a first class of objects from the user behavior record.
A first information obtaining module 402, configured to extract a first class object and an effective identifier from the access record, and obtain feature information corresponding to the first class object, where the effective identifier includes an effective identifier and an ineffective identifier.
A first ordering model training module 403, configured to obtain a first ordering model through training of a sample set composed of the feature information and an effective identifier, where the effective identifier corresponds to a positive sample, and the non-effective identifier corresponds to a negative sample.
A second access record obtaining module 404, configured to obtain a user behavior record, and extract an access record for the second class object from the user behavior record.
A second information obtaining module 405, configured to extract a second class object and an effective identifier from the access record, and count feature information corresponding to the second class object, where the effective identifier includes an effective identifier and an ineffective identifier.
A second ranking model training module 406, configured to obtain a second ranking model through training of a sample set composed of the feature information and an effective identifier, where the effective identifier corresponds to a positive sample, and the ineffective identifier corresponds to a negative sample.
The comprehensive access record obtaining module 407 is configured to obtain a user behavior record, and extract an access record of a first class object or a second class object from the user behavior record, where the access record includes: the display method comprises user identification information, a display type sequence where the first type object or the second type object is located, and keywords corresponding to the display type sequence.
The feature information obtaining module 408 is configured to obtain user feature information corresponding to the user identification information and feature information corresponding to the keyword.
And the type sequence generation model training module 409 is used for obtaining a type sequence generation model through training of a sample set consisting of the user characteristic information, the characteristic information corresponding to the keyword and the display type sequence.
The recall module 410 is configured to obtain a keyword input by a user, and recall the first class object and the second class object according to the keyword.
And the display type sequence generating module 411 is configured to generate a display type corresponding to at least one preset display position through a type sequence generating model obtained through pre-training according to the keyword and the feature information of the current user.
A first display module 412, configured to display the first type object at a display position where the display type is a first type; optionally, in another embodiment of the present disclosure, the first display module 412 includes:
the first ordering submodule 4121 is configured to order the first class of objects by using a first ordering model obtained through pre-training, so as to obtain a first sequence.
The first displaying sub-module 4122 is configured to sequentially display the first type objects in the first sequence at the first type display positions one by one.
A second showing module 413, configured to show the second type object in a showing position where the showing type is the second type. Optionally, in another embodiment of the present disclosure, the second display module 413 includes:
the second sorting submodule 4131 is configured to sort the second class of objects by using a second sorting model obtained through pre-training, so as to obtain a second sequence.
The second displaying sub-module 4132 is configured to sequentially display the second type objects in the second sequence at the displaying positions of the second type one by one.
To sum up, the embodiment of the present disclosure provides a display device for a recommended object, the device including: the recall module is used for acquiring keywords input by a user and recalling the first class object and the second class object respectively according to the keywords; the display type sequence generation module is used for generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the characteristic information of the current user; the first display module is used for displaying the first type of object at a display position with the display type being a first type; and the second display module is used for displaying the second type of object at the display position with the display type being the second type. The display type of each display position can be flexibly determined, and the ordering rate or other effective operation efficiency can be improved.
The fourth embodiment is a device embodiment corresponding to the second embodiment, and the detailed description may refer to the second embodiment, which is not repeated herein.
An embodiment of the present disclosure further provides an electronic device, with reference to fig. 6, including: a process 501, a memory 502 and a computer program 5021 stored on the memory 502 and operable on the processor 501, the processor 501 implementing the aforementioned presentation method of the recommended object when executing the program.
The embodiment of the present disclosure also provides a readable storage medium, and when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the aforementioned presentation method for a recommended object.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, this disclosure is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the present disclosure as described herein, and any descriptions above of specific languages are provided for disclosure of enablement and best mode of the present disclosure.
In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the disclosure may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the disclosure, various features of the disclosure are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, the claimed disclosure requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this disclosure.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Various component embodiments of the disclosure may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in a sequenced display device in accordance with embodiments of the present disclosure. The present disclosure may also be embodied as an apparatus or device program for performing a portion or all of the methods described herein. Such programs implementing the present disclosure may be stored on a computer-readable medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the disclosure, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The disclosure may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
The above description is only for the specific embodiments of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present disclosure, and all the changes or substitutions should be covered within the scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (10)
1. A presentation method of a recommended object, the method comprising:
acquiring keywords input by a user, and recalling a first class object and a second class object respectively according to the keywords;
wherein, the lower node of the second class object is the first class object;
generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the feature information of the current user, wherein the feature information of the keywords reflects the preference of all users to the display type, and the feature information of the current user reflects the preference of the current user to the display type;
displaying the first type of object at a display position with the display type being a first type by adopting a first sequencing model obtained by pre-training;
displaying the second type of object at a display position with the display type being a second type by adopting a second sequencing model obtained by pre-training;
wherein the display positions with the display type of the first type and the display positions with the display type of the second type are alternately arranged.
2. The method according to claim 1, wherein the step of presenting the first type of object at the presentation position with the presentation type being the first type by using the pre-trained first ordering model comprises:
sequencing the first class of objects by adopting a first sequencing model obtained by pre-training to obtain a first sequence;
and displaying the first-class objects in the first sequence at the display positions of the first type one by one according to the sequence.
3. The method according to claim 1, wherein the step of presenting the second type object at the presentation position with the presentation type being the second type by using the second ranking model obtained by pre-training comprises:
sequencing the second class of objects by adopting a second sequencing model obtained by the pre-training to obtain a second sequence;
and displaying the second type objects in the second sequence at the display positions of the second type one by one according to the sequence.
4. The method of claim 2, wherein the first order model is trained by:
acquiring a user behavior record, and extracting an access record aiming at a first class of objects from the user behavior record;
extracting a first class object and an effective identifier from the access record, and acquiring characteristic information corresponding to the first class object, wherein the effective identifier comprises an effective identifier and an ineffective identifier;
and training a sample set consisting of the characteristic information and the effective identification to obtain a first sequencing model, wherein the effective identification corresponds to a positive sample, and the ineffective identification corresponds to a negative sample.
5. The method of claim 3, wherein the second order model is trained by:
acquiring a user behavior record, and extracting an access record aiming at a second class of objects from the user behavior record;
extracting a second class object and an effective identifier from the access record, and counting characteristic information corresponding to the second class object, wherein the effective identifier comprises an effective identifier and an ineffective identifier;
and training a sample set consisting of the characteristic information and the effective identification to obtain a second sequencing model, wherein the effective identification corresponds to the positive sample, and the ineffective identification corresponds to the negative sample.
6. The method of claim 1, wherein the type sequence generation model is trained by the following steps:
acquiring a user behavior record, and extracting an access record of a first class object or a second class object from the user behavior record, wherein the access record comprises: user identification information, a display type sequence where the first type object or the second type object is located, and keywords corresponding to the display type sequence;
acquiring user characteristic information corresponding to the user identification information and characteristic information corresponding to the keywords;
and training a sample set consisting of the user characteristic information, the characteristic information corresponding to the keywords and the display type sequence to obtain a type sequence generation model.
7. The method of claim 1, wherein the first type of object is a merchant and the second type of object is a good or service provided by the merchant.
8. A presentation device for a recommended object, the device comprising:
the recall module is used for acquiring keywords input by a user and recalling the first class object and the second class object respectively according to the keywords;
wherein, the lower node of the second class object is the first class object;
the display type sequence generation module is used for generating a display type corresponding to at least one preset display position through a type sequence generation model obtained through pre-training according to the keywords and the feature information of the current user, wherein the feature information of the keywords reflects the preference of all users to the display type, and the feature information of the current user reflects the preference of the current user to the display type;
the first display module is used for displaying the first type of objects at a display position with the display type being a first type by adopting a first sequencing model obtained by pre-training;
the second display module is used for displaying the second type of object at a display position with the display type being a second type by adopting a second sequencing model obtained by pre-training;
wherein the display positions with the display type of the first type and the display positions with the display type of the second type are alternately arranged.
9. An electronic device, comprising:
processor, memory and computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the program, implements the presentation method of a recommended object according to one or more of claims 1 to 7.
10. A readable storage medium, characterized in that instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform a presentation method of a recommended object according to one or more of method claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810989073.6A CN109241455B (en) | 2018-08-28 | 2018-08-28 | Recommended object display method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810989073.6A CN109241455B (en) | 2018-08-28 | 2018-08-28 | Recommended object display method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109241455A CN109241455A (en) | 2019-01-18 |
CN109241455B true CN109241455B (en) | 2021-08-20 |
Family
ID=65068826
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810989073.6A Active CN109241455B (en) | 2018-08-28 | 2018-08-28 | Recommended object display method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109241455B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110083688B (en) * | 2019-05-10 | 2022-03-25 | 北京百度网讯科技有限公司 | Search result recall method, device, server and storage medium |
CN110275779B (en) * | 2019-06-20 | 2022-07-08 | 北京百度网讯科技有限公司 | Resource acquisition method, device, equipment and storage medium |
CN110415032A (en) * | 2019-07-24 | 2019-11-05 | 深圳乐信软件技术有限公司 | A kind of exposure conversion ratio predictor method, device, equipment and storage medium |
CN113360777B (en) * | 2021-08-06 | 2021-12-07 | 北京达佳互联信息技术有限公司 | Content recommendation model training method, content recommendation method and related equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130325855A1 (en) * | 2012-06-04 | 2013-12-05 | Sense Networks, Inc | Venue Prediction Based on Ranking |
CN104750713A (en) * | 2013-12-27 | 2015-07-01 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
CN105224959A (en) * | 2015-11-02 | 2016-01-06 | 北京奇艺世纪科技有限公司 | The training method of order models and device |
CN105760400A (en) * | 2014-12-19 | 2016-07-13 | 阿里巴巴集团控股有限公司 | Method and device for ranking push messages based on search behavior |
CN106055688A (en) * | 2016-06-07 | 2016-10-26 | 北京金山安全软件有限公司 | Search result display method and device and mobile terminal |
CN108334536A (en) * | 2017-11-30 | 2018-07-27 | 中国电子科技集团公司电子科学研究院 | A kind of information recommendation method, equipment and storage medium |
-
2018
- 2018-08-28 CN CN201810989073.6A patent/CN109241455B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130325855A1 (en) * | 2012-06-04 | 2013-12-05 | Sense Networks, Inc | Venue Prediction Based on Ranking |
CN104750713A (en) * | 2013-12-27 | 2015-07-01 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
CN105760400A (en) * | 2014-12-19 | 2016-07-13 | 阿里巴巴集团控股有限公司 | Method and device for ranking push messages based on search behavior |
CN105224959A (en) * | 2015-11-02 | 2016-01-06 | 北京奇艺世纪科技有限公司 | The training method of order models and device |
CN106055688A (en) * | 2016-06-07 | 2016-10-26 | 北京金山安全软件有限公司 | Search result display method and device and mobile terminal |
CN108334536A (en) * | 2017-11-30 | 2018-07-27 | 中国电子科技集团公司电子科学研究院 | A kind of information recommendation method, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109241455A (en) | 2019-01-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109241455B (en) | Recommended object display method and device | |
US9576251B2 (en) | Method and system for processing web activity data | |
US8515937B1 (en) | Automated identification and assessment of keywords capable of driving traffic to particular sites | |
JP5693746B2 (en) | Product information ranking | |
JP5542812B2 (en) | Query identification and association | |
JP6517818B2 (en) | Improving Website Traffic Optimization | |
CN104699725B (en) | data search processing method and system | |
US9589277B2 (en) | Search service advertisement selection | |
JP2016505993A (en) | Method and apparatus for configuring search terms, delivering advertisements, and retrieving product information | |
CN105447730A (en) | Target user orientation method and device | |
US20210118024A1 (en) | Multi-label product categorization | |
CN108319376B (en) | Input association recommendation method and device for optimizing commercial word promotion | |
CN106445963A (en) | Advertisement index keyword automatic generation method and apparatus for APP platform | |
US20200074509A1 (en) | Business data promotion method, device, terminal and computer-readable storage medium | |
US11941073B2 (en) | Generating and implementing keyword clusters | |
CN110781251A (en) | Insurance knowledge map generation method, device, equipment and storage medium | |
CN106484698A (en) | A kind of method for pushing of search keyword and device | |
CN106557556A (en) | A kind of methods of exhibiting of Webpage, device, server and system | |
CN105786910B (en) | Entry weighing computation method and device | |
CN110363570A (en) | Classification methods of exhibiting, device, electronic equipment and storage medium in | |
CN110570271A (en) | information recommendation method and device, electronic equipment and readable storage medium | |
CN106383857A (en) | Information processing method and electronic equipment | |
WO2006062421A2 (en) | Information display method and system | |
CN116823410A (en) | Data processing method, object processing method, recommending method and computing device | |
KR102082900B1 (en) | System for providing optimal keyword of sale items |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |