CN113641892A - Information recommendation method and device - Google Patents
Information recommendation method and device Download PDFInfo
- Publication number
- CN113641892A CN113641892A CN202110793233.1A CN202110793233A CN113641892A CN 113641892 A CN113641892 A CN 113641892A CN 202110793233 A CN202110793233 A CN 202110793233A CN 113641892 A CN113641892 A CN 113641892A
- Authority
- CN
- China
- Prior art keywords
- information
- service
- recommendation information
- recommendation
- feature vector
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The specification discloses an information recommendation method and device, which can respond to a service request of a target user for a set service type, determine candidate recommendation information under the set service type and historical recommendation information which has a service execution relation with the target user historically, determine a feature vector corresponding to the historical recommendation information according to a pre-constructed service topological graph, determine a feature vector corresponding to the candidate recommendation information for each candidate recommendation information, wherein the service topological graph is used for representing the service relation between the recommendation information under each service type and each user, input the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model, can obtain a recommendation score corresponding to the candidate recommendation information, and finally, according to the recommendation score corresponding to each candidate recommendation information, and information recommendation is performed on the target user, so that the accuracy of information recommendation is improved.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for information recommendation.
Background
With the continuous development of information technology, a service platform can provide various services for users, so that the users can conveniently find the services which the users want to execute, and the service platform can recommend some recommendation information which the users may be interested in to the users according to the preferences of the users.
In the prior art, for one service type, a service platform may recommend recommendation information under the service type to a user according to historical preference of the user in the service under the service type, but the method does not consider the preference of the user under other service types, and the method of performing information recommendation only according to the historical preference of the user under the service type may be relatively inaccurate, and the recommendation information under other service types is generally different from attribute information of the recommendation information under the service type, and is difficult to be introduced into information recommendation service corresponding to the service type.
Therefore, how to improve the accuracy of information recommendation to the user is an urgent problem to be solved.
Disclosure of Invention
The present specification provides an information recommendation method and apparatus, which partially solve the above problems in the prior art.
The technical scheme adopted by the specification is as follows:
the present specification provides a method for information recommendation, including:
responding to a service request of a target user for a set service type, and determining candidate recommendation information corresponding to the target user under the set service type and historical recommendation information which has a service execution relation with the target user historically, wherein the historical recommendation information comprises recommendation information of different service types;
determining a feature vector corresponding to each historical recommendation information according to a pre-constructed service topological graph, and determining a feature vector corresponding to each candidate recommendation information aiming at each candidate recommendation information, wherein the service topological graph is used for representing service relationships between the recommendation information under each service type and each user;
inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model to obtain a recommendation score corresponding to the candidate recommendation information;
and recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
Optionally, the constructing the service topology map specifically includes:
the method comprises the steps of constructing the service topological graph according to integrated service information, wherein the integrated service information comprises at least one of historical service records of service execution of each user aiming at each recommended information, a geographical area corresponding to each recommended information and an information name corresponding to each recommended information, the types of edges between nodes in the service topological graph constructed through different types of integrated service information are different, and each recommended information comprises recommended information under each service type.
Optionally, the service topology map includes an information node, where the information node is used to represent recommendation information;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
determining a service execution sequence of each user aiming at each recommended information according to the historical service record;
and according to the service execution sequence of each user aiming at each recommended information, establishing directed edges among the information nodes so as to establish the service topological graph.
Optionally, the service topology map includes an information node and a region node, where the information node is used to represent recommendation information, and the region node is used to represent a geographic region;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
and aiming at each piece of recommendation information, according to the geographical area corresponding to the recommendation information, constructing edges between the information node corresponding to the recommendation information and the area node of the geographical area corresponding to the recommendation information so as to construct the service topological graph.
Optionally, the service topology map includes an information node, where the information node is used to represent recommendation information;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
and according to the information names of the pieces of recommended information, constructing edges among the information nodes of the pieces of recommended information with similar information names so as to construct the service topological graph.
Optionally, the recommendation model includes a feature extraction layer and a prediction layer;
inputting the feature vector corresponding to the candidate recommendation information and the feature vectors corresponding to the historical recommendation information into a pre-trained recommendation model to obtain recommendation scores corresponding to the candidate recommendation information, and specifically comprising:
inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into the feature extraction layer, so that the feature extraction layer determines the importance degree of each historical recommendation information relative to the candidate recommendation information, and determines a comprehensive feature vector according to the importance degree of each historical recommendation information relative to the candidate recommendation information, the feature vector corresponding to each historical recommendation information and the feature vector corresponding to the candidate recommendation information;
and inputting the comprehensive characteristic vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, before inputting the integrated feature vector to the prediction layer, the method further comprises:
determining a feature compensation value corresponding to the comprehensive feature vector through the feature extraction layer, wherein the feature compensation value is used for representing the correlation degree of each dimension feature in the comprehensive feature vector and the set service type;
determining a compensated comprehensive characteristic vector according to the comprehensive characteristic vector and the characteristic compensation value;
inputting the comprehensive feature vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information, wherein the recommendation score comprises:
and inputting the compensated comprehensive characteristic vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, before inputting the integrated feature vector to the prediction layer, the method further comprises:
determining a feature vector of the candidate recommendation information under the set service type according to the service related information of the candidate recommendation information under the set service type;
inputting the comprehensive feature vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information, wherein the recommendation score comprises:
and inputting at least one of the comprehensive characteristic vector and the compensated comprehensive characteristic vector and the characteristic vector of the candidate recommendation information under the set service type into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, determining a feature vector corresponding to each piece of historical recommendation information according to a service topology map constructed in advance, and determining a feature vector corresponding to each piece of candidate recommendation information, specifically including:
taking the historical recommendation information and the candidate recommendation information as target recommendation information;
determining each information node corresponding to the target recommendation information according to the service topological graph;
and aiming at each information node, determining the characteristic vector corresponding to the information node according to the characteristic vector of the neighbor node of the information node and the weight corresponding to the type of the edge between the information node and the neighbor node.
This specification provides an apparatus for information recommendation, including:
the response module is used for responding to a service request of a target user for a set service type, determining candidate recommendation information corresponding to the target user under the set service type and historical recommendation information which has a service execution relation with the target user historically, wherein the historical recommendation information comprises recommendation information of different service types;
the determining module is used for determining a feature vector corresponding to each historical recommendation information according to a pre-constructed service topological graph, and determining a feature vector corresponding to each candidate recommendation information aiming at each candidate recommendation information, wherein the service topological graph is used for representing service relationships between the recommendation information under each service type and each user;
the input module is used for inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model to obtain a recommendation score corresponding to the candidate recommendation information;
and the recommending module is used for recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
The present specification provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the above-described information recommendation method.
The present specification provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the above information recommendation method when executing the program.
The technical scheme adopted by the specification can achieve the following beneficial effects:
in the information recommendation method provided in this specification, a service platform may determine, in response to a service request of a target user for a set service type, candidate recommendation information corresponding to the target user under the set service type and historical recommendation information having a service execution relationship with the target user historically, where the historical recommendation information includes recommendation information of different service types, and then may determine, according to a service topology map constructed in advance, a feature vector corresponding to each historical recommendation information and a feature vector corresponding to each candidate recommendation information, where the service topology map is used to represent a service relationship between the recommendation information under each service type and each user, and input the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a recommendation model trained in advance, and finally, recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
It can be seen from the above method that the service topology map may include service relationships between users and recommendation information in service types, that is, the service topology map includes service relationships between recommendation information in service types other than the set service type and users, and the feature vector of the recommendation information determined by the service topology map refers to the service relationship between the recommendation information and the user under each service type, so that the feature vector determined by the service topology map can implicitly show the preference of each user for the recommendation information of each service type. The historical recommendation information can clearly show the preference of a target user for recommendation information under different service types, and the feature vectors of the historical recommendation information and the candidate recommendation information are determined through the service topological graph, so that information in other service types can be fully introduced into the recommendation of the recommendation information of the set service type, and the accuracy of information recommendation under the set service type is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the specification and are incorporated in and constitute a part of this specification, illustrate embodiments of the specification and together with the description serve to explain the specification and not to limit the specification in a non-limiting sense. In the drawings:
FIG. 1 is a schematic flow chart of a method for information recommendation in this specification;
fig. 2 is a schematic diagram of a service topology provided in the present specification;
FIG. 3 is a schematic diagram of a proposed model provided in the present specification;
FIG. 4 is a schematic diagram of an information recommendation apparatus provided herein;
fig. 5 is a schematic diagram of an electronic device corresponding to fig. 1 provided in the present specification.
Detailed Description
In order to make the objects, technical solutions and advantages of the present disclosure more clear, the technical solutions of the present disclosure will be clearly and completely described below with reference to the specific embodiments of the present disclosure and the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without any creative effort belong to the protection scope of the present specification.
The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for information recommendation in this specification, including the following steps:
s101: responding to a service request of a target user for a set service type, and determining candidate recommendation information corresponding to the target user under the set service type and historical recommendation information which has a service execution relation with the target user historically, wherein the historical recommendation information comprises recommendation information of different service types.
In practical application, the service platform needs to perform information recommendation for services under some service types, for example, in a takeout service, the service platform may recommend merchants, takeout meals, and the like, which may be of interest to the user, to the user. For another example, in an e-commerce business, the business platform may recommend to the user goods that may be of interest to the user. When information recommendation is performed on one service, the service platform can introduce information of services under other service types, for example, for take-away services, the service platform can introduce information in offline catering services, so that information recommendation is performed more accurately by a user.
Based on this, the service platform may respond to a service request of a target user for a set service type, and determine candidate recommendation information corresponding to the target user under the set service type and historical recommendation information having a service execution relationship with the target user historically, where the historical recommendation information includes recommendation information of different service types, that is, the historical recommendation information may include recommendation information of a service type different from the set service type.
The set service type may be a service type that the service platform needs to perform information recommendation, and the service request of the target user for the set service type may be a request sent to the service platform through the terminal when the user needs to view recommendation information under the set service type. Each piece of historical recommendation information may refer to recommendation information of a user that has historically performed business operations such as purchase and click (i.e., has a business execution relationship), and the business platform may sort the recommendation information having a business execution relationship with the user historically according to a time sequence, and set recommendation information before sorting as each piece of determined historical recommendation information.
S102: and determining a characteristic vector corresponding to each historical recommendation information according to a pre-constructed service topological graph, and determining a characteristic vector corresponding to each candidate recommendation information aiming at each candidate recommendation information, wherein the service topological graph is used for representing service relations between the recommendation information under each service type and each user.
After determining each piece of historical recommendation information and each piece of candidate recommendation information, the service platform may determine, according to a service topology map constructed in advance, a feature vector corresponding to each piece of historical recommendation information, and determine, for each piece of candidate recommendation information, a feature vector corresponding to the candidate recommendation information. The service topological graph can comprise information nodes corresponding to the historical recommendation information and the candidate recommendation information and user nodes corresponding to the users, the service platform can determine the characteristic vector of the information node corresponding to the historical recommendation information through the service topological graph to serve as the characteristic vector corresponding to the historical recommendation information, and determine the characteristic vector of the information node corresponding to the candidate recommendation information through the service topological graph to serve as the characteristic vector corresponding to the candidate recommendation information.
The service topology graph is used to represent a service relationship between the recommendation information under each service type and each user, that is, the service topology graph can represent service execution conditions of each user on the recommendation information under each service type. Because the feature vector of the recommendation information determined by the service topological graph can show the topological structure in the neighborhood range where the information node corresponding to the recommendation information is located, the feature vector corresponding to the recommendation information determined by the service topological graph can represent the association between the recommendation information and the recommendation information of different service types and the association between the recommendation information and each user.
The service topology graph may include user nodes and information nodes, each information node is used to represent one piece of recommendation information, each user node is used to represent one user, and an edge (which may be a directed edge or an undirected edge) between a user node corresponding to one user and an information node corresponding to one piece of recommendation information may represent a service operation (such as purchase, click, and the like) that the user has historically existed for the recommendation information, as shown in fig. 2.
Fig. 2 is a schematic diagram of a service topology provided in this specification.
As can be seen from fig. 2, the recommendation information a and the recommendation information B belong to a service type 1, the recommendation information D and the recommendation information E belong to a service type 2, and the recommendation information C belongs to recommendation information common to the service types 1 and 2, for example, assuming that the service type 1 is a takeout service and the service type 2 is an offline catering service, if a certain merchant has both a takeout service and an online catering service, the recommendation information corresponding to the merchant belongs to both service types, the online catering service and the identifier in the takeout service of the merchant can be unified, and a store in the online catering service of the merchant and a store in the takeout service of the merchant are represented by the same node in a service topology map. In the service topology graph, a service operation performed by a user for a certain piece of recommendation information may be represented by a directed edge pointing to an information node by a user node, for example, an edge pointing to an information node corresponding to recommendation information E from a user node corresponding to a user b in fig. 2 may represent that the user b has performed a purchase operation for the recommendation information E historically.
It should be noted that, besides the service relationship between the recommendation information and each user, the service topology map may further include a plurality of service relationships, so that when the service topology map is constructed, the service topology map may be constructed through a plurality of information, specifically, the service platform may construct the service topology map according to the integrated service information, where the integrated service information may include at least one of a historical service record of each user performing a service for each recommendation information, a geographic area corresponding to each recommendation information, and an information name corresponding to each recommendation information, and each recommendation information includes recommendation information recommended by a different service.
The types of edges between nodes in the service topological graph constructed by the service platform through different types of integrated service information are different, that is, the types of the edges constructed through the historical service records, the edges constructed through the geographic areas and the edges constructed through the information names are three different types of edges, the edges representing the service execution condition between the user and the recommended information are also different from the three types of edges, and it can be understood that the service relationships represented by the edges constructed through different types of information are different, and the edges representing different service relationships belong to different types of edges.
The history service record of performing service execution on each piece of recommendation information by each user can show the service execution sequence of each user on each piece of recommendation information, so that the service platform can determine the service execution sequence of each user on each piece of recommendation information according to the history service record, and construct directed edges between information nodes according to the service execution sequence of each user on each piece of recommendation information to construct the service topological graph, for example, if the user performs a purchase operation on the recommendation information a first and then performs a purchase operation on the recommendation information B, a directed edge pointing from the information node corresponding to the recommendation information a to the information node corresponding to the recommendation information B can be constructed, and if the user performs a purchase operation on the recommendation information C subsequently, a directed edge pointing from the information node corresponding to the recommendation information B to the information node corresponding to the recommendation information C can be constructed, in this way, the service topology chart can show that the interest of the user changes for the recommendation information under various service types.
In addition to the information nodes and the user nodes mentioned above, a region node may also exist in the service topology map, where a region node is used to represent a geographic region, and when the service topology map is constructed by the geographic region corresponding to each piece of recommendation information, the service platform may construct, for each piece of recommendation information, an edge (which may be an undirected edge or a directed edge) between the information node corresponding to the recommendation information and the region node of the geographic region corresponding to the recommendation information according to the geographic region corresponding to the recommendation information, so as to construct the topology map, that is, the information nodes belonging to the same geographic region and the region node corresponding to the geographic region may be connected, so as to reflect a relationship between the recommendation information on the geographic region, where the geographic region mentioned here may refer to a city, a district, a county, or the like.
In addition to the relationship of the recommendation information in the geographic area, the service platform may also construct, according to the information names of the recommendation information, edges between information nodes of the recommendation information having similar information names to construct the service topological graph, where the information names mentioned herein may be determined in various ways, for example, the service platform may determine text vectors of the information names of each recommendation information, and take recommendation information having a similarity higher than a set similarity between the text vectors as recommendation information having similar information names, and for example, the service platform may take recommendation information having a higher repetition rate of characters in the information names of the recommendation information as recommendation information having similar information names.
Because each piece of recommendation information contains recommendation information of different service types, even if the recommendation information belongs to different services, if the information names of the recommendation information are similar, similarity may exist among the recommendation information, therefore, by constructing edges among information nodes of the recommendation information with similar information names, a certain relation can be constructed among the recommendation information belonging to different service types but similar to each other, and of course, besides the information names, the service platform can also construct a service topological graph according to comprehensive service information such as recommendation information types corresponding to the recommendation information, service money amounts corresponding to the recommendation information, and the like, so as to establish service relationships among the recommendation information.
Since the service topological graph includes a plurality of different types of edges, that is, the service topological graph includes a plurality of service relationships, when determining the feature vector of a node in the service topological graph, in addition to the determination by the neighboring nodes around the node, the determination may also be performed by the types of the edges between the node and the neighboring nodes of the node, and the importance degree of each type of edge is determined, so that the more important service relationship occupies more proportion in the feature vector, therefore, the service platform may use each historical recommendation information and each candidate recommendation information as the target recommendation information, determine each information node corresponding to the target recommendation information according to the service topological graph, and determine, for each information node, the weight corresponding to the feature vector of the neighboring node of the information node and the determined type of the edge between the information node and the neighboring node, and determining a feature vector corresponding to the information node.
For example, for one type, feature vectors of neighboring nodes connected by the type of edge may be determined, and attention weights for the type of edge may be determined by feature vectors corresponding to the neighboring nodes. Of course, the weight for each type of edge may also be preset.
S103: and inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model to obtain a recommendation score corresponding to the candidate recommendation information.
After determining the feature vector corresponding to each candidate recommendation information and the feature vector corresponding to each historical recommendation information, the service platform may input the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model for each candidate recommendation information, so as to obtain a recommendation score corresponding to the candidate recommendation information. The service topological graph relates to recommendation information under each service type, and can show the service relationship between the recommendation information under each service type and each user and the relationship between the recommendation information under each service type, so that the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information determined by the service topological graph both take the prior information of the service under each service type into consideration, and each historical recommendation information may also relate to recommendation information under other service types except the set service type, so that the recommendation score of the candidate recommendation information obtained by the recommendation model refers to various information (such as user preference) in other service types, and more accurate recommendation score can be determined.
Further, in order to determine a recommendation score corresponding to candidate recommendation information more accurately, when the service platform determines the recommendation score, the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each piece of historical recommendation information may be input to a feature extraction layer included in a recommendation model, so that the feature extraction layer determines an importance degree of each piece of historical recommendation information relative to the candidate recommendation information, determines a comprehensive feature vector corresponding to the candidate recommendation information according to the importance degree of each piece of historical recommendation information relative to the candidate recommendation information, the feature vector corresponding to each piece of historical recommendation information, and the feature vector corresponding to the candidate recommendation information, and further inputs the comprehensive feature vector into a prediction layer to obtain the recommendation score corresponding to the candidate recommendation information.
For example, the service platform may determine, according to a feature vector corresponding to each piece of historical recommendation information and a feature vector corresponding to the candidate recommendation information, an attention weight of each piece of historical recommendation information to the candidate recommendation information as an importance degree corresponding to each piece of historical recommendation information, and for example, the service platform may determine, as an importance degree corresponding to each piece of historical recommendation information, a correlation degree between the feature vector corresponding to each piece of historical recommendation information and the feature vector corresponding to the candidate recommendation information.
The service platform can perform weighted summation on the feature vectors corresponding to the historical recommendation information according to the importance degree of the historical recommendation information relative to the candidate recommendation information, and then obtain the comprehensive feature vector by combining the feature vectors corresponding to the candidate recommendation information.
The manner of determining the feature compensation value corresponding to the integrated feature vector also includes various manners, for example, the feature compensation value may be determined by an attention network, and of course, may also be determined by a gating mechanism. The feature compensation value has the effect of limiting certain features in the composite feature vector of the candidate recommendation information to be used for determining the recommendation score of the candidate recommendation information. That is to say, since the comprehensive feature vector contains information under many other service types, some information contained in the comprehensive feature vector may not be needed by the recommendation model, and the information not needed by the recommendation model can be filtered out by the feature compensation value, so that the information contained in the compensated comprehensive feature vector is needed by the recommendation model, and the recommendation score can be determined more accurately according to the candidate recommendation information.
It should be noted that, although the above-mentioned integrated feature vector or the compensated integrated feature vector is derived from the service topology, that is, information included in both of the integrated feature vector and the compensated integrated feature vector under a plurality of service types, in this embodiment, information recommendation is performed only for a set service type, and therefore, information recommendation using information unique to the set service type is also required. Therefore, the service platform may determine the feature vector of the candidate recommendation information under the set service type according to the service-related information of the candidate recommendation information under the set service type, and the service platform may input at least one of the comprehensive feature vector and the compensated comprehensive feature vector, and the feature vector of the candidate recommendation information under the set service type to the prediction layer, so as to obtain the recommendation score corresponding to the candidate recommendation information.
The recommendation model in the present specification is described above in various aspects, and the overall structure of the recommendation model is described below as an example, as shown in fig. 3.
Fig. 3 is a schematic structural diagram of a recommendation model provided in this specification.
Fig. 3 is an architecture of the recommendation model, a network structure included in the feature extraction layer may be replaced according to other manners in the above description, and as can be seen from fig. 3, a feature matrix formed by feature vectors corresponding to candidate recommendation information and feature vectors corresponding to each historical feature information may be input into an attention network included in the feature extraction layer, a comprehensive feature vector is determined, and the comprehensive feature vector is compensated by a door frame mechanism, so that a compensated comprehensive feature vector is obtained. And inputting the compensated comprehensive characteristic vector and the characteristic vector of the candidate recommendation information under the set service type into a prediction layer to obtain a recommendation score.
S104: and recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
After determining the recommendation score corresponding to each candidate recommendation information in the above manner, the service platform may recommend information to the target user according to the recommendation score corresponding to each candidate recommendation information. Specifically, the candidate recommendation information with a high recommendation score may be recommended to the user at the previous time, and the candidate recommendation with a low recommendation score may be recommended to the user at the later time or may not be recommended to the user.
It can be seen from the above method that the service topology map may include a plurality of service relationships, such as a service relationship between each user and the recommendation information in each service type and a service relationship between each recommendation information, and the feature vector of the recommendation information determined by the service topology map may refer to the service relationship between the recommendation information and the user in each service type, so that the feature vector determined by the service topology map may implicitly present the preference of each user for the recommendation information in each service type. The historical recommendation information can clearly show the preference of a target user for recommendation information under different service types, and the feature vectors of the historical recommendation information and the candidate recommendation information are determined through the service topological graph, so that information in other service types can be fully introduced into the recommendation of the recommendation information of the set service type, and the accuracy of information recommendation under the set service type is improved.
Based on the same idea, the present specification further provides a corresponding information recommendation apparatus, as shown in fig. 4.
Fig. 4 is a schematic diagram of an information recommendation apparatus provided in this specification, including:
a response module 401, configured to respond to a service request of a target user for a set service type, determine candidate recommendation information corresponding to the target user in the set service type and historical recommendation information having a service execution relationship with the target user historically, where the historical recommendation information includes recommendation information of different service types;
a determining module 402, configured to determine, according to a service topological graph constructed in advance, a feature vector corresponding to each piece of historical recommendation information, and determine, for each piece of candidate recommendation information, a feature vector corresponding to the candidate recommendation information, where the service topological graph is used to represent a service relationship between recommendation information in each service type and each user;
an input module 403, configured to input the feature vector corresponding to the candidate recommendation information and the feature vectors corresponding to the historical recommendation information into a pre-trained recommendation model, so as to obtain a recommendation score corresponding to the candidate recommendation information;
and the recommending module 404 is configured to recommend information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
Optionally, the apparatus further comprises:
a building module 405, configured to build the service topology map according to integrated service information, where the integrated service information includes at least one of a historical service record of service execution performed by each user for each piece of recommended information, a geographic area corresponding to each piece of recommended information, and an information name corresponding to each piece of recommended information, and the types of edges between nodes in the service topology map built by different types of integrated service information are different, and each piece of recommended information includes recommended information under each service type.
Optionally, the service topology map includes an information node, where the information node is used to represent recommendation information;
the building module 405 is specifically configured to determine, according to the historical service record, a service execution sequence of each user for each piece of recommended information; and according to the service execution sequence of each user aiming at each recommended information, establishing directed edges among the information nodes so as to establish the service topological graph.
Optionally, the service topology map includes an information node and a region node, where the information node is used to represent recommendation information, and the region node is used to represent a geographic region;
the building module 405 is specifically configured to, for each piece of recommendation information, build, according to a geographic area corresponding to the recommendation information, an edge between an information node corresponding to the recommendation information and an area node of the geographic area corresponding to the recommendation information, so as to build the service topology map.
Optionally, the service topology map includes an information node, where the information node is used to represent recommendation information;
the building module 405 is specifically configured to build, according to the information name of each piece of recommended information, an edge between information nodes of pieces of recommended information having similar information names, so as to build the service topology map.
Optionally, the recommendation model includes a feature extraction layer and a prediction layer;
the input module 403 is specifically configured to input the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each piece of historical recommendation information into the feature extraction layer, so that the feature extraction layer determines the importance degree of each piece of historical recommendation information relative to the candidate recommendation information, and determines a comprehensive feature vector according to the importance degree of each piece of historical recommendation information relative to the candidate recommendation information, the feature vector corresponding to each piece of historical recommendation information, and the feature vector corresponding to the candidate recommendation information; and inputting the comprehensive characteristic vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, before the comprehensive feature vector is input to the prediction layer, the input module 403 is further configured to determine, by the feature extraction layer, a feature compensation value corresponding to the comprehensive feature vector, where the feature compensation value is used to represent a degree of correlation between each dimensional feature in the comprehensive feature vector and the set service type; determining a compensated comprehensive characteristic vector according to the comprehensive characteristic vector and the characteristic compensation value;
the input module 403 is specifically configured to input the compensated comprehensive feature vector into the prediction layer, so as to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, before the comprehensive feature vector is input to the prediction layer, the input module 403 is further configured to determine, according to the service-related information of the candidate recommendation information under the set service type, the feature vector of the candidate recommendation information under the set service type; and inputting at least one of the comprehensive characteristic vector and the compensated comprehensive characteristic vector and the characteristic vector of the candidate recommendation information under the set service type into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
Optionally, the determining module 402 is specifically configured to use the historical recommendation information and the candidate recommendation information as target recommendation information; determining each information node corresponding to the target recommendation information according to the service topological graph; and aiming at each information node, determining the characteristic vector corresponding to the information node according to the characteristic vector of the neighbor node of the information node and the weight corresponding to the type of the edge between the information node and the neighbor node.
The present specification also provides a computer-readable storage medium storing a computer program, which can be used to execute a method of information recommendation provided in fig. 1 above.
This specification also provides a schematic block diagram of an electronic device corresponding to that of figure 1, shown in figure 5. As shown in fig. 5, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, but may also include hardware required for other services. The processor reads a corresponding computer program from the non-volatile memory into the memory and then runs the computer program to implement the information recommendation method described in fig. 1. Of course, besides the software implementation, the present specification does not exclude other implementations, such as logic devices or a combination of software and hardware, and the like, that is, the execution subject of the following processing flow is not limited to each logic unit, and may be hardware or logic devices.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the various elements may be implemented in the same one or more software and/or hardware implementations of the present description.
As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the 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 the like) 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.
As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the 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 the like) having computer-usable program code embodied therein.
This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.
Claims (12)
1. A method for information recommendation, comprising:
responding to a service request of a target user for a set service type, and determining candidate recommendation information corresponding to the target user under the set service type and historical recommendation information which has a service execution relation with the target user historically, wherein the historical recommendation information comprises recommendation information of different service types;
determining a feature vector corresponding to each historical recommendation information according to a pre-constructed service topological graph, and determining a feature vector corresponding to each candidate recommendation information aiming at each candidate recommendation information, wherein the service topological graph is used for representing service relationships between the recommendation information under each service type and each user;
inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model to obtain a recommendation score corresponding to the candidate recommendation information;
and recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
2. The method according to claim 1, wherein constructing the service topology map specifically comprises:
the method comprises the steps of constructing the service topological graph according to integrated service information, wherein the integrated service information comprises at least one of historical service records of service execution of each user aiming at each recommended information, a geographical area corresponding to each recommended information and an information name corresponding to each recommended information, the types of edges between nodes in the service topological graph constructed through different types of integrated service information are different, and each recommended information comprises recommended information under each service type.
3. The method of claim 2, wherein the service topology map comprises an information node, and the information node is used for representing recommendation information;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
determining a service execution sequence of each user aiming at each recommended information according to the historical service record;
and according to the service execution sequence of each user aiming at each recommended information, establishing directed edges among the information nodes so as to establish the service topological graph.
4. The method of claim 2, wherein the service topology map comprises information nodes and area nodes, the information nodes are used for representing recommendation information, and the area nodes are used for representing a geographical area;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
and aiming at each piece of recommendation information, according to the geographical area corresponding to the recommendation information, constructing edges between the information node corresponding to the recommendation information and the area node of the geographical area corresponding to the recommendation information so as to construct the service topological graph.
5. The method of claim 2, wherein the service topology map comprises an information node, and the information node is used for representing recommendation information;
according to the integrated service information, constructing the service topological graph, which specifically comprises the following steps:
and according to the information names of the pieces of recommended information, constructing edges among the information nodes of the pieces of recommended information with similar information names so as to construct the service topological graph.
6. The method of claim 1, wherein the recommendation model includes a feature extraction layer and a prediction layer;
inputting the feature vector corresponding to the candidate recommendation information and the feature vectors corresponding to the historical recommendation information into a pre-trained recommendation model to obtain recommendation scores corresponding to the candidate recommendation information, and specifically comprising:
inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into the feature extraction layer, so that the feature extraction layer determines the importance degree of each historical recommendation information relative to the candidate recommendation information, and determines a comprehensive feature vector according to the importance degree of each historical recommendation information relative to the candidate recommendation information, the feature vector corresponding to each historical recommendation information and the feature vector corresponding to the candidate recommendation information;
and inputting the comprehensive characteristic vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
7. The method of claim 6, wherein prior to inputting the synthesized feature vector to the prediction layer, the method further comprises:
determining a feature compensation value corresponding to the comprehensive feature vector through the feature extraction layer, wherein the feature compensation value is used for representing the correlation degree of each dimension feature in the comprehensive feature vector and the set service type;
determining a compensated comprehensive characteristic vector according to the comprehensive characteristic vector and the characteristic compensation value;
inputting the comprehensive feature vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information, wherein the recommendation score comprises:
and inputting the compensated comprehensive characteristic vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
8. The method of claim 6 or 7, wherein prior to inputting the synthesized feature vector to the prediction layer, the method further comprises:
determining a feature vector of the candidate recommendation information under the set service type according to the service related information of the candidate recommendation information under the set service type;
inputting the comprehensive feature vector into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information, wherein the recommendation score comprises:
and inputting at least one of the comprehensive characteristic vector and the compensated comprehensive characteristic vector and the characteristic vector of the candidate recommendation information under the set service type into the prediction layer to obtain a recommendation score corresponding to the candidate recommendation information.
9. The method according to claim 2, wherein determining the feature vector corresponding to each piece of historical recommendation information according to a pre-constructed service topology map, and determining the feature vector corresponding to each piece of candidate recommendation information for each piece of candidate recommendation information specifically includes:
taking the historical recommendation information and the candidate recommendation information as target recommendation information;
determining each information node corresponding to the target recommendation information according to the service topological graph;
and aiming at each information node, determining the characteristic vector corresponding to the information node according to the characteristic vector of the neighbor node of the information node and the weight corresponding to the type of the edge between the information node and the neighbor node.
10. An apparatus for information recommendation, comprising:
the response module is used for responding to a service request of a target user for a set service type, determining candidate recommendation information corresponding to the target user under the set service type and historical recommendation information which has a service execution relation with the target user historically, wherein the historical recommendation information comprises recommendation information of different service types;
the determining module is used for determining a feature vector corresponding to each historical recommendation information according to a pre-constructed service topological graph, and determining a feature vector corresponding to each candidate recommendation information aiming at each candidate recommendation information, wherein the service topological graph is used for representing service relationships between the recommendation information under each service type and each user;
the input module is used for inputting the feature vector corresponding to the candidate recommendation information and the feature vector corresponding to each historical recommendation information into a pre-trained recommendation model to obtain a recommendation score corresponding to the candidate recommendation information;
and the recommending module is used for recommending information to the target user according to the recommendation scores corresponding to the candidate recommendation information.
11. A computer-readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1 to 9.
12. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 9 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110793233.1A CN113641892A (en) | 2021-07-14 | 2021-07-14 | Information recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110793233.1A CN113641892A (en) | 2021-07-14 | 2021-07-14 | Information recommendation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113641892A true CN113641892A (en) | 2021-11-12 |
Family
ID=78417279
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110793233.1A Withdrawn CN113641892A (en) | 2021-07-14 | 2021-07-14 | Information recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113641892A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417136A (en) * | 2021-12-24 | 2022-04-29 | 珠海大横琴科技发展有限公司 | User object recommendation method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124464A1 (en) * | 2015-10-28 | 2017-05-04 | Fractal Industries, Inc. | Rapid predictive analysis of very large data sets using the distributed computational graph |
| CN109784599A (en) * | 2017-11-14 | 2019-05-21 | 阿里巴巴集团控股有限公司 | A kind of method, device and equipment of model training, risk identification |
| CN112418920A (en) * | 2020-11-11 | 2021-02-26 | 北京三快在线科技有限公司 | Training method of information recommendation model, information recommendation method and device |
| CN112507218A (en) * | 2020-12-03 | 2021-03-16 | 广州华多网络科技有限公司 | Business object recommendation method and device, electronic equipment and storage medium |
| CN113011483A (en) * | 2021-03-11 | 2021-06-22 | 北京三快在线科技有限公司 | Method and device for model training and business processing |
-
2021
- 2021-07-14 CN CN202110793233.1A patent/CN113641892A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124464A1 (en) * | 2015-10-28 | 2017-05-04 | Fractal Industries, Inc. | Rapid predictive analysis of very large data sets using the distributed computational graph |
| CN109784599A (en) * | 2017-11-14 | 2019-05-21 | 阿里巴巴集团控股有限公司 | A kind of method, device and equipment of model training, risk identification |
| CN112418920A (en) * | 2020-11-11 | 2021-02-26 | 北京三快在线科技有限公司 | Training method of information recommendation model, information recommendation method and device |
| CN112507218A (en) * | 2020-12-03 | 2021-03-16 | 广州华多网络科技有限公司 | Business object recommendation method and device, electronic equipment and storage medium |
| CN113011483A (en) * | 2021-03-11 | 2021-06-22 | 北京三快在线科技有限公司 | Method and device for model training and business processing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417136A (en) * | 2021-12-24 | 2022-04-29 | 珠海大横琴科技发展有限公司 | User object recommendation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108268617B (en) | User intention determining method and device | |
| CN110020427B (en) | Policy determination method and device | |
| CN113011483B (en) | Method and device for model training and business processing | |
| CN113688313A (en) | Training method of prediction model, information pushing method and device | |
| CN113641896A (en) | Model training and recommendation probability prediction method and device | |
| CN113344098A (en) | Model training method and device | |
| CN111538794A (en) | Data fusion method, device and equipment | |
| CN110399359A (en) | A kind of data retrogressive method, device and equipment | |
| CN113641894A (en) | Information recommendation method and device | |
| CN114331602A (en) | Model training method based on transfer learning, information recommendation method and device | |
| CN113343085B (en) | Information recommendation method and device, storage medium and electronic equipment | |
| CN113641892A (en) | Information recommendation method and device | |
| CN113343132B (en) | Model training method, information display method and device | |
| CN115660105A (en) | Model training method, business wind control method and business wind control device | |
| CN113205377A (en) | Information recommendation method and device | |
| CN114996570A (en) | Information recommendation method and device | |
| CN113343141A (en) | Webpage obtaining method and device | |
| CN113344197A (en) | Training method of recognition model, service execution method and device | |
| CN115034861B (en) | Learning method, device and equipment for long tail distribution | |
| CN114676751A (en) | Model training and map-based hierarchical recommendation method and device | |
| CN115130000A (en) | Information recommendation method and device, storage medium and electronic equipment | |
| CN116070921A (en) | Service policy generation method, device and equipment | |
| CN112598279A (en) | Information providing method and device | |
| CN115048578A (en) | Recommendation method and device | |
| CN113987340A (en) | Method and device for recommending guide words |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211112 |