阅读说明：本技术 商品对象推荐方法及其装置、设备、介质、产品 (Commodity object recommendation method and device, equipment, medium and product thereof ) 是由 张铨 车天文 钟媛媛 于 2021-09-15 设计创作，主要内容包括：本申请涉及电商信息技术领域,公开一种商品对象推荐方法及其装置、设备、介质、产品,所述方法包括：接收客户端响应用户访问在售商品对象的行为而提交的用户行为消息,将其添加到行为消息队列中；监听从行为消息队列出列的用户行为消息,获取该用户行为消息所指向的在售商品对象；从预设的商品相似度信息矩阵中查询获取与该在售商品对象构成相似的候选商品对象以构造商品推荐列表,所述商品推荐列表包含从所述候选商品对象中优选出的目标商品对象；响应所述目标用户的商品推荐请求,向其推送相应的商品推荐列表。本申请能够即时根据用户访问在售商品对象的行为而产生与该在售商品对象构成相似的目标商品对象,匹配精准,适用于多种应用场景。(The application relates to the technical field of E-commerce information, and discloses a commodity object recommendation method, a device, equipment, a medium and a product thereof, wherein the method comprises the following steps: receiving a user behavior message submitted by a client in response to the behavior of a user accessing an on-sale commodity object, and adding the user behavior message into a behavior message queue; monitoring user behavior messages listed from the behavior message queue, and acquiring the commodity object on sale pointed by the user behavior messages; inquiring and acquiring candidate commodity objects similar to the current commodity object from a preset commodity similarity information matrix to construct a commodity recommendation list, wherein the commodity recommendation list comprises target commodity objects selected from the candidate commodity objects; and responding to the commodity recommendation request of the target user, and pushing a corresponding commodity recommendation list to the target user. The method and the device can generate the target commodity object similar to the commodity object in sale according to the behavior of the user for accessing the commodity object in sale in real time, are accurate in matching and are suitable for various application scenes.)

1. A commodity object recommendation method is characterized by comprising the following steps:

receiving a user behavior message submitted by a client in response to the behavior of a user accessing an on-sale commodity object, and adding the user behavior message into a behavior message queue;

monitoring user behavior messages listed from the behavior message queue, and acquiring the commodity object on sale pointed by the user behavior messages;

inquiring and acquiring candidate commodity objects similar to the current commodity object from a preset commodity similarity information matrix to construct a commodity recommendation list, wherein the commodity recommendation list comprises target commodity objects selected from the candidate commodity objects;

and responding to the commodity recommendation request of the target user, and pushing a corresponding commodity recommendation list to the target user.

2. The commodity object recommendation method according to claim 1, wherein candidate commodity objects similar to the commodity object in sale are obtained by searching in a preset commodity similarity information matrix to construct a commodity recommendation list, comprising the steps of:

acquiring the uniqueness characteristic information of the commodity object on sale, wherein the uniqueness characteristic information and the dimension label of the commodity similarity information matrix have a one-to-one mapping relation;

inquiring a commodity similarity information matrix according to the unique characteristic information, and determining a row vector corresponding to the commodity object for sale, wherein each element of the row vector stores a similarity value for measuring the similarity between the commodity object for sale and a corresponding candidate commodity object;

determining a plurality of candidate commodity objects of which the similarity values meet similar matching conditions according to the row vectors;

and constructing a commodity recommendation list, and adding at least one candidate commodity object meeting the similar matching condition as a target commodity object into the commodity recommendation list.

3. The merchandise object recommendation method according to claim 2, wherein determining a plurality of candidate merchandise objects whose similarity values satisfy a similarity matching condition according to the row vector comprises:

sorting the row vectors corresponding to the commodity objects for sale according to the similar numerical values;

preferably selecting a plurality of elements with the maximum similarity values from the sorted row vectors according to a preset similarity matching condition;

and determining a plurality of corresponding candidate commodity objects meeting the similar matching conditions according to the dimension labels of the preferred elements in the commodity similarity information matrix.

4. The commodity object recommendation method according to claim 2, wherein constructing a commodity recommendation list, and adding at least one candidate commodity object satisfying the similarity matching condition as a target commodity object to the commodity recommendation list comprises the steps of:

calling popularity reference information determined according to the access popularity of the commodity object to obtain commodity popularity data corresponding to a plurality of candidate commodity objects meeting the similar matching conditions;

filtering out candidate commodity objects with commodity popularity data lower than a preset threshold value from the candidate commodity objects meeting the similar matching conditions to obtain at least one remaining target commodity object;

and constructing a commodity recommendation list, wherein the commodity recommendation list stores commodity abstract texts and commodity pictures corresponding to the target commodity objects.

5. The commodity object recommendation method according to claim 2, wherein constructing a commodity recommendation list, and adding at least one candidate commodity object satisfying the similarity matching condition as a target commodity object to the commodity recommendation list comprises the steps of:

calling historical order data of a target user providing the user behavior message to determine a purchased commodity object;

filtering the purchased commodity object from a plurality of candidate commodity objects meeting the similar matching condition to obtain at least one remaining target commodity object;

and constructing a commodity recommendation list, wherein the commodity recommendation list stores commodity abstract texts and commodity pictures corresponding to the target commodity objects.

6. The commodity object recommendation method according to any one of claims 1 to 5, wherein receiving a user behavior message submitted by a client in response to a user's behavior of accessing a commodity object for sale, and adding the user behavior message to a behavior message queue comprises:

receiving a user behavior message submitted by a client in response to the behavior of a user accessing an on-sale commodity object, judging whether a preset user behavior queue is in a congestion state, if so, starting an asynchronous user behavior queue, adding the user behavior message into the asynchronous user behavior queue, otherwise, adding the user behavior message into the preset user behavior queue.

7. The merchandise object recommendation method according to any one of claims 1 to 5, characterized in that the method comprises the steps for constructing the merchandise similarity information matrix as follows:

constructing an image characteristic similarity matrix between every two commodity objects based on the image characteristic information of the commodity pictures of the commodity objects in the commodity database, and storing similarity values between each commodity object and other commodity objects in the same row vector;

extracting classification labels of the commodity objects based on the text information of the commodity objects in the commodity database, determining a classification label similarity value between every two commodity objects, and constructing a text feature similarity matrix;

determining a corresponding relation according to the same two commodity objects, and carrying out linear fusion on the image characteristic similarity matrix and the similar values with the corresponding relation in the text characteristic similarity matrix to construct a commodity similarity information matrix, wherein the similar values between each commodity object and other commodity objects in the matrix are stored in the same row vector;

and sorting the same row vector in the commodity similarity information matrix according to the size of the similarity value.

8. A computer device comprising a central processor and a memory, characterized in that the central processor is adapted to invoke execution of a computer program stored in the memory to perform the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that it stores, in the form of computer-readable instructions, a computer program implemented according to the method of any one of claims 1 to 6, which, when invoked by a computer, performs the steps comprised by the corresponding method.

10. A computer program product comprising computer program/instructions, characterized in that the computer program/instructions, when executed by a processor, implement the steps of the method as claimed in any one of claims 1 to 6.

Technical Field

The present application relates to the field of e-commerce information technologies, and in particular, to a method for recommending a commodity object, and a corresponding apparatus, computer device, computer-readable storage medium, and computer program product.

Background

Similar commodities are often recommended for users in an e-commerce platform, a common application scenario is that commodities similar to a commodity object are recommended in real time according to the commodity object being visited by the user, in practice, the commodity object which the user is interested in is often determined according to historical behavior habits of the user, the historical behavior habits of the user are usually access record data left by the user for visiting the commodity object in the e-commerce platform in the past period, including user behaviors such as ordering and browsing, but the access record data have long timeliness, and commodity object recommendation is performed depending on the data, so that the commodity object recommended to the user is lagged behind the requirements of the user, for example, the user has purchased the same kind of commodities or has lost the purchase requirements of the same kind of commodities. From the current commodity recommendation lists and patent retrieval data of the terminal APP or the webpage of various multi-e-commerce platforms, the problem is not effectively solved in the e-commerce platform industry so far.

For example, an algorithm adopted for recommending commodities in the current e-commerce platform often adopts a collaborative filtering strategy of item-CF and user-CF, and according to the inherent characteristics of the algorithm, the real-time recommendation of commodities interested by a user is difficult to be mined according to the user behavior.

Disclosure of Invention

A primary object of the present application is to solve at least one of the above problems and provide a method for recommending a commodity object, and a corresponding apparatus, a computer device, a computer-readable storage medium, and a computer program product.

In order to meet various purposes of the application, the following technical scheme is adopted in the application:

the commodity object recommendation method adaptive to one of the purposes of the application comprises the following steps:

receiving a user behavior message submitted by a client in response to the behavior of a user accessing an on-sale commodity object, and adding the user behavior message into a behavior message queue;

monitoring user behavior messages listed from the behavior message queue, and acquiring the commodity object on sale pointed by the user behavior messages;

inquiring and acquiring candidate commodity objects similar to the current commodity object from a preset commodity similarity information matrix to construct a commodity recommendation list, wherein the commodity recommendation list comprises target commodity objects selected from the candidate commodity objects;

and responding to the commodity recommendation request of the target user, and pushing a corresponding commodity recommendation list to the target user.

In a further embodiment, the method for constructing the commodity recommendation list by querying and acquiring candidate commodity objects similar to the commodity object in sale from a preset commodity similarity information matrix comprises the following steps:

acquiring the uniqueness characteristic information of the commodity object on sale, wherein the uniqueness characteristic information and the dimension label of the commodity similarity information matrix have a one-to-one mapping relation;

inquiring a commodity similarity information matrix according to the unique characteristic information, and determining a row vector corresponding to the commodity object for sale, wherein each element of the row vector stores a similarity value for measuring the similarity between the commodity object for sale and a corresponding candidate commodity object;

determining a plurality of candidate commodity objects of which the similarity values meet similar matching conditions according to the row vectors;

and constructing a commodity recommendation list, and adding at least one candidate commodity object meeting the similar matching condition as a target commodity object into the commodity recommendation list.

In an embodiment, determining a plurality of candidate commodity objects with the similarity values satisfying the similarity matching condition according to the row vector includes the following steps:

sorting the row vectors corresponding to the commodity objects for sale according to the similar numerical values;

preferably selecting a plurality of elements with the maximum similarity values from the sorted row vectors according to a preset similarity matching condition;

and determining a plurality of corresponding candidate commodity objects meeting the similar matching conditions according to the dimension labels of the preferred elements in the commodity similarity information matrix.

In an embodiment, constructing a commodity recommendation list, and adding at least one candidate commodity object satisfying the similar matching condition as a target commodity object to the commodity recommendation list, includes the following steps:

calling popularity reference information determined according to the access popularity of the commodity object to obtain commodity popularity data corresponding to a plurality of candidate commodity objects meeting the similar matching conditions;

filtering out candidate commodity objects with commodity popularity data lower than a preset threshold value from the candidate commodity objects meeting the similar matching conditions to obtain at least one remaining target commodity object;

and constructing a commodity recommendation list, wherein the commodity recommendation list stores commodity abstract texts and commodity pictures corresponding to the target commodity objects.

In an embodiment, constructing a commodity recommendation list, and adding at least one candidate commodity object satisfying the similar matching condition as a target commodity object to the commodity recommendation list, includes the following steps:

calling historical order data of a target user providing the user behavior message to determine a purchased commodity object;

filtering the purchased commodity object from a plurality of candidate commodity objects meeting the similar matching condition to obtain at least one remaining target commodity object;

and constructing a commodity recommendation list, wherein the commodity recommendation list stores commodity abstract texts and commodity pictures corresponding to the target commodity objects.

In a further embodiment, the receiving a user behavior message submitted by a client in response to a behavior of a user accessing an object of a commodity on sale and adding the user behavior message to a behavior message queue includes:

receiving a user behavior message submitted by a client in response to the behavior of a user accessing an on-sale commodity object, judging whether a preset user behavior queue is in a congestion state, if so, starting an asynchronous user behavior queue, adding the user behavior message into the asynchronous user behavior queue, otherwise, adding the user behavior message into the preset user behavior queue.

In an expanded embodiment, the method for recommending commodity objects includes the following steps for constructing the commodity similarity information matrix:

constructing an image characteristic similarity matrix between every two commodity objects based on the image characteristic information of the commodity pictures of the commodity objects in the commodity database, and storing similarity values between each commodity object and other commodity objects in the same row vector;

extracting classification labels of the commodity objects based on the text information of the commodity objects in the commodity database, determining a classification label similarity value between every two commodity objects, and constructing a text feature similarity matrix;

determining a corresponding relation according to the same two commodity objects, and carrying out linear fusion on the image characteristic similarity matrix and the similar values with the corresponding relation in the text characteristic similarity matrix to construct a commodity similarity information matrix, wherein the similar values between each commodity object and other commodity objects in the matrix are stored in the same row vector;

and sorting the same row vector in the commodity similarity information matrix according to the size of the similarity value.

A commodity object recommending apparatus adapted to one of the objects of the present application includes: the system comprises a message enlisting module, a message dequeuing module, a similar matching module and a commodity recommending module, wherein the message enlisting module is used for receiving a user behavior message submitted by a client in response to the behavior of a user accessing a commodity object for sale and adding the user behavior message into a behavior message queue; the message dequeuing module is used for monitoring the user behavior message dequeued from the behavior message queue and acquiring the commodity object on sale pointed by the user behavior message; the similarity matching module is used for inquiring and acquiring candidate commodity objects similar to the commodity objects on sale from a preset commodity similarity information matrix to construct a commodity recommendation list, and the commodity recommendation list comprises target commodity objects selected from the candidate commodity objects; and the commodity recommendation module is used for responding to the commodity recommendation request of the target user and pushing a corresponding commodity recommendation list to the target user.

In a further embodiment, the affinity matching module comprises: the object acquisition submodule is used for acquiring the unique characteristic information of the commodity object on sale, and the unique characteristic information and the dimension label of the commodity similarity information matrix have one-to-one mapping relation; the similarity query submodule is used for querying a commodity similarity information matrix according to the unique characteristic information and determining a row vector corresponding to the commodity object on sale, wherein each element of the row vector stores a similarity value used for measuring the similarity between the commodity object on sale and a corresponding candidate commodity object; the candidate determining submodule is used for determining a plurality of candidate commodity objects of which the similarity values meet the similarity matching condition according to the row vectors; and the list construction submodule is used for constructing a commodity recommendation list and adding at least one candidate commodity object meeting the similar matching condition into the commodity recommendation list as a target commodity object.

In a specific embodiment, the candidate determination sub-module comprises: the vector sorting unit is used for sorting the row vectors corresponding to the commodity objects for sale according to the similar numerical values; the element optimization unit is used for optimizing a plurality of elements with the maximum similarity values from the sorted row vectors according to a preset similarity matching condition; and the element determining unit is used for determining a plurality of corresponding candidate commodity objects meeting the similar matching conditions according to the dimension labels of the optimized elements in the commodity similarity information matrix.

In an embodied embodiment, the list construction sub-module comprises: the popularity quoting unit is used for calling popularity reference information determined according to the access popularity of the commodity object to obtain commodity popularity data corresponding to a plurality of candidate commodity objects meeting the similar matching conditions; the candidate filtering unit is used for filtering candidate commodity objects with commodity popularity data lower than a preset threshold value from a plurality of candidate commodity objects meeting the similar matching conditions to obtain at least one residual target commodity object; and the list customizing unit is used for constructing a commodity recommendation list which stores the commodity abstract texts and the commodity pictures corresponding to the target commodity objects.

In an embodied embodiment, the list construction sub-module comprises: the visited determining unit is used for calling the historical order data of the target user providing the user behavior message to determine the purchased commodity object; a visited filtering unit, configured to filter the purchased commodity object from a plurality of candidate commodity objects that satisfy the similarity matching condition, and obtain at least one remaining target commodity object; and the list customizing unit is used for constructing a commodity recommendation list which stores the commodity abstract texts and the commodity pictures corresponding to the target commodity objects.

In a further embodiment, the message enqueuing module is configured to receive a user behavior message submitted by a client in response to a behavior of a user accessing an on-sale commodity object, determine whether a preset user behavior queue is in a congestion state, enable an asynchronous user behavior queue if the preset user behavior queue is in the congestion state, add the user behavior message to the asynchronous user behavior queue, and otherwise add the user behavior message to the preset user behavior queue.

In an expanded embodiment, the commodity object recommending device of the present application includes the following structure for constructing the commodity similarity information matrix: the image similarity module is used for constructing an image feature similarity matrix between every two commodity objects based on the image feature information of the commodity images of the commodity objects in the commodity database, so that the similarity value between each commodity object and other commodity objects is stored in the same row vector; the text similarity module is used for extracting the classification label of each commodity object based on the text information of the commodity objects in the commodity database, determining the classification label similarity value between every two commodity objects and constructing a text feature similarity matrix; the linear fusion module is used for determining a corresponding relation according to the same two commodity objects, and carrying out linear fusion on the image characteristic similarity matrix and the similar numerical values with the corresponding relation in the text characteristic similarity matrix to construct a commodity similarity information matrix, wherein the similar numerical values between each commodity object and other commodity objects in the matrix are stored in the same row vector; and the similarity sorting module is used for sorting the same row vector in the commodity similarity information matrix according to the size of the similarity value.

The computer device comprises a central processing unit and a memory, wherein the central processing unit is used for calling and running a computer program stored in the memory to execute the steps of the commodity object recommendation method.

A computer-readable storage medium storing a computer program implemented according to the method for recommending an object of merchandise according to another object of the present application in the form of computer-readable instructions, the computer program, when called by a computer, executing the steps included in the method.

A computer program product adapted to another object of the present application is provided, which includes a computer program/instructions, when the computer program/instructions are executed by a processor, to implement the steps of the commodity object recommendation method described in any one of the embodiments of the present application.

Compared with the prior art, the application has the following advantages:

the method utilizes a message queue mechanism to receive user behavior messages triggered in real time when a user visits commodity objects, after the messages are queued and listed through a message queue, a commodity recommendation list corresponding to the commodity objects visited by the user and contained in the user behavior messages is generated in time at the background by a system, target commodity objects in the commodity recommendation list are derived from a pre-constructed commodity similarity information matrix, and the commodity similarity information matrix stores similarity data between the commodity objects and other commodity objects, so that the method can quickly determine the target commodity objects similar to the commodity objects visited by the user just now from the commodity similarity information matrix and recommend the target commodity objects to the user, the user can obtain similar commodities corresponding to the commodity objects pointed by the user behaviors just implemented by the user, and the recommended commodities can more easily and closely meet the requirements of the user, the matching degree between the data generated by the instant behavior of the user and the recommended target commodity object is realized.

Drawings

The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

FIG. 1 is a schematic flow chart diagram illustrating an exemplary embodiment of a method for recommending merchandise objects according to the present application;

FIG. 2 is a flowchart illustrating a process of querying a commodity similarity information matrix according to an embodiment of the present application;

FIG. 3 is a flowchart illustrating a process of preferential treatment of candidate merchandise objects based on row vectors according to an embodiment of the present application;

FIG. 4 is a flowchart illustrating one process of constructing a recommendation list of merchandise in an embodiment of the present application;

FIG. 5 is a flowchart illustrating a second process for constructing a merchandise recommendation list according to an embodiment of the present application;

FIG. 6 is a flowchart illustrating a process of constructing a commodity similarity information matrix according to an embodiment of the present application;

FIG. 7 is a functional block diagram of an exemplary embodiment of a merchandise object recommendation device of the present application;

fig. 8 is a schematic structural diagram of a computer device used in the present application.

Detailed Description

Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary only for the purpose of explaining the present application and are not to be construed as limiting the present application.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combinations of one or more of the associated listed items.

It will be understood by those within the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

As will be appreciated by those skilled in the art, "client," "terminal," and "terminal device" as used herein include both devices that are wireless signal receivers, which are devices having only wireless signal receivers without transmit capability, and devices that are receive and transmit hardware, which have receive and transmit hardware capable of two-way communication over a two-way communication link. Such a device may include: cellular or other communication devices such as personal computers, tablets, etc. having single or multi-line displays or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service), which may combine voice, data processing, facsimile and/or data communication capabilities; a PDA (Personal Digital Assistant), which may include a radio frequency receiver, a pager, internet/intranet access, a web browser, a notepad, a calendar and/or a GPS (Global Positioning System) receiver; a conventional laptop and/or palmtop computer or other device having and/or including a radio frequency receiver. As used herein, a "client," "terminal device" can be portable, transportable, installed in a vehicle (aeronautical, maritime, and/or land-based), or situated and/or configured to operate locally and/or in a distributed fashion at any other location(s) on earth and/or in space. The "client", "terminal Device" used herein may also be a communication terminal, a web terminal, a music/video playing terminal, such as a PDA, an MID (Mobile Internet Device) and/or a Mobile phone with music/video playing function, and may also be a smart tv, a set-top box, and the like.

The hardware referred to by the names "server", "client", "service node", etc. is essentially an electronic device with the performance of a personal computer, and is a hardware device having necessary components disclosed by the von neumann principle such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, an output device, etc., a computer program is stored in the memory, and the central processing unit calls a program stored in an external memory into the internal memory to run, executes instructions in the program, and interacts with the input and output devices, thereby completing a specific function.

It should be noted that the concept of "server" as referred to in this application can be extended to the case of a server cluster. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided, and in physical space, the servers may be independent from each other but can be called through an interface, or may be integrated into one physical computer or a set of computer clusters. Those skilled in the art will appreciate this variation and should not be so limited as to restrict the implementation of the network deployment of the present application.

One or more technical features of the present application, unless expressly specified otherwise, may be deployed to a server for implementation by a client remotely invoking an online service interface provided by a capture server for access, or may be deployed directly and run on the client for access.

The neural network models referenced or potentially referenced in this application, unless specified in the clear, may be deployed either on a remote server and remotely invoked at the client, or directly invoked at the device-capable client. Those skilled in the art will appreciate that the device can be used as a model training device and a model operating device corresponding to the neural network model as long as the device operating resources are suitable. In some embodiments, when the client-side hardware execution system runs on the client-side, the corresponding intelligence of the client-side hardware execution system can be obtained through migration learning, so that the requirement on the hardware execution resources of the client-side is reduced, and the excessive occupation of the hardware execution resources of the client-side is avoided.

Various data referred to in the present application may be stored in a server remotely or in a local terminal device unless specified in the clear text, as long as the data is suitable for being called by the technical solution of the present application.

The person skilled in the art will know this: although the various methods of the present application are described based on the same concept so as to be common to each other, they may be independently performed unless otherwise specified. In the same way, for each embodiment disclosed in the present application, it is proposed based on the same inventive concept, and therefore, concepts of the same expression and concepts of which expressions are different but are appropriately changed only for convenience should be equally understood.

The embodiments to be disclosed herein can be flexibly constructed by cross-linking related technical features of the embodiments unless the mutual exclusion relationship between the related technical features is stated in the clear text, as long as the combination does not depart from the inventive spirit of the present application and can meet the needs of the prior art or solve the deficiencies of the prior art. Those skilled in the art will appreciate variations therefrom.

The commodity object recommendation method can be programmed into a computer program product and is realized by being deployed in terminal equipment and/or a server to run, so that a client can access an open user interface after the computer program product runs in a webpage program or application program mode to realize man-machine interaction.

Referring to fig. 1, in an exemplary embodiment, the method includes the steps of:

step S1100, receiving a user behavior message submitted by a client in response to the behavior of a user accessing an object of a commodity on sale, and adding the user behavior message into a behavior message queue:

in order to obtain the user behavior message, a buried point acquisition instruction can be implanted in an application program or a related webpage corresponding to the e-commerce platform, when a user accesses a certain commodity object for sale, a user behavior message is constructed in response to the access line and submitted to a server of the e-commerce platform with the technical scheme of the application.

The timing of the user triggering and constructing the user behavior message may be when the user clicks a certain commodity object for sale, or when the user enters a detailed page of the commodity object for sale, or when the user performs an ordering process on the commodity object for sale, which can be flexibly implemented by those skilled in the art.

The implementer for constructing the user behavior message can be implemented by a process where the user is located and an application program corresponding to the e-commerce platform is installed, or can be implemented by a server for triggering deployment of the e-commerce platform when the user accesses a webpage corresponding to the e-commerce platform.

The user behavior message generally includes the on-sale commodity object accessed by the user, which may be represented as SKU or SPU of the on-sale commodity object, or other unique characteristic information, so that the server of the present application may access relevant information of the on-sale commodity object in the e-commerce platform commodity database accordingly, for example, obtain commodity profile information or a commodity picture of the on-sale commodity object for constructing a commodity recommendation list.

And the user behavior messages submitted to the server of the application are added into a behavior message queue maintained by the server of the application, and are sequentially listed and consumed according to a preset queuing mechanism. It can be seen that the message queue can process user behavior messages constructed by triggering of all online users of the whole e-commerce platform, and serves the full platform, and is suitable for providing target commodity objects similar to the commodity objects on sale for any online users of the full platform.

In an embodiment of the application, in order to avoid response failure due to congestion of the behavior message queue, after receiving a user behavior message submitted by a client in response to a behavior of a user accessing an object selling goods, further judging whether the user behavior queue is in a congestion state, if so, starting an asynchronous user behavior queue, adding the user behavior message to the asynchronous user behavior queue, otherwise, adding the user behavior message to an original user behavior queue. By starting the asynchronous processing mechanism, concurrent user behavior messages can be responded and processed in time, so that the processing efficiency of recommending the target commodity object is improved.

Step S1200, monitoring the user behavior messages listed from the behavior message queue, and acquiring the commodity object on sale pointed by the user behavior messages:

the server processes each user behavior message by starting a message consumption process, each user behavior message is consumed by one consumption thread correspondingly, and the consumption threads analyze the user behavior messages and acquire the commodity objects which are accessed by the users and sold, so that the commodity recommendation logic is implemented later.

Step S1300, obtaining candidate commodity objects similar to the current commodity object from a preset commodity similarity information matrix to construct a commodity recommendation list, where the commodity recommendation list includes target commodity objects preferred from the candidate commodity objects:

the consumption thread of the application can call a pre-constructed commodity similarity information matrix for inquiring and acquiring candidate commodity objects similar to the commodity objects on sale.

The commodity similarity information matrix is constructed in advance, is essentially a similarity list, stores similarity values between every two commodity objects in the e-commerce platform, and can be quantitatively determined according to the similarity degree of text characteristic information and/or picture characteristic information between every two commodity objects, or even can be manually determined in some embodiments. Therefore, each commodity object corresponds to a plurality of other commodity objects which are highly similar to the commodity object and has a plurality of corresponding similar numerical values, and the related commodity objects which are similar to the given commodity object can be determined according to the given commodity object.

In order to facilitate positioning of the commodity object in the commodity similarity information matrix, the matrix may use the SKU and the SPU of the commodity object as the dimension labels, or may also establish a mapping relationship between the SKU and the dimension labels of the matrix of the SPU of the commodity object, in short, the commodity object similar to the SKU and the dimension labels of the matrix may be found from the matrix according to the unique characteristic information of the commodity object, and conversely, when a plurality of elements storing similar numerical values similar to the commodity object under sale are searched from the matrix, the specific commodity object should be determined according to the dimension coordinates of the elements.

Considering that each commodity object is actually similar to the same commodity, the commodity similarity information matrix can be independently constructed for each final classification in the final classification structure in the commodity category tree, so long as the final classification is distinguished when called.

Further, considering that the user generally only pays attention to a small number of similar commodity objects, the commodity similarity information matrix may only store a plurality of commodity objects with most matched similarity values in accordance with each commodity object, and those skilled in the art may also flexibly implement the method.

The plurality of commodity objects similar to the commodity object in sale are inquired and obtained from the commodity similarity information matrix, and may be all commodity objects similar to the commodity object in sale or a plurality of commodity objects with the highest similarity value, so that the determined commodity objects are candidate commodity objects, at least one candidate commodity object in the candidate commodity objects is taken as a target commodity object, and the target commodity object is added to the constructed commodity recommendation list.

It can be understood that, in response to each user behavior message, the server dynamically updates the commodity recommendation list corresponding to the user who triggered to submit the user behavior message, where one or more target recommended commodity objects included in the commodity recommendation list constitute similar commodities with the commodity objects on sale that are visited by the user when the user triggered to submit the user behavior message. The target recommended commodity object in the commodity recommendation list is dynamically updated according to the commodity object on sale accessed by the user, namely after the user accesses a new commodity object on sale again, the service logic of the technical scheme of the application is triggered, namely the target commodity object recommended for the user is correspondingly updated, so that the dynamic association between the user instant behavior event and the target commodity object recommended for the user is realized.

Step S1400, responding to the commodity recommendation request of the target user, and pushing a corresponding commodity recommendation list to the target user:

when a user switches to some page triggering acquisition of a commodity recommendation list on an application program or a webpage of a client, for example, a page containing a commodity object advertisement column or enters a special page for acquiring possible likeliness of the user, "guess you like", a commodity recommendation request is also triggered correspondingly, and after the server receives the commodity recommendation request, the latest commodity recommendation list is pushed to the user.

And after receiving the commodity recommendation list, the client of the user analyzes the commodity recommendation list and displays the commodity recommendation list on a graphical user interface to achieve the purpose of commodity recommendation.

It should be understood that, in the exemplary embodiment, the present application utilizes a message queue mechanism to receive a user behavior message triggered in real time when a user accesses a commodity object, and after queuing and listing through a message queue, a system generates a commodity recommendation list corresponding to the commodity object accessed by the user and included in the user behavior message in time at the background, a target commodity object in the commodity recommendation list is derived from a pre-constructed commodity similarity information matrix, and the commodity similarity information matrix stores similarity data between the commodity object and other commodity objects, so that the present application can quickly determine a target commodity object similar to the commodity object accessed by the user from the commodity similarity information matrix and recommend the target commodity object to the user, and the user can obtain similar commodities corresponding to the commodity object pointed by the user behavior just implemented by the user, and the recommended commodities more easily and closely meet the user requirements, the matching degree between the data generated by the instant behavior of the user and the recommended target commodity object is realized.

Referring to fig. 2, in a further embodiment, the step S1300 of querying and obtaining candidate commodity objects similar to the commodity object in sale from a preset commodity similarity information matrix to construct a commodity recommendation list includes the following steps:

step S1310, obtaining the unique feature information of the on-sale commodity object, where the unique feature information and the dimension label of the commodity similarity information matrix have a one-to-one mapping relationship:

as mentioned above, the unique characteristic information of the goods-on-sale object may be the SKU or SPU thereof, in this embodiment, the unique characteristic information corresponding to the goods-on-sale object is further obtained by querying a mapping table, and the unique characteristic information is a dimension label in the goods similarity information matrix. The mapping table stores the mapping relation data between the SKU or SPU of all the commodity objects and the dimension label of the commodity similarity information matrix, so that the corresponding dimension label of the commodity object in the commodity similarity information matrix can be determined according to the mapping table, and the dimension label is used as the unique characteristic information of the commodity object in sale so as to directly implement query operation in the commodity similarity information matrix.

Step S1320, querying a commodity similarity information matrix according to the unique feature information, and determining a row vector corresponding to the commodity object on sale, where each element of the row vector stores a similarity value for measuring a similarity between the commodity object on sale and a corresponding candidate commodity object:

the commodity similarity information matrix can be regarded as a vector matrix, so that each row vector of the commodity similarity information matrix correspondingly stores all similar values of a commodity object which is sold and corresponds to a plurality of other commodity objects, namely each element in the row vector stores the similar value between the commodity object which is sold and a corresponding commodity object, the row coordinate and the column coordinate of the matrix are marked by dimension labels, and the dimension labels are mapped with SKUs or SPUs of the commodity objects through the mapping table, so that the dimension labels are determined, the corresponding commodity objects are actually determined, and vice versa, and the row vectors of the corresponding commodity objects in the matrix can be determined according to one dimension label.

In an embodiment of an alternative implementation of the present application, the merchandise similarity information matrix may also be represented in other ways, for example, stored in a Key-Value form of a Redis database, wherein information pairs consisting of SKU, SPU or dimension tag of each merchandise object being sold as the Value of its Key field (Key), SKU, SPU or dimension tag of the merchandise object being compared with the Key field (Key), and similar values obtained by comparing the SKU, SPU or dimension tag of the merchandise object and the two can have equivalent effects to the present embodiment, and those skilled in the art should understand that the scope covered by the spirit of the present application should not be limited thereby.

Step S1330, determining, according to the row vector, a plurality of candidate commodity objects whose similarity values satisfy a similarity matching condition:

as described above, each of the line vectors of the commodity objects on sale includes a plurality of elements, and the dimension tag corresponding to each element points to a different commodity object, so that the commodity objects corresponding to all the elements of the line vector can be determined as candidate commodity objects satisfying the similar matching condition with the commodity objects on sale. In this case, the matching condition refers to that all the similarity values in the row vector may be, and generally corresponds to a case where the similarity values of a limited number of commodity objects have been screened out according to a certain similarity threshold value during the construction of the commodity similarity information matrix to construct the row vector.

In an alternative embodiment, a higher similarity threshold may be set, or a filtering number may be set as a matching condition for elements in the row vector, so as to further reduce the elements in the row vector, so as to determine a limited plurality of candidate objects, which will be further described in the following embodiments.

Step S1340, constructing a commodity recommendation list, and adding at least one candidate commodity object meeting the similar matching condition as a target commodity object into the commodity recommendation list:

after the candidate commodity objects are obtained, at least one or more of the candidate commodity objects can be added to an emptied commodity recommendation list as target commodity objects, and the construction of the commodity recommendation list is completed. In this process, the candidate commodity object may be further preferred, and for this reason, it will be exemplified later and not shown here.

The embodiment provides a specific implementation scheme for querying and acquiring candidate commodity objects similar to the commodity object in sale from the commodity similarity information matrix and determining the target commodity object to construct the commodity recommendation list from the candidate commodity objects, and it can be seen that by querying the pre-constructed commodity similarity information matrix, massive commodity data do not need to be called by a commodity database of an instant calling platform for real-time comparison, so that the retrieval efficiency of the similar commodity objects can be greatly improved, the response time of a server is shortened, and the human-computer interaction experience of a client is improved.

Referring to fig. 3, in an embodiment of the present invention, the step S1330 of determining the candidate merchandise objects with the similarity values satisfying the similarity matching condition according to the row vector includes the following steps:

step S1331, sorting the row vectors corresponding to the goods on sale objects according to the similarity values:

in some embodiments, in the commodity similarity information matrix, the inside of each row vector is sorted in advance, so this step is mainly implemented when the elements in the row vector are not sorted according to the similar values, and the elements in the row vector are sorted according to the similar values, mainly to facilitate further selecting a limited plurality of candidate commodity objects corresponding to all the elements, so as to implement the simplified selection of the target commodity object.

Step S1332, selecting a plurality of elements with the maximum similarity value from the sorted row vectors according to a preset similarity matching condition:

and adopting a TopN strategy, and giving a fixed value N of the natural numbers to indicate that N elements are required to be selected from the sorted elements so as to screen out N corresponding candidate objects. And selecting the candidate commodity objects according to the direction from the maximum similarity value to the minimum similarity value, and finally selecting the candidate commodity objects corresponding to the N elements with the maximum similarity values.

Step S1333, determining a plurality of corresponding candidate commodity objects meeting the similarity matching condition according to the dimension label of the preferred element in the commodity similarity information matrix:

as mentioned above, the commodity similarity information matrix is a row-column matrix, which is essentially a list, and thus has coordinate information corresponding to rows and columns thereof, and each row or column takes the dimension label indicating the commodity object as an index, so that after the elements are selected, only a plurality of corresponding candidate commodity objects are determined according to the dimension labels indicating the column coordinates of the elements, and naturally, the candidate commodity objects all satisfy the similarity matching condition.

In this embodiment, a candidate commodity object similar to the commodity object on sale can be further optimized based on the similarity value, so as to adapt to the situation that the candidate commodity object is too many, thereby reducing the data amount to be processed subsequently and ensuring the processing and response efficiency of the server.

Referring to fig. 4, in another embodiment, the step S1340 of constructing a product recommendation list, and adding at least one candidate product object satisfying the similar matching condition as a target product object to the product recommendation list includes the following steps:

step S1341, invoking popularity reference information determined according to the access popularity of the commodity object, and obtaining commodity popularity data corresponding to a plurality of candidate commodity objects satisfying the similarity matching condition:

the method can further prepare an access popularity statistical data table, and the access popularity statistical data table is used for counting the access popularity of each commodity object in the latest statistical time period according to the user access behavior data and/or other ranking data recalled aiming at the commodity objects and/or other similar data, and represents the popularity of the commodity object as the name suggests, so that the candidate commodity object can be further selected, and the selected candidate commodity object can meet the user requirements more easily.

In order to construct a commodity recommendation list, the present embodiment first calls the popularity reference information storing the access popularity, and then queries commodity popularity data corresponding to a plurality of candidate commodity objects determined by the present application from the popularity reference information.

Step S1342, filtering out candidate commodity objects whose commodity popularity data is lower than a preset threshold from the plurality of candidate commodity objects satisfying the similarity matching condition, and obtaining at least one remaining target commodity object:

the method comprises the steps of utilizing a preset threshold to represent an access heat threshold of candidate commodity objects, then screening the candidate commodity objects, deleting the candidate commodity objects of which the commodity heat data are lower than the preset threshold, and enabling the rest candidate commodity objects to be selected candidate commodity objects, namely target commodity objects required by constructing a commodity recommendation list according to the method.

Step S1343, constructing a commodity recommendation list, where the commodity recommendation list stores the commodity abstract text and the commodity picture corresponding to each target commodity object:

if a commodity recommendation list is not generated for the user before, an empty table can be created first, if a commodity recommendation list corresponding to the user already exists, the empty table can be obtained by emptying the commodity recommendation list, and on the basis, each target commodity object is added into the empty table.

When the target commodity object is added into the empty table, the summary information of the target commodity object can be constructed in advance, specifically, the summary information can be formed by inquiring and calling corresponding commodity summary texts and commodity pictures from a commodity database of an e-commerce platform according to the SKU, SPU or other unique characteristic information of the target commodity object, and then the summary information is stored in the empty table.

After receiving the commodity recommendation list, the client analyzes and displays the commodity recommendation list correspondingly, and then the summary information corresponding to each target commodity object can be seen.

In the embodiment, the commodity heat data of the commodity object is quoted to select the preferred candidate commodity object, the candidate commodity object with lower visit heat is filtered from the candidate commodity object, and the target commodity object recommended to the user is selected.

Referring to fig. 5, in another embodiment, the step S1340 of constructing a product recommendation list, and adding at least one candidate product object satisfying the similar matching condition as a target product object to the product recommendation list includes the following steps:

step S1341', call the historical order data of the target user who provided the user behavior message to determine the purchased commodity object:

to avoid recommending items for the user that the user has purchased, the user's historical order data may be invoked to determine the items objects that the user has purchased.

According to the method and the device, the message thread consumes the user behavior message, so that the message thread can directly determine the user to which the user behavior message belongs, and the historical order data of the target user is acquired to determine the purchased commodity object of the target user by taking the user behavior message as the target user.

Step S1342', filtering out the purchased commodity object from the plurality of candidate commodity objects satisfying the similarity matching condition, and obtaining at least one remaining target commodity object:

the purchased commodity objects are directly filtered from the candidate commodity objects determined previously, and the remaining candidate commodity objects are the target commodity objects selected in the embodiment.

Step S1343', a product recommendation list is constructed, where the product recommendation list stores the product abstract text and the product picture corresponding to each target product object:

when the target commodity object is added into the empty table, the summary information of the target commodity object can be constructed in advance, specifically, the summary information can be formed by inquiring and calling corresponding commodity summary texts and commodity pictures from a commodity database of an e-commerce platform according to the SKU, SPU or other unique characteristic information of the target commodity object, and then the summary information is stored in the empty table.

After receiving the commodity recommendation list, the client analyzes and displays the commodity recommendation list correspondingly, and then the summary information corresponding to each target commodity object can be seen.

In the embodiment, the historical order data of the target user is quoted to select the preferred candidate commodity object, the commodities purchased by the target user are filtered out, and the target commodity object not purchased by the target user is selected, so that invalid recommendation data generated for the target user is avoided, the matching degree between the target commodity object and the user access behavior can be improved, and the user requirement can be met more accurately.

Referring to fig. 6, in an expanded embodiment, in order to implement the pre-construction of the product similarity information matrix, the product object recommendation method of the present application includes the following steps for constructing the product similarity information matrix:

step S2100, constructing an image feature similarity matrix between every two commodity objects based on the image feature information of the commodity pictures of the commodity objects in the commodity database, and enabling similarity values between each commodity object and other commodity objects to be stored in the same row vector:

the image feature information of the commodity picture of the commodity object in the commodity database of the e-commerce platform can be extracted by utilizing a pre-trained neural network model, such as Resnet, Effecent network model and the like, and each commodity object is preferably selected as one commodity picture. The method comprises the steps of constructing vector indexes of all image characteristic information by utilizing a Faiss or Annoy framework, calculating a similarity value between every two commodity objects by applying a cosine similarity algorithm based on the image characteristic information, and obtaining an image characteristic similarity matrix, wherein each element in the matrix represents the similarity between every two commodity objects, and each row vector comprises the similarity values between the commodity object pointed by the dimension label corresponding to the row vector and all other commodity objects.

In an optimized embodiment, elements in each row vector in the image feature similarity matrix may be optimized, and each row vector optimizes a rated plurality of elements with the largest similarity value, so that each commodity object only retains the similarity values of the rated plurality of commodity objects similar to the commodity object.

Step S2200, extracting the classification label of each commodity object based on the text information of the commodity object in the commodity database, determining the classification label similarity value between every two commodity objects, and constructing as a text feature similarity matrix:

the method can adopt NLP technology to extract text features of information such as titles, descriptions and attributes of commodity objects in a commodity database of an e-commerce platform, data cleaning can be carried out before the text features are extracted, a plurality of classification labels corresponding to each commodity object are determined on the basis of the text features extraction, and therefore the commodity object corresponding to each classification label is determined.

The classification labels owned by every two commodity objects may coincide, so that intersection and comparison indexes determined by the coincidence of the two commodity objects based on the classification labels can be converted into similarity values between every two commodity objects, and a text feature similarity matrix is constructed by referring to the structure of the image feature similarity matrix.

Step S2300, determining a corresponding relation according to the same two commodity objects, and linearly fusing the image feature similarity matrix and the similar numerical values with the corresponding relation in the text feature similarity matrix to construct a commodity similarity information matrix, wherein in the matrix, the similar numerical values between each commodity object and other commodity objects are stored in the same row vector:

the text feature similarity matrix and the image feature similarity matrix have the same structure, so that all elements of the text feature similarity matrix and the image feature similarity matrix correspond one to one and indicate similar numerical values with different properties between the same pair of commodity objects, on the basis, the text feature similarity matrix and the image feature similarity matrix are subjected to vector addition to calculate the mean value to obtain a commodity similarity information matrix, and linear fusion of the two similarity matrices can be realized. Of course, the way of implementing linear fusion can be flexibly changed by those skilled in the art, for example, weighted summation between vectors and then averaging, or direct vector addition can be used.

Step S2400, aiming at the same row vector in the commodity similarity information matrix, sorting according to the size of the similarity value:

finally, on the basis of the commodity similarity information matrix, the commodity similarity information matrix can be converted into a data storage structure such as Redis, so that the dimension labels on the row coordinates in the commodity similarity information matrix can be used for the values of the Key domains in Key-Value, and each element of the whole row vector is associated with the dimension labels of the corresponding column coordinates, and after being combined one by one, the element can be stored as the Value of the Key-Value domain.

The present embodiment provides an example of constructing the commodity similarity information matrix, and of course, those skilled in the art may change various embodiments according to the principle disclosed in this example, as long as the commodity similarity information matrix required by the present application can be constructed, and the commodity similarity information matrix of the present application is merely a name, which is expressed in a data storage form, and is not limited to a matrix form in a mathematical sense, but should be understood as any storage form including various types of databases.

The commodity similarity information matrix provided by the embodiment combines two information sources of the commodity picture and the text information of the commodity object as reference information of the similarity information, wherein the commodity picture determines the similarity between the commodity objects on the image level, the text information determines the similarity between the commodity objects on the text semantic level, and the two information are finally combined together, so that the commodity similarity information matrix can more accurately represent the actual similarity between the commodity objects, the corresponding relation between the target commodity object determined by the application and the commodity object sold in the user behavior message is closer, and the matching efficiency of matching similar commodities for users is greatly improved.

The commodity recommendation method is wide in application scene, for example, when a user enters a columniform webpage like 'guess you like' to check products which possibly meet the potential needs of the user, the commodity recommendation list can be created for the user by adopting the method, and finally commodities in the commodity recommendation list are displayed in the columniform webpage. For another example, the user enters a certain webpage, and the webpage has a commodity advertisement column, and the commodity advertisement column may also show the target commodity object recommended by the application for the user. For another example, when a user switches from one live broadcast room selling some commodity object in sale to another live broadcast room, the method can be used for determining the target commodity object which is sold in the current live broadcast room and is similar to the commodity object in sale. Therefore, the technical scheme of the application can be applied as long as the target commodity object is recommended to the user, and the user requirements are met.

Referring to fig. 7, a commodity object recommendation apparatus provided in the present application, adapted to a commodity object recommendation method of the present application for functional deployment, includes: the system comprises a message enlisting module 1100, a message dequeuing module 1200, a similarity matching module 1300 and a commodity recommending module 1400, wherein the message enlisting module 1100 is used for receiving a user behavior message submitted by a client in response to a behavior of a user accessing a commodity object for sale and adding the user behavior message into a behavior message queue; the message dequeuing module 1200 is configured to monitor a user behavior message dequeued from the behavior message queue, and acquire an on-sale commodity object pointed by the user behavior message; the similarity matching module 1300 is configured to query and acquire a candidate commodity object similar to the current commodity object from a preset commodity similarity information matrix to construct a commodity recommendation list, where the commodity recommendation list includes a target commodity object preferred from the candidate commodity objects; the commodity recommendation module 1400 is configured to respond to the commodity recommendation request of the target user, and push a corresponding commodity recommendation list to the target user.

In a further embodiment, the affinity matching module 1300 comprises: the object acquisition submodule is used for acquiring the unique characteristic information of the commodity object on sale, and the unique characteristic information and the dimension label of the commodity similarity information matrix have one-to-one mapping relation; the similarity query submodule is used for querying a commodity similarity information matrix according to the unique characteristic information and determining a row vector corresponding to the commodity object on sale, wherein each element of the row vector stores a similarity value used for measuring the similarity between the commodity object on sale and a corresponding candidate commodity object; the candidate determining submodule is used for determining a plurality of candidate commodity objects of which the similarity values meet the similarity matching condition according to the row vectors; and the list construction submodule is used for constructing a commodity recommendation list and adding at least one candidate commodity object meeting the similar matching condition into the commodity recommendation list as a target commodity object.

In a specific embodiment, the candidate determination sub-module comprises: the vector sorting unit is used for sorting the row vectors corresponding to the commodity objects for sale according to the similar numerical values; the element optimization unit is used for optimizing a plurality of elements with the maximum similarity values from the sorted row vectors according to a preset similarity matching condition; and the element determining unit is used for determining a plurality of corresponding candidate commodity objects meeting the similar matching conditions according to the dimension labels of the optimized elements in the commodity similarity information matrix.

In an embodied embodiment, the list construction sub-module comprises: the popularity quoting unit is used for calling popularity reference information determined according to the access popularity of the commodity object to obtain commodity popularity data corresponding to a plurality of candidate commodity objects meeting the similar matching conditions; the candidate filtering unit is used for filtering candidate commodity objects with commodity popularity data lower than a preset threshold value from a plurality of candidate commodity objects meeting the similar matching conditions to obtain at least one residual target commodity object; and the list customizing unit is used for constructing a commodity recommendation list which stores the commodity abstract texts and the commodity pictures corresponding to the target commodity objects.

In an embodied embodiment, the list construction sub-module comprises: the visited determining unit is used for calling the historical order data of the target user providing the user behavior message to determine the purchased commodity object; a visited filtering unit, configured to filter the purchased commodity object from a plurality of candidate commodity objects that satisfy the similarity matching condition, and obtain at least one remaining target commodity object; and the list customizing unit is used for constructing a commodity recommendation list which stores the commodity abstract texts and the commodity pictures corresponding to the target commodity objects.

In a further embodiment, the message enqueuing module 1100 is configured to receive a user behavior message submitted by a client in response to a behavior of a user accessing an on-sale commodity object, determine whether a preset user behavior queue is in a congestion state, enable an asynchronous user behavior queue if the preset user behavior queue is in the congestion state, add the user behavior message to the asynchronous user behavior queue, and otherwise add the user behavior message to the preset user behavior queue.

In an expanded embodiment, the commodity object recommending device of the present application includes the following structure for constructing the commodity similarity information matrix: the image similarity module is used for constructing an image feature similarity matrix between every two commodity objects based on the image feature information of the commodity images of the commodity objects in the commodity database, so that the similarity value between each commodity object and other commodity objects is stored in the same row vector; the text similarity module is used for extracting the classification label of each commodity object based on the text information of the commodity objects in the commodity database, determining the classification label similarity value between every two commodity objects and constructing a text feature similarity matrix; the linear fusion module is used for determining a corresponding relation according to the same two commodity objects, and carrying out linear fusion on the image characteristic similarity matrix and the similar numerical values with the corresponding relation in the text characteristic similarity matrix to construct a commodity similarity information matrix, wherein the similar numerical values between each commodity object and other commodity objects in the matrix are stored in the same row vector; and the similarity sorting module is used for sorting the same row vector in the commodity similarity information matrix according to the size of the similarity value.

In order to solve the technical problem, an embodiment of the present application further provides a computer device. As shown in fig. 8, the internal structure of the computer device is schematically illustrated. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected by a system bus. The computer readable storage medium of the computer device stores an operating system, a database and computer readable instructions, the database can store control information sequences, and the computer readable instructions, when executed by the processor, can cause the processor to implement a commodity object recommendation method. The processor of the computer device is used for providing calculation and control capability and supporting the operation of the whole computer device. The memory of the computer device may have stored therein computer readable instructions, which, when executed by the processor, may cause the processor to perform the merchandise object recommendation method of the present application. The network interface of the computer device is used for connecting and communicating with the terminal. Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In this embodiment, the processor is configured to execute specific functions of each module and its sub-module in fig. 7, and the memory stores program codes and various data required for executing the modules or the sub-modules. The network interface is used for data transmission to and from a user terminal or a server. The memory in this embodiment stores program codes and data necessary for executing all modules/sub-modules in the commodity object recommending apparatus of the present application, and the server can call the program codes and data of the server to execute the functions of all sub-modules.

The present application also provides a storage medium storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to perform the steps of the method for recommending merchandise objects according to any of the embodiments of the present application.

The present application also provides a computer program product comprising computer programs/instructions which, when executed by one or more processors, implement the steps of the method for merchandise object recommendation described in any of the embodiments of the present application.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments of the present application can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when the computer program is executed, the processes of the embodiments of the methods can be included. The storage medium may be a computer-readable storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).

In summary, the method and the device for matching the commodity object can generate the target commodity object similar to the commodity object on sale in real time according to the behavior of the user for accessing the commodity object on sale, are accurate in matching, and are suitable for various application scenes.

Those of skill in the art will appreciate that the various operations, methods, steps in the processes, acts, or solutions discussed in this application can be interchanged, modified, combined, or eliminated. Further, other steps, measures, or schemes in various operations, methods, or flows that have been discussed in this application can be alternated, altered, rearranged, broken down, combined, or deleted. Further, steps, measures, schemes in the prior art having various operations, methods, procedures disclosed in the present application may also be alternated, modified, rearranged, decomposed, combined, or deleted.

The foregoing is only a partial embodiment of the present application, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the principle of the present application, and these modifications and decorations should also be regarded as the protection scope of the present application.