CN113807926A

CN113807926A - Recommendation information generation method and device, electronic equipment and computer readable medium

Info

Publication number: CN113807926A
Application number: CN202111126109.6A
Authority: CN
Inventors: 李鹏; 姚亚飞; 李勇; 赫阳; 彭长平; 林战刚; 邵京平
Original assignee: Beijing Wodong Tianjun Information Technology Co Ltd
Current assignee: Beijing Wodong Tianjun Information Technology Co Ltd
Priority date: 2021-09-26
Filing date: 2021-09-26
Publication date: 2021-12-17

Abstract

The embodiment of the disclosure discloses a recommendation information generation method, a recommendation information generation device, an electronic device and a computer readable medium. One embodiment of the method comprises: in response to the detection of the object browsing operation of a target user, acquiring an object recommendation model of the target user, wherein the object recommendation model generates recommended object information through a historical object layering portrait of the target user; determining at least one piece of initial recommended item information of the target user based on the item browsing operation and the item recommendation model; and generating recommendation information based on the attribute information of the at least one piece of initial recommended article information. The embodiment reduces the data processing time, reduces the consumption of data processing capacity, and improves the accuracy of recommending information to the user.

Description

Recommendation information generation method and device, electronic equipment and computer readable medium

Technical Field

The embodiment of the disclosure relates to the technical field of computers, in particular to a recommendation information generation method, a recommendation information generation device, electronic equipment and a computer readable medium.

Background

With the development of information technology, obtaining information through a network becomes an important way for a user to obtain information. In order to improve the efficiency of obtaining information by the user, the information server can push information to the user according to portrait information of the user or historical browsing records of the user.

The prior art has the following defects when pushing information to a user:

information servers typically push information to users while they are browsing for relevant information. In practice, the information on the information website is massive, in order to search for information that may be of interest to the user from the massive information, the information server needs to perform data processing quickly by means of a complex information search model, a large amount of data processing capacity and time need to be consumed in the process, and the accuracy of the finally obtained recommendation information is not high.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

Some embodiments of the present disclosure propose a recommendation information generation method, apparatus, electronic device, and computer readable medium to solve the technical problems mentioned in the background section above.

In a first aspect, some embodiments of the present disclosure provide a recommendation information generating method, including: in response to the detection of the object browsing operation of a target user, acquiring an object recommendation model of the target user, wherein the object recommendation model generates recommended object information through a historical object layering portrait of the target user; determining at least one piece of initial recommended item information of the target user based on the item browsing operation and the item recommendation model; and generating recommendation information based on the attribute information of the at least one piece of initial recommended article information.

In a second aspect, some embodiments of the present disclosure provide a recommendation information generating apparatus, including: an item recommendation model acquisition unit configured to acquire an item recommendation model of a target user in response to detection of an item browsing operation of the target user, the item recommendation model generating recommended item information from a historical item hierarchy representation of the target user; an initial recommended item information determination unit configured to determine at least one piece of initial recommended item information of the target user based on the item browsing operation and an item recommendation model; and a recommendation information generation unit configured to generate recommendation information based on the attribute information of the at least one piece of initially recommended item information.

In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, which when executed by one or more processors, cause the one or more processors to implement the method described in any of the implementations of the first aspect.

In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium on which a computer program is stored, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect.

The above embodiments of the present disclosure have the following beneficial effects: the accuracy of the recommendation information obtained by the recommendation information generation method of some embodiments of the disclosure is improved. Specifically, the reason why the accuracy of the recommendation information is not high is that: the item recommendation model consumes a lot of data processing power and time. Based on this, the recommendation information generation method of some embodiments of the present disclosure obtains the item recommendation model of the target user when detecting the item browsing operation of the target user. The historical object layering portrait describes the object attributes in a layering mode, and the objects are divided according to the attributes, so that data dimensionality and data processing amount are greatly reduced. Then, determining at least one piece of initial recommended item information of the target user through item browsing operation and an item recommendation model; and finally, generating recommendation information based on the attribute information of the at least one piece of initial recommended article information. Due to the adoption of the layered portrait of the historical article, the data processing time is reduced, the consumption of data processing capacity is reduced, the accuracy of recommending information to a user can be improved through the application, and the efficiency of online information pushing is ensured.

Drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and elements are not necessarily drawn to scale.

Fig. 1 is a schematic diagram of an application scenario of a recommendation information generation method of some embodiments of the present disclosure;

FIG. 2 is a flow diagram of some embodiments of a recommendation information generation method according to the present disclosure;

FIG. 3 is a flow diagram of some embodiments of an item recommendation model training method according to the present disclosure;

FIG. 4 is a schematic illustration of a hierarchical representation of a historical item according to the present disclosure;

FIG. 5 is a flow diagram of further embodiments of a recommendation information generation method according to the present disclosure;

FIG. 6 is a schematic diagram of attribute information aggregation to be processed;

FIG. 7 is a schematic block diagram of some embodiments of a recommendation information generating device according to the present disclosure;

FIG. 8 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.

It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings. The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.

It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.

It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 is a schematic diagram of an application scenario of a recommendation information generation method according to some embodiments of the present disclosure.

As shown in fig. 1, a target user may log in information website browsing information on a terminal device 101 (which may be a mobile phone, a tablet computer, etc.). When the information server 102 of the information website detects the article browsing operation of the target user, the target user can be considered to have a need for related information. At this time, the information server 102 may obtain the item recommendation model of the target user, and determine at least one piece of initial recommended item information of the target user through the item browsing operation and the item recommendation model; the item recommendation model generates recommended item information through a historical item layering portrait of a target user. The historical object layering portrait describes the object attribute in a layering mode, and the objects are divided according to the attribute, so that the data dimension and the data processing amount are greatly reduced. Then, generating recommendation information (such as recommendation information 1, recommendation information 2, recommendation information 3 and recommendation information 4 in fig. 1) by the attribute information of the initial recommended article information; finally, the information server 102 may transmit the recommendation information to the terminal device 101. Due to the adoption of the layered portrait of the historical article, the data processing time is reduced, the consumption of data processing capacity is reduced, the timeliness of recommending information to the user is improved, and the efficiency of online information pushing is ensured.

It should be understood that the number of terminal apparatuses 101 and information servers 102 in fig. 1 is merely illustrative. There may be any number of terminal devices 101 and information servers 102, as desired for implementation.

With continued reference to fig. 2, fig. 2 illustrates a flow 200 of some embodiments of a recommendation information generation method according to the present disclosure. The recommendation information generation method comprises the following steps:

step 201, in response to detecting the item browsing operation of the target user, acquiring an item recommendation model of the target user.

In some embodiments, an executing subject (e.g., the information server 102 shown in fig. 1) of the recommendation information generating method may detect an item browsing operation of a target user through a wired connection manner or a wireless connection manner, and after the item browsing operation is detected, the executing subject may obtain an item recommendation model of the target user. It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G/5G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, a uwb (ultra Wide band) connection, and other wireless connection means now known or developed in the future.

In practice, the user can log in information website browsing information (for example, browsing mobile phone information, computer information, etc.) through the terminal device 101. When the execution subject detects an item browsing operation of the target user (for example, clicking some item information, opening an item information page, etc.), it may be considered that the target user has a need for related information. At this time, the executing agent may obtain a pre-trained item recommendation model for the target user. The item recommendation model can generate recommended item information through the historical item layering portrait of the target user. The historical object layered portrait classifies and screens the objects through the layered index of the object attributes, so that the data processing amount and the searching time for searching the objects are greatly reduced, and the timeliness for acquiring the recommendation information is improved.

Step 202, determining at least one initial recommended item credit of the target user based on the item browsing operation and the item recommendation model.

In some embodiments, the execution subject may convert the item browsing operation of the target user into input information of an item recommendation model, so that the item recommendation model processes the input information and outputs at least one piece of initial recommended item information. For example, the execution agent may convert the item browsing operation into a vector or information sequence, and then input the vector or information sequence into the item recommendation model to obtain the initial recommended item credit.

Step 203, generating recommendation information based on the attribute information of the at least one piece of initial recommended item information.

In some embodiments, after the initial recommended item information is obtained, in order to further improve the accuracy of the recommended information, the execution subject may perform operations such as correcting and screening on the initial recommended item information through the attribute information of the initial recommended item information, so as to obtain the recommended information, thereby improving the accuracy and the effectiveness of the recommended information. For example, the execution subject may screen out a plurality of pieces of main attribute information from the attribute information of the initial recommended item information, and then search information within a set time period as recommendation information through the main attribute information; information satisfying a plurality of pieces of main attribute information at the same time may be used as recommendation information.

According to the recommendation information obtained by the recommendation information generation method disclosed by some embodiments of the disclosure, the timeliness of the recommendation information is improved. Specifically, the reason why the timeliness of the recommendation information is not high is that: the item recommendation model consumes a lot of data processing power and time. Based on this, the recommendation information generation method of some embodiments of the present disclosure obtains the item recommendation model of the target user when detecting the item browsing operation of the target user. The historical object layering portrait describes the object attributes in a layering mode, and the objects are divided according to the attributes, so that data dimensionality and data processing amount are greatly reduced. Then, determining at least one piece of initial recommended item information of the target user through item browsing operation and an item recommendation model; and finally, generating recommendation information based on the attribute information of the at least one piece of initial recommended article information. Due to the adoption of the layered portrait of the historical article, the data processing time is reduced, the consumption of data processing capacity is reduced, the accuracy of recommending information to a user can be improved through the application, and the efficiency of online information pushing is ensured.

With continued reference to fig. 3, fig. 3 illustrates a flow 300 of some embodiments of an item recommendation model training method according to the present disclosure. The item recommendation model training method comprises the following steps:

step 301, obtaining a historical article browsing information set in the time period set by the target user.

The information of interest may be different for different users. In order to obtain the recommended item information of the target user in a targeted manner, the execution subject (e.g., the information server 102 shown in fig. 1) may obtain a set of historical item browsing information within a time period set by the target user. The historical item browsing information in the historical item browsing information set can be item information previously browsed on the information website by the user.

And step 302, constructing a hierarchical portrait of the historical item through the historical item browsing information set.

In order to analyze the historical item browsing information and determine the information focused by the target user, the execution subject can construct a historical item layered portrait by the historical item browsing information set. The historical object layered portrait can comprise a plurality of layered attribute information, and each layer of attribute information belongs to a corresponding attribute layer. For example, a hierarchical representation of historical items may include attribute information for an XX cell phone, a YY brand, a ZZ store, and so forth. The attribute layer corresponding to the XX mobile phone attribute information can be an electronic product; the attribute layer corresponding to the YY brand can be brand information; the attribute layer corresponding to the ZZ store may be store information or the like.

In some optional implementations of some embodiments, the constructing a hierarchical representation of historical items through the set of historical item browsing information may include: and attribute aggregation is carried out on the historical item browsing information in the historical item browsing information set through a preset hierarchical index, so as to obtain a historical item hierarchical portrait.

The execution main body can acquire a preset hierarchical index in advance, attribute division is carried out on historical item browsing information in the historical item browsing information set through the hierarchical index, and then the attributes after attribute division are clustered, so that a historical item hierarchical portrait is obtained. As shown in fig. 4, the preset hierarchical index may include an attribute level 1, an attribute level 2, an attribute level 3, and an attribute level 4. The attribute layer 1 may represent item type information; the attribute layer 2 may represent item category information; the attribute layer 3 may represent item brand information; the attribute layer 4 may represent commodity store name information. Attribute layer 1 may have K attributes; the attribute layer 2 may have M attributes; the attribute layer 3 may have N attributes; the attribute layer 4 may have T attributes. Each attribute layer may have a plurality of attribute nodes represented by circles, and each attribute node may represent specific attribute information. For example, when attribute layer 1 represents item type information, the attribute nodes contained in attribute layer 1 may be: food, electronic equipment, household appliances, and the like. Similarly, property layer 2, property layer 3, and property layer 4 may contain corresponding property nodes. The execution main body can divide the historical item browsing information in the historical item browsing information set according to the hierarchical index, and then aggregate the divided attribute nodes to obtain the hierarchical portrait of the historical item of the target user. That is, the hierarchical representation of the historical item is formed by the attribute information corresponding to the black attribute node in each attribute layer in fig. 4, and the white attribute node is an attribute node not related to the target user. The historical item hierarchical representation may characterize which attribute information corresponds to the item that the user prefers.

In some optional implementations of some embodiments, the hierarchical index is obtained by:

firstly, an initial hierarchical index is constructed through preset static attributes.

The execution agent may first obtain the static attributes and build an initial hierarchical index from the static attributes. Wherein the static attribute may include at least one of: item type information, item category information, item brand information, item store name information, and the like. The initial hierarchical index may include at least one property level, the property level including at least one property node.

And secondly, adjusting the attribute nodes of the attribute layer in the initial hierarchical index through the knowledge attributes to obtain the hierarchical index.

In practice, new items are often present at information websites. E.g. a new food product, a new electronic device, etc. These new items may contain attribute information that was not available before, or may compromise multiple previous attribute information. To this end, the execution agent may set these new attribute information as knowledge attributes. That is, the knowledge attribute is an attribute generated based on the historical item browsing information of the target user. The execution body can adjust the attribute nodes of the attribute layer to obtain the hierarchical index. Wherein the adjustment may include adding, aggregating, or splitting, etc. The data processing amount can be greatly reduced through the hierarchical index. For example, there are 1 million kinds of articles. Each of the 1 million items may be divided into attribute nodes in each attribute level in the hierarchical index. Accordingly, each property level of the hierarchical index may correspond to 1 million items. When determining recommended item information through the item recommendation model, the execution subject may determine information that may be needed by the user according to the attribute layers, respectively, where each attribute layer includes a number of attribute nodes that is much smaller than the number of items. As such, layer-by-layer processing or multiple simultaneous layers may be performed to determine what the target user may desire. And then, performing operations such as aggregation on attribute nodes of the articles possibly required by the user in each attribute layer, and further screening out the articles simultaneously meeting the plurality of attribute nodes of the user, so that the articles can be regarded as recommended article information required by the target user. Therefore, the data processing amount is reduced through the hierarchical index (namely the attribute information) of the article, the data processing speed is improved, the information is quickly searched, the online pushing of the information is facilitated, and the timeliness of the information pushing is ensured.

In some optional implementation manners of some embodiments, the performing attribute aggregation on the historical item browsing information in the historical item browsing information set through a preset hierarchical index to obtain a historical item hierarchical representation may include the following steps:

the first step is that the initial article layering portrait of the article corresponding to the historical article browsing information is determined by the layering index according to the historical article browsing information in the historical article browsing information set.

The execution main body can determine the attribute layer where the attribute information of the article corresponding to the historical article browsing information is located through the hierarchical index to obtain an initial article hierarchical portrait. At this time, the black attribute nodes in each attribute layer corresponding to the initial object layered representation are usually dispersed.

And secondly, performing attribute aggregation on the initial object layered portrait set corresponding to the historical object browsing information set to generate a historical object layered portrait.

Each initial object hierarchical representation in the initial object hierarchical representation set has a corresponding initial object hierarchical representation. The execution main body can perform attribute aggregation on corresponding black attribute nodes in the initial article layered portrait set to generate a historical article layered portrait. The hierarchical representation of historical items may represent a distribution of attributes of information desired by the target user. As in fig. 4, attribute level 2 has 2 black attribute nodes adjacent. There are 4 adjacent black attribute nodes in attribute level 4. The adjacent attribute nodes may form an attribute cluster representing that the target user is interested in the items belonging to the attribute cluster.

And 303, training an item recommendation model based on the historical item layering portrait and the historical item browsing information set.

The execution main body can perform model training on the hierarchical portrait of the historical object and the browsing information set of the historical object based on various existing intelligent algorithms to obtain an object recommendation model.

In some optional implementations of some embodiments, the training of the item recommendation model based on the historical item hierarchical representation and the historical item browsing information set may include:

firstly, historical item identifiers of the historical item browsing information in the historical item browsing information set are obtained.

The execution subject can search the historical item identification corresponding to the historical item through the historical item browsing information. The historical item identification may be a number, a cargo number, etc. of the corresponding item.

And secondly, converting the historical object layering image into a historical image feature vector.

The execution subject may convert the historical item hierarchical representation into a historical representation feature vector. For example, the historic representation feature vector may be a determinant, with rows representing the attribute layer and columns representing the ordering of the attribute nodes in the attribute layer. The number in the determinant represents the attribute node corresponding to the number of columns of the data in the attribute layer corresponding to the number of rows of the number.

And thirdly, taking the historical article identifier and the historical image feature vector corresponding to the historical article identifier as input, taking historical article information corresponding to the historical article identifier as output, and training to obtain an article recommendation model.

The execution main body can use historical article identification and historical portrait feature vectors corresponding to the historical article identification as input through various existing intelligent algorithms, and uses historical article information corresponding to the historical article identification as output to train and obtain the article recommendation model. The historical item information may be information such as names and functions of historical items, and may be represented by vectors, images, and the like. The execution body may train the item recommendation model by inputting the historical item identifier and the historical image feature vector corresponding to the historical item identifier and outputting click information of the user on the target item browsing page. According to the actual situation, other ways of training the article recommendation model can be provided, and the details are not repeated here. According to the process for training the item recommendation model, the item recommendation model can construct the corresponding relation among the historical item identification, the historical portrait feature vector and the historical item information.

With further reference to FIG. 5, a flow 500 of further embodiments of a recommendation information generation method is illustrated. The flow 500 of the recommendation information generation method includes the following steps:

step 501, in response to the detection of the item browsing operation of the target user, acquiring an item recommendation model of the target user.

The content of step 501 is the same as that of step 201, and is not described in detail here.

Step 502, obtaining a target item identifier of the target item browsing information corresponding to the item browsing operation.

The execution subject may first obtain a target item identifier of target item browsing information corresponding to the item browsing operation. The target item identification may be a number, a cargo number, etc. of the corresponding item.

Step 503, inputting the target object identifier and the target historical portrait feature vector corresponding to the target object identifier into the object recommendation model to obtain at least one piece of candidate recommended object information.

According to the description, the item recommendation model can construct the corresponding relation among the historical item identification, the historical portrait feature vector and the historical item information. Therefore, the executive body can input the target item identification and the target historical portrait feature vector corresponding to the target item identification into the item recommendation model to obtain at least one piece of candidate recommended item information.

And 504, matching the at least one piece of candidate recommended item information based on the target item identifier to obtain at least one piece of initial recommended item information.

After obtaining the at least one piece of candidate recommended item information, the execution subject may match the at least one piece of candidate recommended item information again through the target item identifier to obtain the initial recommended item information, thereby further improving the effectiveness of the initial recommended item information. For example, the target item identifier corresponds to a mobile phone, and the at least one piece of candidate recommended item information includes a mobile phone, a tablet computer, a basketball, and the like. The initial recommended article information obtained after matching the target article identifier may be a mobile phone and a tablet computer.

And 505, acquiring to-be-processed attribute information of the initial recommended item information in the at least one piece of initial recommended item information.

The execution subject may obtain attribute information of an article corresponding to each piece of initially recommended article information, and set the attribute information as to-be-processed attribute information.

Step 506, aggregating the attribute information sets to be processed corresponding to the at least one piece of initial recommended article information, and determining at least one target attribute set.

The execution subject may perform an aggregation operation on the set of attribute information to be processed, and determine at least one target attribute set. Fig. 6 is a schematic diagram of attribute information aggregation to be processed. As shown in fig. 6, the execution body may aggregate the set of attribute information to be processed to obtain a closed interval a, a closed interval B, and a closed interval C in fig. 6. The closed interval a, the closed interval B, and the closed interval C are combined corresponding to the three target attributes.

And 507, setting the article information corresponding to the at least one target attribute set as recommendation information.

The execution subject may set item information satisfying the closed section a, the closed section B, and/or the closed section C as recommendation information. Further, as can be seen from fig. 6, there is an intersection of the closed interval a, the closed interval B, and the closed interval C, and the intersection includes two common attribute nodes (attribute nodes of the mesh background in fig. 6). The execution main body can set the article information meeting the two common attribute nodes as recommendation information, and the accuracy and the effectiveness of the recommendation information are further improved. The recommendation information may be the same as or different from the historical item browsing information of the target user.

And step 508, constructing an information pushing window based on the recommendation information, and sending the information pushing window to the terminal equipment where the target user is located.

After determining the recommendation information, the execution subject may construct an information push window based on the recommendation information, and send the information push window to the terminal device where the target user is located, as shown in fig. 1.

With further reference to fig. 7, as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a recommendation information generating apparatus, which correspond to those shown in fig. 2, and which may be applied in various electronic devices in particular.

As shown in fig. 7, a recommendation information generating apparatus 700 of some embodiments includes: an item recommendation model acquisition unit 701, an initial recommended item information determination unit 702, and a recommendation information generation unit 703. The item recommendation model acquisition unit 701 is configured to acquire an item recommendation model of a target user in response to detection of an item browsing operation of the target user, wherein the item recommendation model generates recommended item information through a historical item layering portrait of the target user; an initial recommended item information determining unit 702 configured to determine at least one piece of initial recommended item information of the target user based on the item browsing operation and an item recommendation model; a recommendation information generating unit 703 configured to generate recommendation information based on the attribute information of the at least one piece of initially recommended item information.

In an optional implementation manner of some embodiments, the recommendation information generating apparatus 700 may further include an item recommendation model building unit (not shown in the figure) configured to build an item recommendation model, where the item recommendation model building unit may include: the system comprises a historical item browsing information collection subunit (not shown in the figure), a historical item hierarchical sketch construction subunit (not shown in the figure) and an item recommendation model training subunit (not shown in the figure). A historical article browsing information set subunit configured to acquire a historical article browsing information set within the target user set time period; a historical item layered representation construction subunit configured to construct a historical item layered representation from the set of historical item browsing information; and the article recommendation model training subunit is configured to train an article recommendation model based on the historical article layering portrait and the historical article browsing information set.

In an optional implementation manner of some embodiments, the historical item hierarchical representation constructing subunit may include: and the historical item layered portrait construction module (not shown in the figure) is configured to perform attribute aggregation on the historical item browsing information in the historical item browsing information set through a preset layered index to obtain a historical item layered portrait.

In an optional implementation manner of some embodiments, the historical item hierarchical representation building module may include: an initial item hierarchical representation determination sub-module (not shown) and a historical item hierarchical representation generation sub-module (not shown). The initial article layered portrait determining sub-module is configured to determine an initial article layered portrait of an article corresponding to historical article browsing information through the layered index for the historical article browsing information in the historical article browsing information set; and the historical item layered image generation sub-module is configured to generate a historical item layered image by performing attribute aggregation on the initial item layered image set corresponding to the historical item browsing information set.

In an optional implementation manner of some embodiments, the item recommendation model training subunit may include: the system comprises a historical item identification acquisition module (not shown in the figure), a historical portrait feature vector acquisition module (not shown in the figure) and an item recommendation model training module (not shown in the figure). The historical article identifier acquisition module is configured to acquire a historical article identifier of the historical article browsing information in the historical article browsing information set; a historical image feature vector acquisition module configured to convert the historical item hierarchical image into a historical image feature vector; and the article recommendation model training module is configured to train the historical article identifier and the historical image feature vector corresponding to the historical article identifier to obtain an article recommendation model by taking the historical article identifier and the historical image feature vector corresponding to the historical article identifier as input and taking the historical article information corresponding to the historical article identifier as output.

In an optional implementation manner of some embodiments, the recommendation information generating apparatus 700 may include a hierarchical index building unit (not shown in the figure) configured to build a hierarchical index, where the hierarchical index building unit may include: an initial hierarchical index building subunit (not shown in the figure) and a hierarchical index building subunit (not shown in the figure). The initial hierarchical index constructing subunit is configured to construct an initial hierarchical index by using preset static attributes, where the static attributes include at least one of the following: the initial hierarchical index comprises at least one attribute layer, and the attribute layer comprises at least one attribute node; and the hierarchical index structure building subunit is configured to adjust the attribute nodes of the attribute layer in the initial hierarchical index through knowledge attributes to obtain the hierarchical index, wherein the knowledge attributes are attributes generated according to the historical item browsing information of the target user, and the adjustment includes aggregation or splitting.

In an optional implementation manner of some embodiments, the initial recommended item information determining unit 702 may include: a target item identification obtaining sub-unit (not shown in the figure), a candidate recommended item information obtaining sub-unit (not shown in the figure), and an initial recommended item information determining sub-unit (not shown in the figure). The target article identifier acquiring subunit is configured to acquire a target article identifier of target article browsing information corresponding to the article browsing operation; a candidate recommended article information obtaining subunit configured to input the target article identifier and a target historical portrait feature vector corresponding to the target article identifier into the article recommendation model, so as to obtain at least one piece of candidate recommended article information; and the initial recommended article information determining subunit is configured to match the at least one piece of candidate recommended article information based on the target article identifier to obtain at least one piece of initial recommended article information.

In an optional implementation manner of some embodiments, the recommendation information generating unit 703 may include: a pending attribute information obtaining subunit (not shown in the figure), a target attribute set determining subunit (not shown in the figure), and a recommendation information generating subunit (not shown in the figure). The attribute information to be processed acquiring subunit is configured to acquire attribute information to be processed of the initial recommended item information in the at least one piece of initial recommended item information; the target attribute set determining subunit is configured to aggregate the attribute information sets to be processed corresponding to the at least one piece of initially recommended article information, and determine at least one target attribute set; and the recommendation information generation subunit is configured to set the article information corresponding to the at least one target attribute set as recommendation information.

In an optional implementation manner of some embodiments, the recommendation information generating apparatus 700 may further include: and an information pushing unit (not shown in the figure) configured to construct an information pushing window based on the recommendation information, and send the information pushing window to the terminal device where the target user is located.

It will be understood that the elements described in the apparatus 700 correspond to various steps in the method described with reference to fig. 2. Thus, the operations, features and resulting advantages described above with respect to the method are also applicable to the apparatus 700 and the units included therein, and will not be described herein again.

As shown in fig. 8, an electronic device 800 may include a processing means (e.g., central processing unit, graphics processor, etc.) 801 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage means 808 into a Random Access Memory (RAM) 803. In the RAM803, various programs and data necessary for the operation of the electronic apparatus 800 are also stored. The processing apparatus 801, the ROM 802, and the RAM803 are connected to each other by a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

Generally, the following devices may be connected to the I/O interface 805: input devices 806 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 807 including, for example, a Liquid Crystal Display (LCD), speakers, vibrators, and the like; storage 808 including, for example, magnetic tape, hard disk, etc.; and a communication device 809. The communication means 809 may allow the electronic device 800 to communicate wirelessly or by wire with other devices to exchange data. While fig. 8 illustrates an electronic device 800 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 8 may represent one device or may represent multiple devices as desired.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In some such embodiments, the computer program may be downloaded and installed from a network through communications device 809, or installed from storage device 808, or installed from ROM 802. The computer program, when executed by the processing apparatus 801, performs the above-described functions defined in the methods of some embodiments of the present disclosure.

It should be noted that the computer readable medium described above in some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: in response to the detection of the object browsing operation of a target user, acquiring an object recommendation model of the target user, wherein the object recommendation model generates recommended object information through a historical object layering portrait of the target user; determining at least one piece of initial recommended item information of the target user based on the item browsing operation and the item recommendation model; and generating recommendation information based on the attribute information of the at least one piece of initial recommended article information.

Computer program code for carrying out operations for embodiments of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in some embodiments of the present disclosure may be implemented by software, and may also be implemented by hardware. The described units may also be provided in a processor, and may be described as: a processor includes an item recommendation model acquisition unit, an initial recommended item information determination unit, and a recommendation information generation unit. The names of these units do not in some cases constitute a limitation to the unit itself, and for example, the recommendation information generation unit may also be described as a "unit for generating recommendation information by a user".

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.

Claims

1. A recommendation information generation method includes:

in response to the detection of the item browsing operation of a target user, acquiring an item recommendation model of the target user, wherein the item recommendation model generates recommended item information through a historical item layering portrait of the target user;

determining at least one piece of initial recommended item information of the target user based on the item browsing operation and an item recommendation model;

and generating recommendation information based on the attribute information of the at least one piece of initial recommended article information.

2. The method of claim 1, wherein the item recommendation model is trained by:

acquiring a historical article browsing information set in a time period set by the target user;

constructing a hierarchical portrait of the historical object through the historical object browsing information set;

and training an item recommendation model based on the historical item layering portrait and the historical item browsing information set.

3. The method of claim 2, wherein said constructing a hierarchical representation of historical items from said set of historical item browsing information comprises:

and attribute aggregation is carried out on the historical item browsing information in the historical item browsing information set through a preset hierarchical index, so as to obtain a historical item hierarchical portrait.

4. The method of claim 3, wherein the attribute aggregation of the historical item browsing information in the historical item browsing information set through a preset hierarchical index to obtain a historical item hierarchical representation comprises:

for historical item browsing information in the historical item browsing information set, determining an initial item layering portrait of an item corresponding to the historical item browsing information through the layering index;

and generating the historical item layered portrait by performing attribute aggregation on the initial item layered portrait set corresponding to the historical item browsing information set.

5. The method of claim 2, wherein the training of an item recommendation model based on the historical item layering representation and the set of historical item browsing information comprises:

acquiring historical article identifiers of historical article browsing information in the historical article browsing information set;

converting the historical object layering portrait into a historical portrait feature vector;

and taking the historical article identification and the historical portrait feature vector corresponding to the historical article identification as input, taking historical article information corresponding to the historical article identification as output, and training to obtain an article recommendation model.

6. The method of claim 3, wherein the hierarchical index is obtained by:

constructing an initial hierarchical index by using preset static attributes, wherein the static attributes comprise at least one of the following items: the initial hierarchical index comprises at least one attribute layer, and the attribute layer comprises at least one attribute node;

and adjusting attribute nodes of an attribute layer in the initial hierarchical index through knowledge attributes to obtain the hierarchical index, wherein the knowledge attributes are attributes generated according to the historical item browsing information of the target user, and the adjustment comprises aggregation or splitting.

7. The method of claim 1, wherein the determining at least one piece of initial recommended item information for the target user based on the item browsing operations and an item recommendation model comprises:

acquiring a target article identifier of target article browsing information corresponding to the article browsing operation;

inputting the target object identification and the target historical portrait feature vector corresponding to the target object identification into the object recommendation model to obtain at least one piece of candidate recommended object information;

and matching the at least one piece of candidate recommended article information based on the target article identification to obtain at least one piece of initial recommended article information.

8. The method of claim 1, wherein the generating recommendation information based on the attribute information of the at least one piece of initial recommended item information comprises:

acquiring attribute information to be processed of the initial recommended item information in the at least one piece of initial recommended item information;

aggregating attribute information sets to be processed corresponding to the at least one piece of initial recommended article information to determine at least one target attribute set;

and setting the item information corresponding to the at least one target attribute set as recommendation information.

9. The method of claim 1, wherein the method further comprises:

and constructing an information push window based on the recommendation information, and sending the information push window to the terminal equipment where the target user is located.

10. A recommendation information generating apparatus comprising:

an item recommendation model acquisition unit configured to acquire an item recommendation model of a target user in response to detection of an item browsing operation of the target user, the item recommendation model generating recommended item information from a historical item hierarchy representation of the target user;

an initial recommended item information determination unit configured to determine at least one piece of initial recommended item information of the target user based on the item browsing operation and an item recommendation model;

a recommendation information generating unit configured to generate recommendation information based on the attribute information of the at least one piece of initially recommended item information.

11. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-9.

12. A computer-readable medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the method of any one of claims 1 to 9.