Method and device for determining target object

Abstract

The present invention discloses a method and apparatus for determining target items, relating to the field of computer technology. A specific implementation of the method includes: calculating the click-through rate of each trigger branch in a historical period and its recommendation ratio among all trigger branches; selecting items from each trigger branch using the recommendation ratio and internal weights, and arranging the selected items in descending order of the trigger branch's click-through rate; and determining the top-ranked items as target items. This implementation can guide the item sorting and target item determination of an information object recommendation system during the triggering phase based on the recommendation ratio and click-through rate of each trigger branch in a historical period, thereby accurately measuring the importance of each trigger branch.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for determining a target object. Background Art

[0002] Current information object recommendation systems operate in two phases: the triggering phase and the recommendation phase. In the triggering phase, a certain number of target items are selected from different triggering branches as preliminary recommendations. These items are then sent to the recommendation phase for a final decision on whether to recommend them. In the triggering phase, items are ranked according to their importance for selection. Although each item has a weight within the triggering branch, the weights between branches are difficult to estimate.

[0003] In the prior art, a fixed weight and offset are generally set for each trigger branch. When sorting items in all trigger branches, the total weight of each item is the product of its internal weight and the fixed weight of the trigger branch to which it belongs, plus the sum of the offset of the trigger branch to which it belongs.

[0004] In the process of implementing the present invention, the inventors discovered that the prior art has at least the following problems:

[0005] 1. Improper settings of the above-mentioned fixed weights and offsets can seriously affect the target item selection results. However, existing technologies rely solely on subjective settings of fixed weights and offsets, fail to utilize historical data patterns, and fail to accurately reflect the importance of each trigger branch.

[0006] 2. When adding a new trigger branch, setting and adjusting fixed weights and offsets will become difficult. Summary of the Invention

[0007] In view of this, an embodiment of the present invention provides a method and device for determining target items, which can guide the information object recommendation system to sort items and determine target items in the triggering stage based on the recommendation ratio and click-through rate of each trigger branch in the historical period, thereby achieving accurate measurement of the importance of each trigger branch.

[0008] To achieve the above objective, according to one aspect of the present invention, a method for determining a target object is provided.

[0009] The method for determining a target item in an embodiment of the present invention is used to determine a target item from multiple trigger branches; wherein each trigger branch has at least one item with a set internal weight; the method includes: counting the click-through rate of each trigger branch in a historical period and its recommendation ratio in all trigger branches; using the recommendation ratio and the internal weight to select items from each trigger branch, and arranging the selected items in descending order according to the click-through rate of the trigger branch; and determining the multiple items arranged in front as target items.

[0010] Optionally, the click rate of a triggering branch is the click rate of information objects associated with the triggering branch items, and the recommendation ratio of a triggering branch is the ratio of the number of recommended items of the triggering branch to all triggering branches.

[0011] Optionally, the selecting of items from each triggering branch using the recommendation ratio and the internal weight specifically includes: arranging the items in each triggering branch in descending order according to their internal weights in the triggering branch; and selecting the items that are arranged at the front and have a proportion equal to the recommendation ratio of the triggering branch.

[0012] Optionally, the method further includes: after arranging the selected items in descending order according to the click-through rate of the trigger branch, arranging the remaining items in each trigger branch that are arranged in descending order of internal weights in descending order according to the click-through rate of the trigger branch after the first item queue, to obtain a second item queue; wherein, the first item queue is: the queue formed by arranging the selected items in descending order according to the click-through rate of the trigger branch; and combining duplicate items in the second queue into one item to form a third item queue.

[0013] Optionally, determining the plurality of items arranged in front as target items specifically includes: determining a preset number of items arranged in front in the third item queue as target items.

[0014] To achieve the above objective, according to another aspect of the present invention, a device for determining a target object is provided.

[0015] The device for determining a target item in an embodiment of the present invention can be used to determine the target item from multiple trigger branches; wherein each trigger branch has at least one item with a set internal weight; the device may include: a statistical unit, used to count the click-through rate of each trigger branch in a historical period and its recommendation ratio in all trigger branches; a sorting unit, used to select items from each trigger branch using the recommendation ratio and the internal weight, and arrange the selected items in descending order according to the click-through rate of the trigger branch; an item determination unit, used to determine the multiple items arranged in front as target items.

[0016] Optionally, the click-through rate of a trigger branch is the click-through rate of the information object associated with the trigger branch item, and the recommendation ratio of a trigger branch is the ratio of the number of recommended items of the trigger branch to all trigger branches; the sorting unit can be further used to: arrange the items of each trigger branch in descending order according to their internal weight in the trigger branch; and select the items that are ranked first and account for the recommendation ratio of the trigger branch.

[0017] Optionally, the device may further include an item replenishment unit, which is used to: after arranging the selected items in descending order according to the click rate of the trigger branch, arrange the remaining items in each trigger branch that are arranged in descending order of internal weights after the first item queue in descending order according to the click rate of the trigger branch to obtain a second item queue; combine duplicate items in the second queue into one item to form a third item queue; wherein, the first item queue is: the queue formed by arranging the selected items in descending order according to the click rate of the trigger branch; the item determination unit may be further used to: determine a preset number of items arranged in front in the third item queue as target items.

[0018] To achieve the above objective, according to another aspect of the present invention, an electronic device is provided.

[0019] An electronic device of the present invention includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the target item provided by the present invention.

[0020] To achieve the above objective, according to another aspect of the present invention, a computer-readable storage medium is provided.

[0021] A computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, implements the method for determining a target object provided by the present invention.

[0022] According to the technical solution of the present invention, one embodiment of the above invention has the following advantages or beneficial effects:

[0023] First, we calculate the click-through rate (CTR) of each trigger branch and its recommendation ratio among all trigger branches over a historical period to measure the importance of each trigger branch. The CTR indicates the conversion rate of clicks to the information objects associated with the trigger branch items, while the recommendation ratio reflects the proportion of recommended items in the trigger branch. Using these two metrics as trigger branch weights objectively measures the relative value of each trigger branch.

[0024] Secondly, after calculating click-through rate and recommendation ratio, each trigger branch is ranked by click-through rate. Within the trigger branch, items that have already been ranked by internal weight are selected based on the recommendation ratio. This allows the precise position of each item in the queue to be determined, ensuring that high-value items in the trigger branch are not missed. The triggering stage of the information object recommendation system is then completed by selecting the top items as target items. This overcomes the drawbacks of fixed weights in existing technologies and eliminates the need to change existing strategies when new trigger branches are added.

[0025] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are provided for a better understanding of the present invention and are not intended to limit the present invention.

[0027] Figure 1 is a schematic diagram of the main steps of a method for determining a target object according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of a specific implementation of the method for determining a target object according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of components of an apparatus for determining a target object according to an embodiment of the present invention;

[0030] Figure 4 is an exemplary system architecture diagram in which embodiments of the present invention may be applied;

[0031] Figure 5 2 is a schematic diagram of the structure of an electronic device used to implement the method for determining a target object in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following description of exemplary embodiments of the present invention is made in conjunction with the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding. These details should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0033] Before specifically introducing the technical solution of the present invention, we first introduce an application scenario of the method for determining the target item to which the embodiment of the present invention can be applied. In the current information object recommendation system, the working process can be divided into a triggering stage and a recommendation stage. Among them, the information object refers to a data object or virtual information in a specific form as an information carrier, such as an online advertisement (online advertisement refers to an advertisement using web pages on the Internet as a carrier). In the triggering stage, after receiving the user request, each trigger branch can extract a certain number of items from its vocabulary based on the user's historical behavior or the label of the user portrait. The trigger branch refers to a trigger path established based on the user behavior category. For example, the trigger branches corresponding to the user's browsing behavior and transaction behavior are the browsing trigger branch and the transaction trigger branch respectively. Each trigger branch is provided with one or more vocabulary tables, which store items associated with the user's browsing, transaction and other behaviors and similar items to the item. Each similar item is also provided with a similarity with its reference item.

[0034] Taking the browsing trigger branch as an example, after receiving the user request, the vocabulary can be queried based on the items that the user has recently browsed, and a certain number of similar items of the above items can be obtained. The weight of the item within the browsing trigger branch can be re-determined based on the similarity of each item, that is, the internal weight to be introduced later. This internal weight can be used to measure the importance of the item within the browsing trigger branch. It can be understood that when selecting the final target item (that is, the item to be sent to the recommendation stage), the items with larger internal weights should be given priority. After each trigger branch selects a certain number of items from the vocabulary, the method of the present invention can be used to sort all items between different trigger branches and determine the target item. Finally, the target item is sent to the recommendation stage for subsequent processing. At this time, the trigger stage is completed.

[0035] During the recommendation stage, the click-through rate can be estimated for each target item, and whether the target item will be recommended to the user can be determined based on the click-through rate estimation results, user historical behavior, and item historical evaluation data (recommendation refers to displaying the information object associated with the item to the user, and after the recommendation, the user may click on it). The above is the online workflow of the information object recommendation system.

[0036] After the above process has been executed for a period of time, the click-through rate and recommendation ratio of each trigger branch during this period can be obtained offline. These two indicators are parameters selected by the present method to evaluate the importance of each trigger branch. The click-through rate of a trigger branch refers to the ratio of the number of clicks on the information objects associated with the trigger branch items over a period of time to the number of recommended items (or the number of information objects associated with the recommended items) for that trigger branch. It can be understood that the click-through rate can reflect the conversion effect of the trigger branch from recommendation to click. The recommendation ratio of a trigger branch refers to the ratio of the number of recommended items (or the number of information objects associated with recommended items) for that trigger branch over all trigger branches over a period of time. The recommendation ratio can reflect the proportion of the trigger branch in the recommendation process. In specific applications, since recommended items and clicked information objects are both tagged with information about the trigger branch they belong to, the above indicators can be obtained through statistical analysis. In addition, the number of clicks and the number of recommended items are cumulative totals. If an information object associated with an item is clicked twice or an item is recommended twice, the number of clicks or the number of recommended items increases by 2.

[0037] It should be noted that the following application scenarios are only used to help understand the technical solution of the present invention and do not constitute any limitation on the application of the method of the present invention. In fact, in addition to the above information object recommendation system, any applicable scenario and environment can adopt the method of the present invention.

[0038] It should be understood that the terms "first," "second," and so forth are used herein to describe various objects, but these objects are not limited by these terms. These terms are used solely to distinguish one object from another. For example, a first queue of items could be referred to as a second queue of items, and a second queue of items could be referred to as a first queue of items, without departing from the scope of the present invention. The first and second queues are both queues of items, but they are not the same queue.

[0039] The technical solutions of the present invention will be described in detail below. It should be noted that the embodiments of the present invention and the technical features in the embodiments can be combined with each other without conflict.

[0040] Figure 1 2 is a schematic diagram of the main steps of a method for determining a target object according to an embodiment of the present invention.

[0041] like Figure 1 As shown, the method for determining a target object according to an embodiment of the present invention can be specifically performed according to the following steps:

[0042] Step S101: Count the click rate of each triggering branch in the historical period and its recommendation ratio in all triggering branches.

[0043] In this step, the historical period, relative to the current moment, can be any interval in the past, such as the past day, week, or month. Preferably, in practical applications, the method of the present invention can be used to periodically adjust the trigger branches. For example, the target item selection strategy for each trigger branch within a week starting from the current moment can be adjusted based on the past week's historical data. It is understood that the method of the present invention can also be used in scenarios where non-periodic adjustments are required.

[0044] In an embodiment of the present invention, the click rate and recommendation ratio of each triggering branch can be counted offline, and these two indicators can be used as a basis for evaluating the importance of the triggering branch to select subsequent items.

[0045] Step S102: Select items from each triggering branch using the recommendation ratio and internal weight, and arrange the selected items in descending order according to the click rate of the triggering branch.

[0046] As a preferred solution, in this step, the following steps can be performed for each trigger branch: the items in the trigger branch are sorted in descending order based on their internal weights, and the top-ranked items with a recommended percentage for the trigger branch are selected. Then, for each selected item in the trigger branch, the items are sorted in descending order based on their click-through rate, maintaining their order within the trigger branch, to form a first item queue.

[0047] For example, within a historical period, the browsing trigger branch has a click-through rate of ctrA, a recommendation ratio of imprA, and a number of items of countA; the transaction trigger branch has a click-through rate of ctrB, a recommendation ratio of imprB, and a number of items of countB. If ctrA is greater than ctrB, then when executing step S102, the items in the browsing trigger branch are first sorted in descending order by internal weight, and the items ranked first with a number of imprA*countA are selected. The items in the transaction trigger branch are then sorted in descending order by internal weight, and the items ranked first with a number of imprB*countB are selected. The existing order of the items selected from the two trigger branches is then maintained, and the items with a number of imprA*countA are placed at the front, followed by the items with a number of imprB*countB, in descending order of click-through rate. The resulting item queue is the first item queue. In specific applications, when selecting items based on the recommendation ratio, if the product of the recommendation ratio and the number of items in the trigger branch is not an integer, rounding up or down can be performed according to preset rules.

[0048] After forming the first item queue according to the above method, each trigger branch often has a certain number of items remaining. It can be understood that these items are still arranged in descending order according to the internal weight within the trigger branch to which they belong. At this time, the remaining items of each trigger branch can be arranged after the first item queue in descending order according to the click-through rate of the trigger branch to obtain the second item queue. In other words, traverse each trigger branch in descending order of click-through rate and arrange the remaining items in the trigger branch in sequence after the first item queue. For example, in the above example, after the two trigger branches select items, the number of remaining items in the browsing trigger branch may be (1-impr)A*countA, and the number of remaining items in the transaction trigger branch may be (1-imprB)*countB. At this time, the remaining items of the browsing trigger branch can be first arranged after the first item queue, and then the remaining items of the transaction trigger branch can be arranged after the current queue to form the second item queue.

[0049] Since the above process may select the same product from different trigger branches, there may be duplicate items in the second item queue. In this case, the duplicate items can be combined into one item according to the preset strategy. The queue formed at this time is the third item queue. The above strategy can be flexibly set according to the actual scenario. For example, it can be set as follows:

[0050] 1. Convert the click rate of each trigger branch into the external weight of the trigger branch. The total weight of each item in the second item queue is the product of its internal weight and the external weight of the trigger branch to which it belongs. The above external weight must ensure that the order of the second item queue by total weight is the same as the current situation.

[0051] 2. Convert the duplicate items into one item, and add up the total weights of the duplicate items as the total weight of the converted item. Re-determine its position in the second item queue according to the total weight of the item.

[0052] After the above steps, a ranking reflecting the importance of items in different trigger branches can be obtained, from which the target item can be determined.

[0053] Step 103: Determine the plurality of items arranged in front as target items.

[0054] In this step, target items may be obtained according to business requirements, for example, a preset number of items arranged at the front of the third item queue may be determined as target items, thereby completing the final determination of the target items.

[0055] Figure 2 This is a schematic diagram of a specific implementation of the method for determining a target object according to an embodiment of the present invention, from which the method of the present invention can be intuitively understood. Figure 2 In this context, trigger branch fusion refers to the process of determining target items from each trigger branch in the above steps. In practical applications, after offline statistics on click-through rate and recommendation ratio are compiled, they can be written to an online configuration file. This data can then be retrieved when determining target items online.

[0056] In the technical solution of the embodiment of the present invention, the click-through rate (CTR), which reflects the effectiveness of click-through conversions, and the recommendation ratio (RRR), which reflects the historical recommendation ratio, are used as the basis for measuring the importance of each trigger branch. After calculating the CTR and RRR, each trigger branch is ranked by CTR. Within the trigger branch, items that have already been sorted by internal weight are selected based on the RRR. This allows the precise position of each item in the queue to be determined, ensuring that high-value items in the trigger branch are not missed. The triggering stage of the information object recommendation system is then completed by selecting the top items as target items. This overcomes the drawbacks of fixed weights in the prior art and eliminates the need to change existing strategies when new trigger branches are added.

[0057] It should be noted that for the aforementioned method embodiments, for ease of description, they are expressed as a series of action combinations. However, those skilled in the art should be aware that the present invention is not limited to the order of the actions described, and certain steps can actually be performed in other orders or simultaneously. In addition, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required to implement the present invention.

[0058] In order to better implement the above solutions of the embodiments of the present invention, relevant devices for implementing the above solutions are also provided below.

[0059] See also Figure 3As shown, the device 300 for determining a target item provided by an embodiment of the present invention can be used to determine a target item from a plurality of trigger branches; wherein each trigger branch has at least one item with a set internal weight; the device may include: a statistical unit 301, a sorting unit 302 and an item determination unit 303.

[0060] Among them, the statistical unit 301 can be used to count the click rate of each trigger branch in the historical period and its recommendation ratio in all trigger branches; the sorting unit 302 can be used to select items from each trigger branch using the recommendation ratio and the internal weight, and arrange the selected items in descending order according to the click rate of the trigger branch; the item determination unit 303 can be used to determine multiple items arranged in front as target items.

[0061] In an embodiment of the present invention, the click-through rate of a trigger branch is the click-through rate of the information object associated with the item in the trigger branch, and the recommendation ratio of a trigger branch is the ratio of the number of recommended items in the trigger branch to all trigger branches; the sorting unit 302 can be further used to: sort the items in each trigger branch in descending order according to their internal weight in the trigger branch; and select the items that are ranked first and account for the recommendation ratio of the trigger branch.

[0062] In an optional implementation, the device may further include an item replenishment unit, which is used to: after arranging the selected items in descending order according to the click rate of the trigger branch, arrange the remaining items in each trigger branch that are arranged in descending order of internal weights after the first item queue according to the click rate of the trigger branch to obtain a second item queue; combine the duplicate items in the second queue into one item to form a third item queue; wherein, the first item queue is: the queue formed by arranging the selected items in descending order according to the click rate of the trigger branch; the item determination unit 303 may be further used to: determine a preset number of items arranged in front in the third item queue as target items.

[0063] In the technical solution of the embodiment of the present invention, the click-through rate (CTR), which reflects the effectiveness of click-through conversions, and the recommendation ratio (RRR), which reflects the historical recommendation ratio, are used as the basis for measuring the importance of each trigger branch. After calculating the CTR and RRR, each trigger branch is ranked by CTR. Within the trigger branch, items that have already been sorted by internal weight are selected based on the RRR. This allows the precise position of each item in the queue to be determined, ensuring that high-value items in the trigger branch are not missed. The triggering stage of the information object recommendation system is then completed by selecting the top items as target items. This overcomes the drawbacks of fixed weights in the prior art and eliminates the need to change existing strategies when new trigger branches are added.

[0064] Figure 4An exemplary system architecture 400 is shown to which the method for determining a target item or the apparatus for determining a target item according to an embodiment of the present invention may be applied.

[0065] like Figure 4 As shown, system architecture 400 may include terminal devices 401, 402, and 403, a network 404, and a server 405 (this architecture is merely an example, and the components included in the specific architecture may be adjusted based on the specific application). Network 404 is used to provide a medium for communication links between terminal devices 401, 402, and 403 and server 405. Network 404 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0066] Users can use terminal devices 401, 402, 403 to interact with server 405 via network 404 to receive or send messages, etc. Terminal devices 401, 402, 403 can be installed with various client applications, such as information object recommendation applications, web browser applications, etc. (only for example).

[0067] The terminal devices 401 , 402 , and 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0068] Server 405 may be a server that provides various services, such as an information object processing server (for example only) that supports information object recommendation applications operated by users using terminal devices 401, 402, and 403. The information object processing server may process received target item extraction requests and provide feedback on the processing results (for example, the determined target item (for example only)) to terminal devices 401, 402, and 403.

[0069] It should be noted that the method for determining the target item provided in the embodiment of the present invention is generally executed by the server 405 . Accordingly, the device for determining the target item is generally provided in the server 405 .

[0070] It should be understood that Figure 4 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0071] The present invention also provides an electronic device. The electronic device in an embodiment of the present invention includes: one or more processors; and a storage device configured to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining a target item provided by the present invention.

[0072] Reference below Figure 5, which shows a schematic structural diagram of a computer system 500 of an electronic device suitable for implementing an embodiment of the present invention. Figure 5 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0073] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage unit 508 into a random access memory (RAM) 503. Various programs and data required for the operation of the computer system 500 are also stored in the RAM 503. The CPU 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0074] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, and the like; an output section 507 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 508 including a hard disk; and a communication section 509 including a network interface card such as a LAN card or a modem. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 510 as needed, so that computer programs read therefrom can be installed in the storage section 508 as needed.

[0075] In particular, according to embodiments disclosed herein, the processes described in the main step diagrams above can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the main step diagrams. In the above embodiments, the computer program can be downloaded and installed from a network via the communication section 509 and / or installed from removable media 511. When the computer program is executed by the central processing unit 501, the above-described functions defined in the system of the present invention are performed.

[0076] It should be noted that the computer-readable medium described in the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical cable, RF, or any suitable combination thereof.

[0077] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0078] The units described in the embodiments of the present invention may be implemented in software or hardware. The units described may also be provided within a processor. For example, a processor may be described as comprising a statistics unit, a ranking unit, and an item determination unit. The names of these units do not, in some cases, limit the units themselves. For example, the statistics unit may also be described as "a unit that provides the ranking unit with the click-through rate and recommendation ratio of the triggering branch."

[0079] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment, or may exist independently and not incorporated into the device. The computer-readable medium carries one or more programs, which, when executed by the device, cause the device to perform the following steps: calculating the click-through rate of each trigger branch in a historical period and its recommendation ratio among all trigger branches; selecting items from each trigger branch using the recommendation ratio and internal weights, and sorting the selected items in descending order of the click-through rate of the trigger branch; and determining the top-ranked items as target items.

[0080] In the technical solution of the embodiment of the present invention, the click-through rate (CTR), which reflects the effectiveness of click-through conversions, and the recommendation ratio (RRR), which reflects the historical recommendation ratio, are used as the basis for measuring the importance of each trigger branch. After calculating the CTR and RRR, each trigger branch is ranked by CTR. Within the trigger branch, items that have already been sorted by internal weight are selected based on the RRR. This allows the precise position of each item in the queue to be determined, ensuring that high-value items in the trigger branch are not missed. The triggering stage of the information object recommendation system is then completed by selecting the top items as target items. This overcomes the drawbacks of fixed weights in the prior art and eliminates the need to change existing strategies when new trigger branches are added.

[0081] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for determining a target object, for determining a target object from a plurality of trigger branches; wherein: Each trigger branch has at least one item with a set internal weight; characterized in that the method includes: Statistics are collected for each trigger branch's click-through rate and its recommendation ratio among all trigger branches during the historical period. The click-through rate of a trigger branch is the click-through rate of the information objects associated with the items in that trigger branch. The recommendation ratio of a trigger branch is the ratio of the number of recommended items in that trigger branch to all trigger branches. An information object refers to a specific form of data object or virtual information that serves as an information carrier. Selecting items from each triggering branch using the recommendation ratio and the internal weight includes: arranging the items in each triggering branch in descending order according to their internal weights in the triggering branch, selecting the top-ranked items that account for the recommendation ratio of the triggering branch; and arranging the selected items in descending order according to click-through rates of the triggering branch; and After arranging the selected items in descending order according to the click-through rates of the triggering branches, the remaining items in each triggering branch that are arranged in descending order of internal weight are arranged after the first item queue in descending order according to the click-through rates of the triggering branches, thereby obtaining a second item queue; wherein the first item queue is: a queue formed by arranging the selected items in descending order according to the click-through rates of the triggering branches; Combine the duplicate items in the second queue into one item to form a third item queue; The plurality of items arranged in front are determined as target items.

2. The method according to claim 1, characterized in that The step of determining the plurality of items arranged in front as target items specifically includes: A preset number of items arranged at the front of the third item queue are determined as target items.

3. A device for determining a target object, for determining a target object from a plurality of trigger branches; wherein, Each trigger branch has at least one item with a set internal weight; characterized in that the device includes: A statistical unit is used to calculate the click-through rate of each trigger branch during a historical period and its recommendation ratio among all trigger branches. The click-through rate of a trigger branch is the click-through rate of the information objects associated with the items in the trigger branch. The recommendation ratio of a trigger branch is the ratio of the number of recommended items in the trigger branch to the total number of trigger branches. An information object refers to a specific form of data object or virtual information that serves as an information carrier. a sorting unit, configured to select items from each triggering branch using the recommendation ratio and the internal weight, and arrange the selected items in descending order according to the click rate of the triggering branch; an item determination unit, configured to determine a plurality of items arranged in front as target items; The sorting unit is further configured to: sort the items of each triggering branch in descending order according to their internal weight in the triggering branch; select the items that are ranked first and have a proportion equal to the recommended proportion of the triggering branch; The item replenishment unit is used to: after arranging the selected items in descending order according to the click-through rate of the trigger branch, arrange the remaining items in each trigger branch that are arranged in descending order of internal weight after the first item queue in descending order according to the click-through rate of the trigger branch to obtain a second item queue; combine duplicate items in the second queue into one item to form a third item queue; wherein the first item queue is: the queue formed by arranging the selected items in descending order according to the click-through rate of the trigger branch.

4. The device according to claim 3, characterized in that The item determination unit is further configured to determine a preset number of items arranged in front of the third item queue as target items.

5. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 2 is implemented.

Images