Search result sorting method, device, storage medium and electronic device
By compensating the click-through rate, subscription rate and playback rate of search results, and combining them to determine the sorting reference factor, it solves the solidification problem caused by the "Matthew effect" in search sorting, and improves the richness of the search result page Improve user experience.
Patent Information
- Application Number
- CN202210124774.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-02-10
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2042-02-10
AI Technical Summary
In search scenarios, the "Matthew effect" is prone to appear when directly using click-through rate weighted sorting, resulting in the solidification of the sorting results, reducing the richness and personalization of the result page, and affecting the user experience.
By compensating the position bias of click statistical indicators, subscription statistical indicators and play statistical indicators, combined with their combination, we determine the sorting reference factor of search results, which is used to sort search results, reduce the "Matthew effect" and improve the richness of the result page. and personalized.
It effectively reduces the solidification of sorting results, improves the richness and personalization of the result page, ensures that niche content can be found, maintains the vitality of search sorting, and improves user experience.
Smart Images

Figure CN114461913B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of search, and in particular, to a search result sorting method, device, storage medium and electronic device. Background Art
[0002] In the search scenario, the user's click behavior information can effectively reflect the relevance between the search query and the content item, and can be used to improve the ranking of the content item in the result page. However, due to the influence of factors such as the position and type of the content item, directly using the click-through rate (CTR) weighting to apply to the ranking results is often prone to the "Matthew effect". The better the position of the content item, the higher its CTR tends to be, and the greater the weighting of the ranking results, the better the position will be, and vice versa. This will eventually lead to the solidification of the ranking results, reducing the richness and personalization of the result page. Some niche items cannot be found, and the search ranking loses its vitality, affecting the user experience. Summary of the invention
[0003] The purpose of the present application is to provide a search result ranking method, device, storage medium and electronic device to at least partially improve the above-mentioned problems.
[0004] In order to achieve the above purpose, the technical solution adopted in the embodiment of the present application is as follows:
[0005] In a first aspect, an embodiment of the present application provides a search result sorting method, the method comprising:
[0006] Determine a ranking reference factor for the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results;
[0007] The search results are the results obtained by searching for the target content, the click statistical index is the index obtained by compensating the click rate of the search results by position bias, the subscription statistical index is the index obtained by compensating the subscription rate of the search results by position bias, and the play statistical index is the index obtained by compensating the play rate of the search results by position bias;
[0008] The search results are ranked based on the ranking reference factor of each of the search results.
[0009] In a second aspect, an embodiment of the present application provides a search result sorting device, the device comprising:
[0010] A processing unit, configured to determine a ranking reference factor for the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results;
[0011] The search results are the results obtained by searching for the target content, the click statistical index is the index obtained by compensating the click rate of the search results by position bias, the subscription statistical index is the index obtained by compensating the subscription rate of the search results by position bias, and the play statistical index is the index obtained by compensating the play rate of the search results by position bias;
[0012] A sorting unit is used to sort the search results based on a sorting reference factor of each search result.
[0013] In a third aspect, an embodiment of the present application provides a storage medium on which a computer program is stored, and the computer program implements the above method when executed by a processor.
[0014] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory, wherein the memory is used to store one or more programs; when the one or more programs are executed by the processor, the above method is implemented.
[0015] Compared with the prior art, the search result sorting method, device, storage medium and electronic device provided in the embodiments of the present application determine the sorting reference factor of the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators and playback statistical indicators corresponding to the search results; wherein the search results are the results obtained by searching for the target content, the click statistical indicators are the indicators obtained after position bias compensation for the click rate of the search results, the subscription statistical indicators are the indicators obtained after position bias compensation for the subscription rate of the search results, and the playback statistical indicators are the indicators obtained after position bias compensation for the playback rate of the search results; based on the sorting reference factor of each search result, the search results are sorted. The click statistical indicators, subscription statistical indicators and playback statistical indicators obtained after position bias compensation are used for sorting, which can reduce the "Matthew effect", avoid the solidification of the sorting results, and improve the richness and personalization of the result page. Some niche items can be found, maintaining the vitality of search sorting and improving user experience.
[0016] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0018] Figure 1 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;
[0019] Figure 2 A flowchart of a search result sorting method provided in an embodiment of the present application;
[0020] Figure 3 A schematic diagram of sub-steps of S107 provided in an embodiment of the present application;
[0021] Figure 4 One of the sub-step schematic diagrams of S107 provided in an embodiment of the present application;
[0022] Figure 5 One of the flowcharts of the search result sorting method provided in the embodiment of the present application;
[0023] Figure 6 A schematic diagram of a unit of a search result ranking device provided in an embodiment of the present application.
[0024] In the figure: 10 - processor; 11 - memory; 12 - bus; 13 - communication interface; 201 - processing unit; 202 - sorting unit. DETAILED DESCRIPTION
[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.
[0026] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0027] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0028] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0029] In the description of the present application, it should be noted that the terms "upper", "lower", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the product of the application is usually placed when in use. They are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.
[0030] In the description of this application, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "disposed" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.
[0031] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.
[0032] Directly using the CTR weighting method to sort results is prone to the "Matthew effect". The better the position of the content item, the higher its CTR tends to be. The greater the weighting of the sorting results, the better the position will be, and vice versa. In view of the above situation, the academic community began to study the Bayesian-based probabilistic click model, which can deduct the influence of different biases. However, due to the instability of the probability model parameter estimation and the lack of robustness of the results, it is rarely used in the industry.
[0033] In one possible implementation, click relevance can be measured by eliminating position bias, which is defined as follows:
[0034]
[0035]
[0036] Among them, c r represents the number of clicks on the search result at position r, i r Indicates the number of times the search result is exposed at position r, CTR r represents the click-through rate of all search results at position r, c i,r represents the number of clicks on the i-th search result at position r, e i,r Represents the number of exposures of the i-th search result at position r.
[0037] It should be understood that the denominator in the above formula is the expected number of clicks ECs after considering the position bias, while the numerator is the actual number of clicks Cs. The final COEC (Clicks Over Expected Clicks) represents the normalized number of clicks.
[0038] For example, the user enters question A in the search box of the browser, and the background performs a search and comparison to obtain search results A1 to A15, and obtains the COECs of A1 to A15 respectively.
[0039] Take A5's COEC as an example, the c corresponding to A5 r represents the number of clicks on search result A5 at position r, i r Indicates the number of times search result A5 is exposed at position r, CTR r It represents the click-through rate of all search results at position r, that is, the corresponding click-through rate at position r.
[0040] However, after a lot of practice and summary, the inventors found that when sorting by COEC obtained by the above formula, there is a situation where the position bias is excessively deducted and the correlation becomes low, and the effect of using it in the search scenario is not ideal. The reason is that the basic assumption of COEC is that the position only contains bias information and has nothing to do with the correlation, that is, a certain Item (search result) is randomly ranked in a certain position. In the search scenario, this basic assumption is obviously unreasonable. An Item is ranked in the front position, indicating that it has a high correlation with the search Query (target content, the content that the user wants to know). If the COEC assumption is used, deducting the position bias will actually deduct part of the correlation.
[0041] In order to overcome the above problems, the present application embodiment provides an electronic device, which may be a server device or a computer device. Figure 1 , a schematic diagram of the structure of an electronic device. The electronic device includes a processor 10, a memory 11, and a bus 12. The processor 10 and the memory 11 are connected via the bus 12, and the processor 10 is used to execute an executable module stored in the memory 11, such as a computer program.
[0042] The processor 10 may be an integrated circuit chip having signal processing capabilities. In the implementation process, each step of the search result sorting method may be completed by an integrated logic circuit of hardware in the processor 10 or by instructions in the form of software. The above-mentioned processor 10 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, or discrete hardware components.
[0043] The memory 11 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
[0044] The bus 12 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. Figure 1 Although only one bidirectional arrow is used in the figure, it does not mean that there is only one bus 12 or only one type of bus 12 .
[0045] The memory 11 is used to store programs, such as programs corresponding to the search result sorting device. The search result sorting device includes at least one software function module that can be stored in the memory 11 in the form of software or firmware or fixed in the operating system (OS) of the electronic device. After receiving the execution instruction, the processor 10 executes the program to implement the search result sorting method.
[0046] Possibly, the electronic device provided in the embodiment of the present application further includes a communication interface 13. The communication interface 13 is connected to the processor 10 via a bus. Optionally, the electronic device can communicate with the user terminal via the communication interface 13, for example, to receive a user's search request, or to feed back a search result webpage to the user.
[0047] It should be understood that Figure 1 The structure shown is only a schematic diagram of a portion of the electronic device. The electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown. Figure 1 Each component shown in the figure can be implemented by hardware, software or a combination thereof.
[0048] The search result sorting method provided in the embodiment of the present application can be applied to but not limited to Figure 1 For detailed procedures, please refer to the electronic equipment shown in Figure 2 , the search result sorting method includes: S107 and S108, which are specifically described as follows.
[0049] S107: Determine a ranking reference factor for the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results.
[0050] Among them, the search results are the results obtained by searching for the target content, the click statistics are the indicators obtained by compensating the click rate of the search results for position bias, the subscription statistics are the indicators obtained by compensating the subscription rate of the search results for position bias, and the playback statistics are the indicators obtained by compensating the playback rate of the search results for position bias.
[0051] It should be understood that the click rate corresponding to a search result can be determined based on the total number of exposures and the total number of clicks of a search result, the subscription rate corresponding to a search result can be determined based on the total number of exposures and the total number of subscriptions of a search result, and the play rate corresponding to a search result can be determined based on the total number of exposures and the total number of plays of a search result. The number of exposures should be understood as the number of times a user can view the search result. For example, there are 3 pages of search results. When the first page is displayed, the search results on the first page are in an exposed state. If the user does not switch to the second and third pages, the search results on the second and third pages are in an unexposed state.
[0052] Optionally, the target content is the content entered by the user in the browser or search window. For example, if the target content entered by the user is "Beijing Winter Olympics", the corresponding search results may be news, videos, pictures or peripheral products related to "Beijing Winter Olympics".
[0053] S108, sorting the search results based on the sorting reference factor of each search result.
[0054] It should be understood that the ranking reference factor is associated with the above-mentioned click statistical indicators, subscription statistical indicators and playback statistical indicators. The size of the ranking reference factor is associated with the ranking of the search results. Optionally, the larger the ranking reference factor, the higher the ranking of the search results.
[0055] The above-mentioned click statistics, subscription statistics and playback statistics are obtained after position bias compensation for click rate, subscription rate and playback rate. Compared with sorting directly based on click rate, subscription rate and playback rate, the click statistics, subscription statistics and playback statistics obtained after position bias compensation are used for sorting, which can reduce the "Matthew effect", avoid the solidification of sorting results, improve the richness and personalization of the result page, and some niche items can be found, so as to maintain the vitality of search sorting and improve user experience.
[0056] At the same time, while taking position bias into consideration, the actual click-through rate, subscription rate, and play rate of the search results are also combined to avoid over-compensation, which would result in the top-ranked search results having a lower relevance to the target content, thereby improving the user experience.
[0057] It should be understood that the sorted content is displayed on a web page or user page for user reference.
[0058] In summary, the embodiment of the present application provides a search result sorting method, which determines the sorting reference factor of the search result based on any one or more combinations of click statistical indicators, subscription statistical indicators and playback statistical indicators corresponding to the search results; wherein the search result is the result obtained by searching for the target content, the click statistical indicator is the indicator obtained after position bias compensation for the click rate of the search result, the subscription statistical indicator is the indicator obtained after position bias compensation for the subscription rate of the search result, and the playback statistical indicator is the indicator obtained after position bias compensation for the playback rate of the search result; based on the sorting reference factor of each search result, the search results are sorted. The click statistical indicators, subscription statistical indicators and playback statistical indicators obtained after position bias compensation are used for sorting, which can reduce the "Matthew effect", avoid solidification of sorting results, and improve the richness and personalization of the result page. Some niche items can be found, maintaining the vitality of search sorting and improving user experience.
[0059] exist Figure 2 On the basis of how to obtain click statistics, subscription statistics and playback statistics, the present application embodiment also provides a possible implementation method, please refer to Figure 3 , the search result ranking method also includes: S101, S102, S103, S104, S105 and S106, which are specifically described as follows.
[0060] S101, based on the number of clicks and exposures of the search results at different ranking positions, and the click rate corresponding to each ranking position, determine a first position compensation factor corresponding to the click rate.
[0061] S102: Perform position offset compensation on the click rate of the search result based on the first position compensation factor to obtain a click statistical indicator.
[0062] Alternatively, please refer to the following formula:
[0063]
[0064] Among them, c i,r represents the number of clicks on the i-th search result at position r, e i,r represents the number of times the i-th search result is exposed at position r, CTR r Represents the click-through rate of all search results at position r, CTR i represents the click-through rate of the i-th search result, CCOEC i Represents the click statistics of the i-th search result.
[0065] Assume that P represents the first position compensation factor corresponding to the i-th search result.
[0066] It should be understood that when different types of search results are ranked in the same position, their corresponding click-through rates may be different, such as video, picture, news link, and advertisement.
[0067] On this basis, the embodiment of the present application also provides a possible implementation method for simultaneously performing position offset compensation and type offset compensation, please refer to the following formula:
[0068]
[0069] Among them, CTR r,t Indicates the click-through rate of search results of type t at position r. Type t can be any of the above types.
[0070] Assume that P represents the first position compensation factor corresponding to the i-th search result. At this time, P can be used to perform position bias compensation and type bias compensation at the same time.
[0071] S103, determining a second position compensation factor corresponding to the subscription rate based on the number of subscriptions and the number of exposures of the search results at different sorting positions, and the subscription rate corresponding to each sorting position.
[0072] S104: Perform position offset compensation on the subscription rate of the search result based on the second position compensation factor to obtain a subscription statistical indicator.
[0073] Alternatively, please refer to the following formula:
[0074]
[0075] Among them, s i,r represents the number of subscriptions for the i-th search result at position r, e i,r Indicates the number of times the i-th search result is exposed at position r, STR r represents the subscription rate of all search results at position r, STR i represents the subscription rate of the i-th search result, SCOEC i Indicates the subscription statistics indicator of the i-th search result.
[0076] Assume that Q represents the second position compensation factor corresponding to the i-th search result.
[0077] It should be understood that when different types of search results are ranked in the same position, their corresponding subscription rates may be different, such as video, picture, news link, and advertisement.
[0078] On this basis, the embodiment of the present application also provides a possible implementation method for simultaneously performing position offset compensation and type offset compensation, please refer to the following formula:
[0079]
[0080] Among them, STR r,t Indicates the subscription rate of search results of type t at position r. Type t can be any of the above types.
[0081] Assume that Q represents the second position compensation factor corresponding to the i-th search result. At this time, Q can be used to perform position bias compensation and type bias compensation at the same time.
[0082] S105, determining a third position compensation factor corresponding to the play rate based on the play times and exposure times of the search results at different sorting positions and the play rate corresponding to each sorting position.
[0083] S106: Perform position offset compensation on the play rate of the search result based on the third position compensation factor to obtain a play statistical indicator.
[0084] Alternatively, please refer to the following formula:
[0085]
[0086] Among them, p i,r represents the number of times the i-th search result at position r is played, e i,r PTR represents the number of times the i-th search result is exposed at position r. r Represents the play rate of all search results at position r, PTR i represents the play rate of the i-th search result, PCOEC i Indicates the playback statistics of the i-th search result.
[0087] Assume that M represents the third position compensation factor corresponding to the i-th search result.
[0088] It should be understood that when different types of search results are ranked in the same position, their corresponding play rates may be different, such as video, picture, news link, and advertisement.
[0089] On this basis, the embodiment of the present application also provides a possible implementation method for simultaneously performing position offset compensation and type offset compensation, please refer to the following formula:
[0090]
[0091] Among them, PTR r,tIndicates the playback rate of search results of type t at position r. Type t can be any of the above types.
[0092] Assume that M represents the third position compensation factor corresponding to the i-th search result. At this time, M can be used to perform position bias compensation and type bias compensation at the same time.
[0093] exist Figure 2 Based on the content in S107, the present application embodiment also provides a possible implementation method, please refer to Figure 4 , S107 includes: S107-1, which is specifically described as follows.
[0094] S107-1, determining any one of the click statistical index, the subscription statistical index, and the play statistical index as a ranking reference factor for the search results.
[0095] In a possible implementation, the rationality of the first sequence, the second sequence, and the third sequence corresponding to all search results can also be verified, wherein the first sequence, the second sequence, and the third sequence are sequences corresponding to click statistical indicators, subscription statistical indicators, and playback statistical indicators, respectively, and the corresponding indicator parameters in each sequence are arranged from high to low.
[0096] Optionally, obtain a correlation arrangement array of the search results and the target content in each sequence, and complete rationality verification based on the change trend of each correlation in the correlation arrangement array. When the change trend is unified and the fluctuation range is small, the rationality is high.
[0097] The statistical index corresponding to the sequence with the highest rationality is determined as the reference factor for sorting.
[0098] exist Figure 2 Based on the content in S107, the present application embodiment also provides a possible implementation method, please refer to Figure 5 , S107 includes: S107-2, which is specifically described as follows.
[0099] S107-2, determining a ranking reference factor for the search results based on the click statistical index, the subscription statistical index, the play statistical index, and the weights of each behavior.
[0100] Among them, behavioral weights include click weight, subscription weight, and play weight.
[0101] Optionally, the ranking reference factor of the search results is determined according to the following formula:
[0102] R i =α×CCOEC i +β×SCOEC i+γ×PCOEC i ;
[0103] Among them, parameters α, β, and γ represent click weight, subscription weight, and play weight, respectively.
[0104] It should be understood that R i You can continue to add other parameters and obtain a final statistical indicator (R i ), can not only optimize the problem of excessive position bias in COEC, but also further consider type bias and other user behaviors, and ultimately effectively measure the relevance of search queries and items, bringing significant improvements to online search indicators.
[0105] In a possible implementation, the average value of the click statistical index, the subscription statistical index, and the play statistical index may be determined as a ranking reference factor for the search results.
[0106] See also Figure 6 , Figure 6 A search result sorting device is provided in an embodiment of the present application. Optionally, the search result sorting device is applied to the electronic device described above.
[0107] The search result sorting device includes: a processing unit 201 and a sorting unit 202 .
[0108] The processing unit 201 is used to determine a ranking reference factor of the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results.
[0109] Among them, the search results are the results obtained by searching for the target content, the click statistics are the indicators obtained by compensating the click rate of the search results for position bias, the subscription statistics are the indicators obtained by compensating the subscription rate of the search results for position bias, and the playback statistics are the indicators obtained by compensating the playback rate of the search results for position bias.
[0110] The sorting unit 202 is used to sort the search results based on the sorting reference factor of each search result.
[0111] Optionally, the processing unit 201 is further used to determine a ranking reference factor for search results based on click statistics, subscription statistics, playback statistics and various behavior weights; wherein the behavior weights include click weights, subscription weights and playback weights.
[0112] Optionally, the processing unit 201 may execute the above S101 to S107, and the sorting unit may execute the above S108.
[0113] It should be noted that the search result ranking device provided in this embodiment can execute the method flow shown in the above method flow embodiment to achieve the corresponding technical effect. For the sake of brief description, for parts not mentioned in this embodiment, please refer to the corresponding content in the above embodiment.
[0114] The present application also provides a storage medium that stores computer instructions and programs, which, when read and executed, execute the search result sorting method of the above embodiment. The storage medium may include a memory, a flash memory, a register, or a combination thereof.
[0115] The following provides an electronic device, which may be a server device. Figure 1 As shown, the above-mentioned search result sorting method can be implemented; specifically, the electronic device includes: a processor 10, a memory 11, and a bus 12. The processor 10 can be a CPU. The memory 11 is used to store one or more programs. When the one or more programs are executed by the processor 10, the search result sorting method of the above-mentioned embodiment is executed.
[0116] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a 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 a different order from the order marked in the accompanying drawings. For example, two consecutive boxes 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 and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0117] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
[0118] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0119] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
[0120] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. A search result ranking method, characterized in that: The method comprises: Determine a ranking reference factor for the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results; The search results are the results obtained by searching for the target content, the click statistical index is the index obtained by compensating the click rate of the search results by position bias, the subscription statistical index is the index obtained by compensating the subscription rate of the search results by position bias, and the play statistical index is the index obtained by compensating the play rate of the search results by position bias; sorting the search results based on a sorting reference factor of each of the search results; Before determining the ranking reference factor of the search results, the method further includes: Determine a first position compensation factor corresponding to the click rate based on the number of clicks and exposures of the search results at different ranking positions, and the click rate corresponding to each ranking position; Performing position offset compensation on the click rate of the search result based on the first position compensation factor to obtain the click statistical indicator; Among them, c i,r represents the number of clicks on the i-th search result at position r, e i,r represents the number of times the i-th search result is exposed at position r, CTR r,t Represents the click-through rate of search results of type t at position r, CTR i represents the click-through rate of the i-th search result, CCOEC i Represents the click statistics of the i-th search result.
2. The search result ranking method according to claim 1, characterized in that: Before determining the ranking reference factor of the search results, the method further includes: Determine a second position compensation factor corresponding to the subscription rate based on the number of subscriptions and the number of exposures of the search results at different ranking positions, and the subscription rate corresponding to each ranking position; The subscription rate of the search result is compensated for position bias based on the second position compensation factor to obtain the subscription statistical indicator.
3. The search result ranking method according to claim 1, characterized in that: Before determining the ranking reference factor of the search results, the method further includes: Based on the number of plays and exposures of the search results at different sorting positions, and the play rate corresponding to each sorting position, determining a third position compensation factor corresponding to the play rate; The playback rate of the search result is compensated for position offset based on the third position compensation factor to obtain the playback statistical indicator.
4. The search result ranking method according to claim 1, characterized in that: The step of determining the ranking reference factor of the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results includes: Any one of the click statistical indicator, the subscription statistical indicator, and the play statistical indicator is determined as a ranking reference factor for the search results.
5. The search result ranking method according to claim 1, characterized in that: The step of determining the ranking reference factor of the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results includes: Determine a ranking reference factor for the search results based on the click statistical index, the subscription statistical index, the playback statistical index, and each behavior weight; The behavior weights include click weight, subscription weight and play weight.
6. A search result sorting device, characterized in that: The device comprises: a processing unit, configured to determine a first position compensation factor corresponding to a click rate based on the number of clicks and exposures of the search results at different ranking positions, and the click rate corresponding to each ranking position; and perform position offset compensation on the click rate of the search results based on the first position compensation factor to obtain the click statistical indicator; Among them, c i,r represents the number of clicks on the i-th search result at position r, e i,r represents the number of times the i-th search result is exposed at position r, CTR r,t Represents the click-through rate of search results of type t at position r, CTR i represents the click-through rate of the i-th search result, CCOEC i Indicates the statistical indicator of clicks on the i-th search result; A processing unit, configured to determine a ranking reference factor for the search results based on any one or more combinations of click statistical indicators, subscription statistical indicators, and playback statistical indicators corresponding to the search results; The search results are the results obtained by searching for the target content, the click statistical index is the index obtained by compensating the click rate of the search results by position bias, the subscription statistical index is the index obtained by compensating the subscription rate of the search results by position bias, and the play statistical index is the index obtained by compensating the play rate of the search results by position bias; A sorting unit is used to sort the search results based on a sorting reference factor of each search result.
7. The search result ranking device according to claim 6, characterized in that: The processing unit is also used to determine the ranking reference factor of the search results based on the click statistical indicators, the subscription statistical indicators, the playback statistical indicators and various behavior weights; wherein the behavior weights include click weights, subscription weights and playback weights.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
9. An electronic device, characterized in that: include: A processor and a memory, the memory being used to store one or more programs; When the one or more programs are executed by the processor, the method according to any one of claims 1 to 5 is implemented.
Citation Information
Patent Citations
Method and device for determining sequence of search items, electronic equipment and computer readable storage medium
CN110598102A
Sorting method and device, electronic equipment and computer readable storage medium
CN110674400A
Model training method and device, electronic equipment and readable storage medium
CN112487283A
Energy compensation method and device, equipment and storage medium
CN113515828A
Personalization of placed content ordering in search results
US20050240580A1
