Reordering method, device and system for recommended content and computing equipment
By constructing a data table and using a sorting window threshold judgment method, the multi-dimensional and mandatory constraint problems of recommendation content shuffling algorithms in existing technologies are solved, achieving efficient recommendation data sorting and improving user experience and system performance.
Patent Information
- Application Number
- CN202410939926.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-15
AI Technical Summary
Existing recommendation content shuffling algorithms suffer from problems such as failure to shuffle at the end, inability to comprehensively consider multi-dimensional classifications, and inability to meet mandatory shuffling constraints, resulting in a poor user experience.
By acquiring multiple attribute information from the recommendation data, a corresponding data table is constructed. Based on the sorting window threshold, it is determined whether each attribute information meets the conditions. Recommendations with similar dimensions are separated, and a sorting queue of multi-dimensional attributes is generated to satisfy the scattering constraint.
It achieves multi-dimensional attribute shuffling of recommendation data, improves user experience, satisfies forced shuffling constraints, exhibits excellent performance, and buffers the interaction between the recommendation system and users.
Smart Images

Figure CN118964724A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a method, device, system, computing equipment, computer program product and storage medium for reordering recommended content. Background Art
[0002] Breaking up is a reordering process based on the results of content recommendation and search systems to improve the user's visual experience. Its goal is to disperse content with similar attributes, increase freshness, and avoid monotonous and repetitive content that may cause aesthetic fatigue to users.
[0003] The recommendation results directly given by various content systems often have the following pain points: 1. Similar categories of content are easy to be clustered. If the features of the content are similar, the recommendation scores they get are also easy to be similar, and the monotonous continuous display of similar content is not the result expected by users; 2. Avoid recommendation algorithm attacks. The same author and a large number of content published at the same time are likely to impact the display opportunities of other similar content; 3. It is easy to fall into the optimal solution of recommendation. For users who have almost no usage traces, it is easy to amplify the only features, which is easy to produce wrong recommendations; 4. The user's preferences are captured too strongly. At the psychological level, users are sensitive to the fact that their privacy or preferences are perfectly captured. Too accurate results are not only likely to cause user disgust, but also to limit the conversion of user potential.
[0004] Most of the existing scattering algorithms have shortcomings, such as failure of scattering at the end, easy clustering, only considering classification in one dimension, and being unable to comprehensively consider other factors. Summary of the invention
[0005] In view of the above problems, the present invention is proposed to provide a method, apparatus, system, computing device, computer program product and storage medium for reordering recommended content that overcome the above problems or at least partially solve the above problems.
[0006] According to one aspect of the present invention, there is provided a method for reordering recommended content, which is executed in a computing device, and comprises: obtaining a plurality of recommended data in the recommended content, wherein the recommended data comprises a plurality of attribute information, and for each attribute information, constructing a corresponding data table; judging whether each attribute information contained in the currently-to-be-ordered recommended data satisfies a condition, and when all the attribute information satisfies the condition, writing the recommended data into a sorting queue, and if any attribute information does not satisfy the condition, writing the recommended information into a data table corresponding to the attribute information that does not satisfy the condition; traversing all the recommended data until all the recommended data are reordered, generating a final sorting queue, and using the final sorting queue as a recommendation result.
[0007] Optionally, in the method for reordering recommended content according to the present invention, it is determined whether each attribute information contained in the recommended data to be currently ordered meets the conditions, including: for each attribute information of the recommended data to be currently ordered, determining whether the attribute information meets the sorting window threshold, the sorting window threshold respectively indicating the sorting interval corresponding to each attribute information; if so, continuing to determine whether other attribute information of the recommended data meets the sorting window threshold; if not, writing the recommended data into the data table of the attribute information.
[0008] Optionally, in the method for reordering recommended content according to the present invention, judging whether the attribute information satisfies the sorting window threshold includes: determining the sorting sequence number of the sorted recommended data having the attribute information in the sorting queue; calculating the difference between the maximum value among the sorting sequence numbers and the sequence number corresponding to the current entry position in the sorting queue; judging whether the difference is greater than the sorting window threshold corresponding to the attribute information; if greater, confirming that the attribute information satisfies the sorting window threshold; if not greater, confirming that the attribute information does not satisfy the sorting window threshold.
[0009] Optionally, in the method for reordering recommended content according to the present invention, all recommended data are traversed until all recommended data are reordered, including: for each recommended data in the data table, when it is detected that the recommended data is written to the data table, and the sum of the difference between the number of newly written recommended data in the sorting queue and the corresponding attribute information pointed to by the data table is greater than the sorting window threshold, the recommended data is re-used as the current recommended data to be sorted.
[0010] Optionally, the method for reordering recommended contents according to the present invention further includes: if each ordered recommended data does not include each attribute information included in the currently ordered recommended data, confirming that each attribute information of the currently ordered recommended data satisfies a condition.
[0011] Optionally, the method for reordering recommended content according to the present invention further includes: if there is no ordered recommended data in the ordering queue, directly writing the currently ordered recommended data into the ordering queue.
[0012] Optionally, in the method for reordering recommended content according to the present invention, it also includes: when all recommended data are sorted, if it is detected that the number of sorted recommended data in the sorting queue is less than a preset number threshold, the sorting window threshold is reduced so as to reorder the recommended data in each data table again.
[0013] Optionally, the method for reordering recommended content according to the present invention further includes: if it is detected that the amount of ordered recommended data in the ordering queue is not less than a preset quantity threshold, deleting the recommended data that has not been reordered and the recommended data in each data table.
[0014] According to another aspect of the present invention, there is provided an apparatus for reordering recommended content, which resides in a computing device, and comprises: an acquisition module, adapted to acquire a plurality of recommended data in the recommended content, wherein the recommended data comprises a plurality of attribute information, and for each attribute information, a corresponding data table is constructed; a judgment module, adapted to judge whether each attribute information contained in the currently ordered recommended data satisfies a condition, and when all attribute information satisfies the condition, writes the recommended data into a sorting queue, and if any attribute information does not satisfy the condition, writes the recommended information into a data table corresponding to the attribute information that does not satisfy the condition; and a sorting module, adapted to traverse all recommended data until all recommended data are reordered, generate a final sorting queue, and use the final sorting queue as a recommendation result.
[0015] According to another aspect of the present invention, there is provided a recommendation system, comprising the above-mentioned device for reordering recommended content, suitable for reordering the recommended data determined after the recall stage and the sorting stage, obtaining the reordering result of the recommended data, and outputting the recommended data according to the reordering result.
[0016] According to another aspect of the present invention, there is provided a computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be suitable for being executed by the at least one processor, and the program instructions include instructions for executing the above method.
[0017] According to another aspect of the present invention, a readable storage medium storing program instructions is provided. When the program instructions are read and executed by a computing device, the computing device executes the above method.
[0018] According to the solution of the present invention, by obtaining the attribute information of each recommended data and judging whether each attribute information satisfies the conditions one by one, the recommended contents of similar dimensions are separated, the interaction between the recommendation system and the user is buffered, the user experience is improved, and the multi-dimensional attribute dispersion of the recommended data is achieved, the dispersion degree is good, the performance is excellent, and the mandatory dispersion constraints are satisfied.
[0019] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:
[0021] Figure 1 A block diagram of the physical components (ie, hardware) of a computing device 100 is shown;
[0022] Figure 2 A flowchart of a method 200 for reordering recommended content according to an embodiment of the present invention is shown;
[0023] Figure 3 A schematic diagram of a process for determining whether each attribute information satisfies a condition according to an embodiment of the present invention is shown;
[0024] Figure 4 A schematic structural diagram of a device 400 for reordering recommended contents according to an embodiment of the present invention is shown;
[0025] Figure 5 FIG. 5 is a schematic diagram showing the structure of a recommendation system 500 according to an embodiment of the present invention. DETAILED DESCRIPTION
[0026] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0027] The scattering algorithm is to reorder an ordered list. Since the original order is given according to certain scores, weights, time limits, etc., the reordering process should preserve the original order as much as possible while satisfying the scattering constraints.
[0028] Currently, the commonly used scattering algorithms include column-based scattering, weight distribution scattering, window sliding scattering and their variants.
[0029] The following introduces the existing scattering algorithms:
[0030] 1. Scatter by column:
[0031] The idea is to put the recommended content of different attributes into different buckets, and try to choose a different bucket each time. This way, the elements can be scattered as much as possible. In order to ensure that the original arrangement order is retained, each column is sorted in the original order.
[0032] The advantages of this algorithm are: simple and direct, easy to implement; the scattering effect is relatively good. Although sorting may cause elements to be accidentally adjacent at the beginning and end of the column, the maximum number of adjacent elements before the end is 2, which does not affect the experience; the performance is relatively stable and is not easily affected by the input structure.
[0033] At the same time, the disadvantages of this algorithm are: the end-breaking function fails and clustering is prone to occur; the respect for the original order is not strong, and even objects with very low recommendation coefficients are forced to appear in the front; it can only consider the classification of one attribute information and cannot comprehensively consider other attribute information. At the same time, this algorithm is quite dependent on the number of categories. If the categories are detailed enough, the shortcomings of this algorithm can be partially covered up. In terms of performance, the time and space consumption are both O(n).
[0034] 2. Weight distribution method:
[0035] The principle is to define a new weight for each object:
[0036]
[0037] Among them, w i The coefficient assigned to each attribute information represents the priority of the scattering, and Count i It represents the performance of attribute information (the number of times the same attribute has appeared). Intuitively speaking, the more times similar attributes have appeared, the greater the weight value will be, and in the function calculation process, the original order factor is naturally considered, so after calculating the weight, no other processing is required, just sort by weight from small to large.
[0038] The main disadvantage of this algorithm is that it cannot rigidly meet the strong dispersion constraint conditions, such as when there is a mandatory requirement for content authors to publish at most one article for every 10 articles. The above algorithm can meet the requirements when the author dispersion is met. If there are a few authors who publish articles in a concentrated manner, the constraint conditions cannot be met. Because of the cumulative effect of weight calculation, the algorithm is also prone to failure at the end.
[0039] 3. Sliding window breakup:
[0040] The design principle is to stipulate that there cannot be repeated elements in the window, that is, to simulate a display page seen by the user without duplication. If a duplicate element is found in the window, a suitable element will be detected later to exchange with the current element.
[0041] The advantages of this algorithm are: only local calculation is required, no complete fragmentation is required, and it meets the needs of topN.
[0042] The main disadvantages of this algorithm are: its robustness is not good, it is greatly affected by the sequence distribution, and it also cannot avoid the defect of end accumulation.
[0043] It can be seen that most of the existing fragmentation methods have shortcomings and cannot simultaneously meet the requirements of fragmentation degree, fragmentation dimension, performance, and hard constraints.
[0044] In order to solve the problems existing in the above-mentioned prior art, the solution of the present invention is proposed. An embodiment of the present invention provides a method for reordering recommended content, which obtains attribute information of each recommended data and determines whether each attribute information satisfies the conditions for breaking up. The method separates the recommended content of similar dimensions, buffers the interaction between the recommendation system and the user, improves the user experience, and realizes the multi-dimensional attribute breaking up of the recommended data, with a good degree of breaking up, excellent performance, and meets the mandatory breaking up constraint conditions.
[0045] The method may be executed in a computing device 100. The computing device 100 may be implemented as a server, such as an application server, a Web server, etc.; it may also be implemented as a desktop computer, a laptop computer, a processor chip, a tablet computer, etc., but is not limited thereto. Figure 1 1 shows a block diagram of the physical components (i.e., hardware) of a computing device 100. In a basic configuration, the computing device 100 includes at least one processing unit 102 and a system memory 104. According to one aspect, depending on the configuration and type of the computing device, the system memory 104 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memories.
[0046] According to one aspect, system memory 104 includes operating system 105. System memory 104 also includes program modules 150. According to one aspect, operating system 105, for example, is suitable for controlling the operation of computing device 100. Furthermore, examples are practiced in conjunction with graphics libraries, other operating systems, or any other application programs, and are not limited to any particular application or system. Figure 1 This basic configuration is illustrated in FIG. 1 by those components within dashed line 108. According to one aspect, computing device 100 has additional features or functionality. For example, according to one aspect, computing device 100 includes additional data storage devices (removable and / or non-removable), such as magnetic disks, optical disks, or tapes. Such additional storage Figure 1 1 is illustrated by a removable storage device 109 and a non-removable storage device 110.
[0047] As stated above, according to one aspect, a program module 150 is stored in the system memory 104. According to one aspect, the program module 150 can be implemented as one or more computer program products, and the present application does not limit the type of computer program products, for example, it can include: email, word processing application, spreadsheet application, database application, slide show application, painting or computer-aided application, web browser, etc. In some embodiments according to the present application, the computer program / instructions related to the article typesetting method 200 are packaged as a computer program product, and when these computer programs / instructions are executed by the processor (i.e., the processing unit 102), the article typesetting method 200 according to the present application is implemented.
[0048] According to one aspect, the examples may be practiced on a circuit comprising discrete electronic components, a packaged or integrated electronic chip containing logic gates, a circuit utilizing a microprocessor, or a single chip containing electronic components or a microprocessor. Figure 1 Each or many components shown in can be integrated into a system on a chip (SOC) on a single integrated circuit to practice examples. According to one aspect, such a SOC device may include one or more processing units, a graphics unit, a communication unit, a system virtualization unit, and various application functions, all of which are integrated (or "burned") into a chip substrate as a single integrated circuit. When operated via SOC, the functions described in this article can be operated via a dedicated logic integrated with other components of the computing device 100 on a single integrated circuit (chip). Embodiments of the present invention can also be practiced using other technologies capable of performing logical operations (such as AND, OR, and NOT), including but not limited to mechanical, optical, fluid, and quantum technologies. In addition, embodiments of the present invention can be practiced in a general-purpose computer or in any other circuit or system.
[0049] According to one aspect, the computing device 100 may also have one or more input devices 112, such as a keyboard, a mouse, a pen, a voice input device, a touch input device, etc. Output devices 114 may also be included, such as a display, a speaker, a printer, etc. The aforementioned devices are examples and other devices may also be used. The computing device 100 may include one or more communication connections 116 that allow communication with other computing devices 118, which may be printing devices, such as printers. Examples of suitable communication connections 116 include, but are not limited to: RF transmitters, receivers and / or transceiver circuits; Universal Serial Bus (USB), parallel and / or serial ports.
[0050] The term computer-readable medium as used herein includes computer storage media. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (e.g., computer-readable instructions, data structures, or program modules). System memory 104, removable storage device 109, and non-removable storage device 110 are all examples of computer storage media (i.e., memory storage). Computer storage media may include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, cassettes, tapes, disk storage or other magnetic storage devices, or any other products that can be used to store information and can be accessed by computer device 100. According to one aspect, any such computer storage medium can be a part of computing device 100. Computer storage media does not include carrier waves or other propagated data signals.
[0051] According to one aspect, communication media is implemented by computer readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transport mechanism), and includes any information delivery media. According to one aspect, the term "modulated data signal" describes a signal that has one or more characteristics set or changed in a manner that encodes information in the signal. By way of example and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.
[0052] In an embodiment according to the present invention, the computing device 100 is configured to execute the method 200 for reordering recommended content according to the present invention. The computing device 100 includes one or more processors and one or more readable storage media storing program instructions, and when the program instructions are configured to be executed by one or more processors, the computing device executes the method 200 for reordering recommended content in the embodiment of the present invention.
[0053] Figure 2 A flowchart of a method 200 for reordering recommended content according to an embodiment of the present invention is shown. The method 200 is suitable for being executed in a computing device (such as the aforementioned computing device 100).
[0054] like Figure 2 As shown, the purpose of method 200 is to provide a method for reordering recommended content by performing multi-dimensional attribute scattering on recommendation data, with a good scattering degree, excellent performance, and satisfying mandatory scattering constraints.
[0055] It should be noted that the recommendation system process generally has three steps: recall stage → sorting stage → re-ranking stage, among which the sorting stage is further divided into coarse sorting stage and fine sorting stage.
[0056] In the recall phase, the candidate data set is very large, and the most fundamental requirement is speed. In addition, a small number of features should be used, which is the characteristic of the recall phase. In the recall phase, the fastest speed is as good as possible, but user interests must also be taken into account. In simple terms, recall will reduce a large number of materials to hundreds or thousands, and then throw them to the subsequent sorting phase.
[0057] In the coarse sorting stage, because the number of items returned by each user recall link is still too large, and there is a fear that the sorting link cannot keep up with the speed, a coarse sorting link can be added between the recall and the fine sorting. Through a small number of user and item features and a simple model, the recall results are roughly sorted. Under the premise of ensuring a certain degree of accuracy, the number of items transmitted later can be further reduced. Coarse sorting is often optional and can be used or different, depending on the scenario.
[0058] The characteristics of the refined ranking stage are completely different from those of the recall stage. The refined ranking stage requires the model to be accurate enough, which is its fundamental requirement. In addition, because the amount of data processed in this stage is relatively small, a complex model that can withstand the speed limit of the computing device 100 can be used to use any useful features that can be thought of to personalize and sort the recommended content as accurately as possible. In the final analysis, it is for one thing: how accurate it is.
[0059] The re-ordering stage is to break up or meet specific forced insertion requirements of business operations, and is committed to breaking up and re-ordering the sorted recommended content to a certain extent to improve the user experience.
[0060] The method 200 for reordering recommended contents provided in this embodiment is executed in the reordering stage of the recommendation system, and is suitable for reordering the recommended data determined in the recall stage and the sorting stage.
[0061] The method 200 starts at step 202 . In step 202 , a plurality of recommendation data in the recommendation content is obtained, wherein the recommendation data includes a plurality of attribute information, and a corresponding data table is constructed for each attribute information.
[0062] In a specific example, taking the case where each piece of recommended data contains two pieces of attribute information, the results shown in the following Table 1 are obtained.
[0063] Table 1
[0064] Recommended data ID Attribute information 1 Attribute information 2 1 a A 2 a B 3 b B 4 b B 5 c C 6 c C 7 d D 8 d D 9 e D 10 c A 11 f E 12 a D 13 g B
[0065] As shown in Table 1 above, each piece of recommended data corresponds to a recommended data ID, and the recommended data ID generally indicates the ranking of each piece of recommended data after the ranking stage. Attribute information refers to the attribute characteristics of the recommended content. For example, if the recommended data is an article, the attribute information may be the author of the article, the type of article, the time of article publication, etc.
[0066] According to the number of attribute information involved in the dispersion of each recommended data, a corresponding data table is constructed. In this example, since each recommended data contains two attribute information, two data tables (attribute information 1 data table and attribute information 2 data table) are constructed, wherein the attribute information 1 data table is used to temporarily store the recommended data whose attribute information 1 does not meet the preset conditions in the recommended data in step 204, and the attribute information 2 data table is used to temporarily store the recommended data whose attribute information 2 does not meet the preset conditions in the recommended data.
[0067] Subsequently, in step 204, it is determined whether each attribute information contained in the recommended data to be sorted meets the conditions, and when all attribute information meets the conditions, the recommended data is written into the sorting queue; if any attribute information does not meet the conditions, the recommended information is written into the data table corresponding to the attribute information that does not meet the conditions.
[0068] Figure 3 The figure shows a flow chart of judging whether each attribute information satisfies a condition according to an embodiment of the present invention.
[0069] like Figure 3 As shown, judging whether each attribute information meets the conditions includes the following sub-steps:
[0070] Step 2040: for each attribute information of the recommended data to be sorted, determine whether the attribute information meets the sorting window threshold. The sorting window threshold indicates the sorting interval corresponding to each attribute information. For example, for attribute information 1, its corresponding sorting window threshold is 3, which means that there are at least 3 other recommended data that do not contain attribute information 1 between each recommended data containing attribute information 1 in the sorting queue.
[0071] It is worth noting that for different attribute information, the sorting window threshold can be different. For example, the sorting window threshold corresponding to attribute information 1 is 3, and the sorting window threshold corresponding to attribute information 2 is 4, etc. By setting different sorting window thresholds for different attribute information, the degree of dispersion of each recommended data can be improved.
[0072] Furthermore, the size of the sorting window threshold can be set according to the amount of recommended data. For example, when the amount of data is small, the size of the sorting window threshold can be appropriately increased, and vice versa. In addition, the size of the sorting window threshold can also be set according to attribute information. For example, attribute information 1 is the author, attribute information 2 is the article type, and the user prefers articles of the same author, then the sorting window threshold corresponding to attribute information 1 can be appropriately reduced. When the user prefers articles of the same type, the sorting window threshold corresponding to attribute information 2 can be appropriately reduced. This application does not limit this.
[0073] Specifically, the sorting sequence number of the sorted recommended data with the attribute information in the sorting queue is first determined. The attribute information here refers to the attribute information currently being judged under the condition. For example, if the attribute information 1 of the recommended data to be sorted under the current condition is attribute information 1, then the attribute information refers to the attribute information 1; if the attribute information 2 of the recommended data to be sorted under the current condition is attribute information 2, then the attribute information refers to the attribute information 2.
[0074] The sorted recommended data refers to the recommended data that has been re-sorted. These data are stored in a sorting queue. The sorting queue allocates a sorting sequence number to each sorted data to indicate the order relationship of each sorted data after re-sorting.
[0075] Exemplarily, the sorted recommended data in the current sorting queue only includes the first recommended data 1: (aA) in the above Table 1, where 1 is the ID of the recommended data, a refers to the attribute information 1 of the recommended data, and A refers to the attribute information A of the recommended data. The sorting sequence number of the recommended data 1: (aA) in the sorting queue is 1.
[0076] At this time, it is necessary to determine whether the recommended data 2: (aB) can be written into the sorting queue, so the attribute information a and B of the recommended data 2: (aB) are first used as the judgment attribute information (ie, the first attribute information) for comparison with the recommended data 1: (aA).
[0077] After comparison, it is found that the recommended data 1: (aA) contains the attribute information a of the recommended data 2: (aB), and the sorting number 1 of the recommended data 1: (aA) is obtained.
[0078] Then, the difference between the maximum value among the sorting sequence numbers and the sequence number corresponding to the current entry position in the sorting queue is calculated.
[0079] Continuing with the above example, since there is only one piece of recommended data 1: (aA) in the current sorting queue, the corresponding sequence number at the current sorting position (i.e. the sorting sequence number of the next recommended data written into the sorting queue) is 2, and the calculated difference is 2-1=1.
[0080] It should be noted that there may be multiple recommended data in the sorting queue that contain the attribute information of the current recommended data to be sorted for condition judgment. Therefore, it is necessary to determine the recommended data with the largest sorting sequence number from these recommended data, and use this maximum value to calculate the difference with the sequence number corresponding to the current sorting position.
[0081] Afterwards, it is determined whether the difference is greater than the sorting window threshold corresponding to the attribute information. If so, it is confirmed that the attribute information meets the sorting window threshold. If not, it is confirmed that the attribute information does not meet the sorting window threshold.
[0082] If all the attribute information of the recommended data to be sorted currently meets the sorting window threshold, the process proceeds to step 2044 to continue to determine whether other attribute information of the recommended data meets the sorting window threshold.
[0083] If any of the attribute information of the recommended data to be sorted does not meet the sorting window threshold, the process proceeds to step 2046 to write the recommended data into the data table of the attribute information. The attribute information here refers to the attribute information that does not meet the sorting window threshold.
[0084] In some embodiments, if each of the sorted recommended data does not contain each of the attribute information contained in the currently sorted recommended data, it is marked that each of the attribute information of the currently sorted recommended data meets the condition. Each of the sorted recommended data does not contain each of the attribute information contained in the currently sorted recommended data, indicating that each of the attribute information of the currently sorted recommended data is different from that of the sorted recommended data, and can be directly written into the sorting queue.
[0085] In some embodiments, if there is no sorted data in the sorting queue, the recommended data to be sorted is directly written into the sorting queue, that is, the first recommended data to be re-sorted can be directly written into the sorting queue.
[0086] When the current recommended data to be sorted is judged, the process proceeds to step 206 to traverse all the recommended data until all the recommended data are re-sorted, and a final sorting queue is generated, which is used as the recommendation result.
[0087] Specifically, for each recommended data in the data table, when it is detected that the recommended data is written into the data table, and the sum of the difference between the number of newly written recommended data in the sorting queue and the corresponding first attribute information is greater than the sorting window threshold, the recommended data is re-used as the current recommended data to be sorted.
[0088] Continuing with the above example, for the attribute information a of the recommended data 2: (aB), since it is the same as the attribute information a of the sorted sorted data 1: (aA), the recommended data 2: (aB) is written into the data table corresponding to the attribute information a, and its corresponding difference value is 1, and the sorting window threshold is 3. When it is detected that the sum of the number of newly written data in the sorting queue and the difference value 1 is not less than the sorting window threshold 3, the recommended data 2: (aB) is re-used as the current recommended data to be sorted. In other words, when the sum of the number of newly written data and the difference value is not less than the sorting window threshold, it means that the interval between the current sorting position and the recommended data 1: (aA) is greater than the sorting window threshold. When both the attribute information of the recommended data 2: (aB) and the attribute information of the newly written data meet the conditions, the recommended data 2: (aB) can be written into the sorting queue.
[0089] In a specific example, still taking the sorting window threshold of 3 as an example, the above-mentioned attribute information 1 data table 2, attribute information 2 data table 3 and sorting queue table 4 are shown below.
[0090] Table 2
[0091] Property information Attribute informationa 2:(aB) 12:(aD) Attribute informationb 4:(bB) Attribute informationc 6:(cC) 10:(cA) Attribute information 8:(dD) Attribute Information Attribute information 9:(fD) Attribute information
[0092] Table 3
[0093] Property information Attribute Information A 10:(cA) Attribute Information B 2:(aB) 4:(bB) 13:(gB) Attribute Information C 6:(cC) Attribute information D 8:(dD) 9:(fD) 12:(aD) Attribute Information
[0094] Table 4
[0095]
[0096]
[0097] In addition, in some embodiments, before all recommended data are sorted, if it is detected that the number of sorted recommended data in the sorting queue is less than a preset number threshold, the sorting window threshold is reduced so as to further reduce and re-sort the recommended data in each data table.
[0098] Exemplarily, the sorting queue is preset to need to write 10 data. After the above-mentioned reordering method 200, only 8 data are written, which does not reach the preset number. In this case, the sorting window threshold is reduced (such as from 3 to 2), and the recommended data in each data table is reordered again using the reduced sorting window threshold.
[0099] In some embodiments, if it is detected that the number of sorted recommended data in the sorting queue is less than a preset number threshold, the recommended data that has not been re-sorted and the recommended data in each data table are deleted.
[0100] Exemplarily, the sorting queue is preset to need to write 10 data. After the above re-sorting method 200, after writing 10 data, there are still recommended data to be sorted, then the remaining recommended data and the data in each data table are deleted.
[0101] The method provided by the present invention obtains the attribute information of each recommended data, and determines one by one whether each attribute information satisfies the conditions for breaking up, separates the recommended contents of similar dimensions, buffers the interaction between the recommendation system and the user, improves the user experience, and realizes multi-dimensional attribute breaking up of the recommended data, with a good degree of breaking up, excellent performance, and satisfaction of the mandatory breaking up constraints.
[0102] Figure 4 The schematic diagram of the structure of a device 400 for reordering recommended content according to an embodiment of the present invention is shown. The device 400 resides in a computing device. The device 400 includes an acquisition module 402, a judgment module 404 and a layout module 406 which are coupled to each other. The acquisition module 402 is suitable for acquiring multiple recommended data in the recommended content, wherein the recommended data includes multiple attribute information, and for each attribute information, a corresponding data table is constructed. The judgment module 404 is suitable for judging whether each attribute information included in the current recommended data to be sorted meets the conditions, and when all the attribute information meets the conditions, the recommended data is written into the sorting queue, and if any attribute information does not meet the conditions, the recommended information is written into the data table corresponding to the attribute information that does not meet the conditions. The layout module 406 is suitable for traversing all the recommended data until all the recommended data are reordered, a final sorting queue is generated, and the final sorting queue is used as the recommendation result.
[0103] It should be noted that the working principle and process of the device 400 provided in this embodiment are similar to those of the above-mentioned method 200. For relevant details, reference may be made to the description of the above-mentioned method 200, which will not be repeated here.
[0104] Figure 4 The structure diagram of the recommendation system 500 according to one embodiment of the present invention is shown. The recommendation system 500 includes the above-mentioned device 400 for reordering recommended content, and the system 500 is suitable for reordering the recommended data determined after the recall stage and the sorting stage, obtaining the reordering result of the recommended data, and outputting the recommended data according to the reordering result.
[0105] In addition, embodiments of the present invention further include: A8. A method as described in A7, further comprising: if it is detected that the number of sorted recommended data in the sorting queue is not less than a preset number threshold, the recommended data that has not been re-sorted and the recommended data in each data table are deleted. A12. A computer program product, comprising a computer program / instructions, wherein the computer program / instructions implement the method as described in any one of A1-A7 when executed by a processor. A13. A readable storage medium storing program instructions, which, when read and executed by a computing device, enables the computing device to execute the method as described in any one of A1-A7.
[0106] The various techniques described herein may be implemented in combination with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or some aspects or parts of the method and apparatus of the present invention may be in the form of program codes (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, a USB flash drive, a floppy disk, a CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.
[0107] In the case where the program code is executed on a programmable computer, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store the program code; the processor is configured to execute the method of the present invention according to the instructions in the program code stored in the memory.
[0108] By way of example and not limitation, readable media include readable storage media and communication media. Readable storage media stores information such as computer readable instructions, data structures, program modules or other data. Communication media generally embody computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and include any information transfer medium. Any combination of the above is also included within the scope of readable media.
[0109] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages can be utilized to implement the content of the present invention described herein, and the above description of specific languages is for the purpose of disclosing the preferred embodiment of the present invention.
[0110] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.
[0111] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.
[0112] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.
[0113] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.
[0114] In addition, those skilled in the art will appreciate that, although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims below, any one of the claimed embodiments may be used in any combination.
[0115] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.
[0116] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.
[0117] Although the present invention has been described according to a limited number of embodiments, it will be apparent to those skilled in the art, with the benefit of the above description, that other embodiments may be envisioned within the scope of the invention thus described. In addition, it should be noted that the language used in this specification is selected primarily for readability and teaching purposes, rather than for explaining or defining the subject matter of the present invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the present invention is illustrative and not restrictive with respect to the scope of the present invention, and the scope of the present invention is defined by the appended claims.
Claims
1. A method for reordering recommended content, executed in a computing device, the method comprising: Acquire multiple pieces of recommendation data in the recommendation content, wherein the recommendation data includes multiple pieces of attribute information, and for each piece of attribute information, a corresponding data table is constructed; Determine whether each attribute information contained in the recommended data to be sorted meets the conditions, and if all attribute information meets the conditions, write the recommended data into the sorting queue; if any attribute information does not meet the conditions, write the recommended information into the data table corresponding to the attribute information that does not meet the conditions; All the recommended data are traversed until all the recommended data are re-sorted, a final sorting queue is generated, and the final sorting queue is used as the recommendation result.
2. The method of claim 1, wherein: Determine whether the attribute information contained in the current recommended data to be sorted meets the conditions, including: For each attribute information of the recommended data to be sorted, determine whether the attribute information meets a sorting window threshold, where the sorting window threshold indicates a sorting interval corresponding to each attribute information; If so, continue to determine whether other attribute information of the recommended data meets the sorting window threshold; If not satisfied, the recommended data is written into the data table of the attribute information.
3. The method of claim 2, wherein: Determine whether the attribute information meets the sorting window threshold, including: Determine the sorting sequence number of the sorted recommended data with the attribute information in the sorting queue; Calculate the difference between the maximum value among the sorting sequence numbers and the sequence number corresponding to the current entry position in the sorting queue; Determine whether the difference is greater than a sorting window threshold corresponding to the attribute information; If it is greater, it is confirmed that the attribute information meets the sorting window threshold; If not, it is confirmed that the attribute information does not meet the sorting window threshold.
4. The method of claim 2, wherein: Traversing all the recommended data until all the recommended data are re-ranked includes: For each recommended data in the data table, when it is detected that the recommended data is written into the data table, and the sum of the difference between the number of newly written recommended data in the sorting queue and the corresponding attribute information pointed to by the data table is greater than the sorting window threshold, the recommended data is re-used as the current recommended data to be sorted.
5. The method of claim 2, wherein: Also includes: If each of the sorted recommended data does not include each of the attribute information included in the currently recommended data to be sorted, it is confirmed that each of the attribute information of the currently recommended data to be sorted meets the conditions.
6. The method of claim 1, wherein: Also includes: If there is no sorted recommended data in the sorting queue, the currently recommended data to be sorted is directly written into the sorting queue.
7. The method of claim 2, wherein: Also includes: After all the recommended data are sorted, if it is detected that the number of sorted recommended data in the sorting queue is less than a preset number threshold, the sorting window threshold is reduced so as to re-sort the recommended data in each data table.
8. A device for reordering recommended content, resident in a computing device, the device comprising: An acquisition module, adapted to acquire a plurality of recommendation data in the recommendation content, wherein the recommendation data includes a plurality of attribute information, and a corresponding data table is constructed for each attribute information; A judgment module, adapted to judge whether each attribute information contained in the recommended data to be sorted satisfies the conditions, and when all the attribute information satisfies the conditions, writes the recommended data into the sorting queue; if any attribute information does not satisfy the conditions, writes the recommended information into the data table corresponding to the attribute information that does not satisfy the conditions; The sorting module is adapted to traverse all the recommended data until all the recommended data are re-sorted, generate a final sorting queue, and use the final sorting queue as the recommendation result.
9. A recommendation system, comprising a device for reordering recommended content as described in claim 8, suitable for reordering recommended data determined after a recall phase and a sorting phase, obtaining a reordering result of the recommended data, and outputting the recommended data according to the reordering result.
10. A computing device comprising: at least one processor; and A memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, and the program instructions include instructions for executing the method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Method and device for reordering recommended information
CN104239468A
Sorting processing method and device, electronic equipment and storage medium
CN116186435A
Sequence recommendation method based on spatial position and time span
CN117891999A
