Resource allocation method and resource allocation device

Abstract

The embodiment of the present application provides a resource allocation method and a resource allocation device, which belongs to the field of communication technology. The resource allocation method includes: when receiving a recommended service request from a target platform, determining a recommended queue length of the recommended service request according to the resource amount of the target platform; and outputting a recommendation result according to the recommended queue length.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a resource allocation method and a resource allocation device. Background Art

[0002] Currently, many Internet businesses have high real-time requirements, such as application distribution, content distribution, and product recommendations. In these recommendation services, the recommendation candidate queue generally goes through the stages of retrieval, recall, rough sorting, and fine sorting, and finally selects one or a series of applications, content, and products to be presented to the user. In order to meet system performance requirements and improve system real-time performance, the recommendation candidate queue is truncated at each stage.

[0003] In the related art, for all users and all time periods, fixed truncation is used in each link. For example, 100,000 applications are recalled in full, and the recommendation candidate queue is truncated, 5,000 applications enter the rough sorting queue, and 1,000 applications enter the fine sorting queue. However, for recommendation candidate queues with different attribute parameters, if the same truncation method is used, a high number of candidates with low attribute parameters will waste limited machine resources, and a low number of candidates with high-quality traffic will not be able to fully utilize resources, resulting in a problem of resource waste. Summary of the invention

[0004] The embodiments of the present application provide a resource allocation method and a resource allocation device, which can adopt different candidate truncation methods for traffic of different qualities, globally consider the allocation between the overall traffic quality and platform machine resources, and optimize the recommended queue allocation.

[0005] In a first aspect, an embodiment of the present application provides a resource allocation method, including:

[0006] When receiving a recommended service request from a target platform, determining a recommended queue length of the recommended service request according to the resource amount of the target platform;

[0007] Output the recommended results according to the recommended queue length.

[0008] In a second aspect, an embodiment of the present application provides a resource allocation device, including:

[0009] A determination module, configured to determine, upon receiving a recommended service request from a target platform, a recommended queue length of the recommended service request according to the amount of resources of the target platform;

[0010] The output module is used to output the recommendation results according to the recommended queue length.

[0011] In a third aspect, an embodiment of the present application provides an electronic device, including a resource allocation device provided by an embodiment of the second aspect.

[0012] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a program or instruction stored in the memory and running on the processor, which, when executed by the processor, implements the steps of the resource allocation method provided in the first aspect.

[0013] In a fifth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the resource allocation method provided in the first aspect are implemented.

[0014] In a sixth aspect, an embodiment of the present application provides a chip, the chip including a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the steps of the resource allocation method provided in the first aspect.

[0015] In an embodiment of the present application, the resource allocation method includes: upon receiving a recommended service request from a target platform, determining the recommended queue length of the recommended service request according to the amount of resources of the target platform; and outputting the recommendation result according to the recommended queue length. Thus, the amount of resources of the target platform is fully considered, the corresponding recommended queue length (the number of candidates for the recommended queue) is configured for the recommended service request, and the recommendation result is output according to the recommended queue length. Furthermore, in the case where the machine resources of the target platform are limited, the length of the recommended queue with low attribute parameters (such as quality flow, recommendation success rate) is reduced, that is, the resources required for the recommended queue are reduced, so that as many resources as possible are allocated to the recommended queue with high attribute parameters. Without increasing the number of machines, the allocation between the overall traffic quality and the amount of platform resources is considered globally, the recommended queue allocation is optimized, and the platform recommendation efficiency is maximized. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 One of the flow charts of a resource allocation method according to an embodiment of the present application is shown;

[0017] Figure 2 A second flowchart of a resource allocation method according to an embodiment of the present application is shown;

[0018] Figure 3 A third flowchart of a resource allocation method according to an embodiment of the present application is shown;

[0019] Figure 4 A fourth flowchart of a resource allocation method according to an embodiment of the present application is shown;

[0020] Figure 5 A fifth flowchart of a resource allocation method according to an embodiment of the present application is shown;

[0021] Figure 6FIG6 shows a sixth flowchart of a resource allocation method according to an embodiment of the present application;

[0022] Figure 7 FIG7 shows a seventh flow chart of a resource allocation method according to an embodiment of the present application;

[0023] Figure 8 A structural block diagram of a resource allocation device according to an embodiment of the present application is shown;

[0024] Fig. 9 A structural block diagram of an electronic device according to an embodiment of the present application is shown;

[0025] Fig.10 A hardware structure block diagram of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0027] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0028] Refer to the following Figures 1 to 10 A resource allocation method and a resource allocation device according to some embodiments of the present application are described.

[0029] In one embodiment of the present application, Figure 1 One of the flow charts of the resource allocation method according to an embodiment of the present application is shown, including:

[0030] Step 102, when receiving a recommended service request from a target platform, determining a recommended queue length of the recommended service request according to the resource amount of the target platform;

[0031] The recommendation queue is a candidate set of recommendation results. The recommendation queue contains one or more recommendation results. The recommendation result can be an application, a product, a content, etc. The recommendation service request indicates the application, product or content to be recommended. The recommendation queue length is the number of recommendation result candidates. For example, the longer the recommendation queue length, the more recommendation results are contained in the recommendation queue.

[0032] It is understandable that in the process of recommendation business, the recommendation queue usually goes through at least one link, such as retrieval, recall, rough sorting, and fine sorting. Each link has a corresponding proportional coefficient. The sum of the proportional coefficients of all links is 1. The proportional coefficients of different links can be the same or different. In the case that the recommendation business has multiple links, the amount of resources allocated by the target platform to each link is obtained by multiplying the total amount of resources of the target platform and the proportional coefficient corresponding to each link, and then the length of the recommendation queue for each link is determined according to the amount of resources allocated by the target platform to each link.

[0033] Step 104: output the recommendation result according to the recommended queue length.

[0034] In this embodiment, the target platform resources are fully considered, the corresponding recommended queue length (the number of candidates for the recommended queue) is configured for the recommended service request, and the recommended result is output according to the recommended queue length to complete the feedback of the recommended service request. Furthermore, when the machine resources of the target platform are limited, the length of the recommended queue with low attribute parameters is reduced, that is, the resources required for the recommended queue are reduced, so that as many resources as possible are allocated to the recommended queue with high attribute parameters. Without increasing the number of machines, the allocation between the overall traffic quality and the platform resources is considered globally, the recommended queue allocation is optimized, and the platform recommendation efficiency is maximized.

[0035] In actual applications, when the terminal sends a recommendation service request to a server (target platform) for providing recommendation services, the server can select a recommendation algorithm for the recommendation service request, determine a recommendation result for the recommendation service request based on the recommendation algorithm, and configure the recommendation queue length in combination with the server resource amount in the time period, and send the recommendation result to the terminal based on the recommendation queue length. In this way, the terminal can respond according to the recommendation result, such as clicking to browse the recommendation result (that is, successfully responding to the recommendation result). Of course, the terminal can also choose to ignore the recommendation result, that is, not to respond to the recommendation result or not to respond successfully.

[0036] Specifically, the resource allocation method is applicable to electronic devices, including but not limited to mobile terminals, tablet computers, computers, laptop computers, wearable devices, vehicle-mounted terminals, etc.

[0037] In one embodiment of the present application, Figure 2As shown, step 102, determining the recommended queue length of the recommended service request according to the resource amount of the target platform, includes:

[0038] Step 202, obtaining multiple service categories of the target platform;

[0039] Among them, the attribute parameters of different business categories in the target platform are different. The attribute parameters include: quality flow, recommendation success rate, etc. The attribute parameters need to be determined based on the response feedback of the collected historical recommendation results. The larger the attribute parameter, the greater the business click volume of this category, the better the conversion effect of the recommendation result, and the higher the recommendation income. Multiple business categories of the target platform can be obtained based on the historical business data analysis of the target platform.

[0040] It should be noted that since attribute parameters such as quality traffic and recommendation success rate are affected by many factors, they are often not stable in a short period of time. For example, the recommendation success rate is relatively low every day before a promotion, but on the day of the promotion, the recommendation success rate suddenly increases a lot. The recommendation success rate on the day of the promotion cannot accurately represent the recommendation success rate of the recommendation business. Therefore, in the specific statistical process, multiple different time periods are usually selected for optimization, and each time period has the same statistical end time and a different statistical start time.

[0041] Step 204, determining the target business category to which the data sample of the recommended business request belongs;

[0042] The multiple business categories include a target business category.

[0043] In this embodiment, by comparing the data sample of the current service request with multiple service categories of the target platform, the target service category most similar to the current service request is determined. The target service category can be used to determine the attribute parameters of the current service request, so that the platform resources can be reasonably allocated through the attribute parameters in the future to maximize the benefits.

[0044] Step 206, determining the initial queue length of the recommended service request according to the resource amount of the target platform;

[0045] In this embodiment, the target platform is preset with a default resource allocation policy, based on which the same resources are allocated to all service requests. Thus, according to the target platform's resource volume and the default resource allocation policy, the initial queue length of the recommended service request is determined, that is, the default recommended queue length.

[0046] Specifically, the resource allocation strategy includes a recommended queue allocation model, as shown in formula (1):

[0047]

[0048] Among them, Q represents the amount of resources of the target platform of the corresponding link. The value of Q is constrained by the platform machine resources and is related to the number of recommended service requests and the length of the recommended queue in the same period. req represents the recommended service request, t represents the time period in which the recommended service request is located, i represents the number of all recommended service requests initiated in the same time period, and count represents the length of the recommended queue of the corresponding link. Different links have different recommended queue lengths.

[0049] Step 208: Determine the recommended queue length according to the initial queue length and the length coefficient corresponding to the target service category.

[0050] The larger the attribute parameter of the target service category is, the larger the length coefficient is, so as to ensure that the recommended queue with high attribute parameters can obtain more resources.

[0051] In this embodiment, the default initial queue length is modified according to the length coefficient corresponding to the target service category to convert the recommended queue length required by the attribute parameters suitable for the current service request. Therefore, different truncation methods can be used for traffic of different qualities, so that the higher the traffic quality, the larger the count. Furthermore, when the machine resources of the target platform are limited, the recommended queue length of low attribute parameters is reduced, that is, the resources required for the recommended queue are reduced, so that as many resources as possible are allocated to the recommended queue of high attribute parameters. Without increasing the number of machines, the allocation between the overall traffic quality and the platform resources is considered globally, the recommended queue allocation is optimized, and the platform recommendation efficiency is maximized.

[0052] Specifically, according to the initial queue length and the length coefficient corresponding to the target service category, the recommended queue length is determined using the following formula (2):

[0053] count=count0×(1+r), (2)

[0054] Where count represents the recommended queue length, count0 represents the initial queue length, and r represents the length coefficient.

[0055] In one embodiment of the present application, Figure 3 As shown, step 204, determining the target business category to which the data sample of the recommended business request belongs, includes:

[0056] Step 302, determining a first distance between the data sample and a cluster center of a cluster of each service category in a plurality of service categories;

[0057] Step 304: taking the service category corresponding to the minimum distance value among the plurality of first distances as the target service category.

[0058] In this embodiment, the first distance between the data sample of the recommended business request and the cluster center (centroid vector) of each business category in the multiple business categories is calculated, and the similarity between the current recommended business request and the multiple business categories is determined by the first distance between the data sample and each cluster center. The smaller the first distance, the higher the similarity. The business category corresponding to the minimum distance value among the multiple first distances is taken as the target business category closest to the current recommended business request, that is, the recommended business request is classified into the cluster of the target business category. Therefore, the attribute parameters of the current business request can be determined through the target business category, so that the platform resources can be reasonably allocated through the attribute parameters in the future to maximize the benefits.

[0059] In one embodiment of the present application, Figure 4 As shown, step 202, obtaining multiple business categories of the target platform, including:

[0060] Step 402, obtaining historical business data of the target platform;

[0061] The historical business data includes information such as historical business requests of the target platform, historical recommendation queues and revenue corresponding to the historical business requests, etc.

[0062] Step 404: cluster the historical business data to determine multiple business categories of the target platform.

[0063] In this embodiment, the historical business data of the platform is taken as a data set, and the historical business data is clustered to count multiple business categories that the target platform has ever executed, so as to determine the attribute parameters corresponding to different business categories, thereby better allocating more resources to recommended business requests with higher recommendation success rates or quality traffic.

[0064] For example, K-means is used to cluster to obtain multiple business categories. K-means is a clustering algorithm based on Euclidean distance, which believes that the closer the distance between two targets is, the greater the similarity is. Specifically, k first historical data samples and m second historical data samples x in the historical business data are determined. m , where k>1, m≥1; the k first historical data samples are random k historical data samples from all historical data samples in the historical business data, and the m second historical data samples are historical data samples from all historical data samples except the first historical data samples. Take the k first historical data samples as the clustering center (μ j , j = 1, 2, ..., k) to establish k clusters (S j =φj=1, 2, ..., k), a cluster is a collection of similar historical data samples. Calculate the second distance d between each second historical data sample and the cluster center of each cluster in the multiple clusters. ij, clustering each second historical data sample into the cluster corresponding to the minimum distance value among the multiple second distances, that is, the cluster with the minimum second distance d ij The clustering category λ i The second historical data sample x is marked as i Most similar categories and categorized Since the i second historical data samples are continuously classified into clusters, the cluster centers of the clusters are updated, that is, the centroid vectors of the clusters are recalculated. Repeat the above steps. When the third distance between the cluster center before the update and the cluster center after the update of any cluster is less than or equal to the preset distance, it means that all k cluster centers have not changed. At this time, the final clusters S of multiple business categories are obtained. j ={S1, S2, ..., S k}, and take the category of any cluster as a business category.

[0065] The second distance is calculated using the following formula (3):

[0066]

[0067] The centroid vector of the cluster is calculated as follows (4):

[0068]

[0069] In one embodiment of the present application, Figure 5 The fifth flowchart of the resource allocation method according to the embodiment of the present application is shown, including:

[0070] Step 502, sorting multiple business categories according to a preset sorting rule;

[0071] For example, the preset sorting rule can be to sort from high to low according to the value of business requests (the ratio of revenue to the number of requests), and the value of different business requests reflects the size of the attribute parameters of the category of business. Among them, the calculation method of revenue is related to the platform business. Taking the distribution of mobile phone manufacturers' applications as an example, the distribution revenue is the distribution income; taking the shopping platform as an example, the product distribution revenue is the platform revenue; taking message distribution as an example, the message distribution revenue is the user's consumption time or number of articles.

[0072] Step 504: according to the arrangement order of the multiple service categories, the initial length coefficient and the preset offset, a length coefficient corresponding to each service category in the multiple service categories is set.

[0073] Specifically, the initial length coefficient and the preset offset are determined according to the preset sorting rules, or are reasonably set according to user needs. For example, if the preset sorting rules are to sort from high to low according to the service request value, at this time, the higher the value created by the service category at the first position in the arrangement order, the larger the attribute parameters of the service category, and the initial length coefficient is the length coefficient corresponding to the first service category, and the length coefficients of other service categories are based on the length coefficient corresponding to the previous service category in the arrangement order, and the preset offset is the change amount, which gradually decreases. On the contrary, if the preset sorting rules are to sort from low to high according to the service request value, at this time, the lower the value created by the service category at the first position in the arrangement order, the smaller the attribute parameters of the service category, and the initial length coefficient is the length coefficient corresponding to the first service category, and the length coefficients of other service categories are based on the length coefficient corresponding to the previous service category in the arrangement order, and the preset offset is the change amount, which gradually increases.

[0074] It is understandable that, assuming that there are 10 service categories, they are sorted from high to low according to the service request value. The corresponding 10 preset offsets can be the same or different. The calculated length coefficients are shown in Table 1 below:

[0075] Table 1

[0076] Serial number Length factor Serial number Length factor 1 0.5 6 -0.1 2 0.4 7 -0.2 3 0.3 8 -0.3 4 0.2 9 -0.4 5 0.1 10 -0.5

[0077] Among them, the value of the service category corresponding to sequence numbers 1 to 5 is higher, that is, the attribute parameter is larger, and more resources can be allocated to it. The initial length coefficient and the preset offset are reasonably set to make the length coefficient of the service category corresponding to sequence numbers 1 to 5 greater than 0. Similarly, the value of the service category corresponding to sequence numbers 6 to 10 is lower, that is, the attribute parameter is smaller. In order to avoid resource waste, the resources of this service category can be appropriately reduced. The initial length coefficient and the preset offset are reasonably set to make the length coefficient of the service category corresponding to sequence numbers 1 to 5 less than 0.

[0078] In this embodiment, the corresponding length coefficients are set for different service categories by presetting the sorting rules, initial length coefficients and presetting the offsets to ensure that the recommendation queues with high attribute parameters can obtain more resources. Thus, without adding more machines, the allocation between the overall traffic quality and the platform resources is considered globally, and more resources can be better allocated to the recommendation service requests with a higher recommendation success rate, thereby improving the optimization efficiency and accuracy.

[0079] In one embodiment of the present application, Figure 6 The sixth flowchart of the resource allocation method according to the embodiment of the present application is shown, including:

[0080] Step 602: When the target service category belongs to the preset service category, the length coefficient is corrected according to the difference between the resource amount and the upper limit of the resource amount of the target platform.

[0081] In this embodiment, the preset service category includes a service category that needs to increase the queue length based on the initial queue length. Considering that the system can allocate more resources to recommended service requests with larger attribute parameters, when determining the recommended queue length, it may happen that the resources required to achieve the recommended queue length have exceeded the upper limit of the resources that the target platform can provide, which is not conducive to the effective allocation of resources. For this reason, when the target service category belongs to the preset service category, the length coefficient is corrected through PID calculation according to the difference between the resource amount and the upper limit of the resource amount of the target platform. Thereby, while ensuring that the amount of resources obtained by the recommended service request with larger attribute parameters will not exceed the maximum amount of resources of the corresponding link, as many resources as possible are allocated to the recommended queue with high attribute parameters, which is conducive to improving the optimization efficiency and accuracy.

[0082] Specifically, the maximum Q of each link platform can be determined by referring to formula (1) and the historical business data of the target platform: max (Upper limit of resource volume), different links correspond to Q max Different, for example, the recall phase corresponds to Q max1 , the coarse-grained link corresponds to Q max2 , the fine sorting link corresponds to Q max3 Etc. Among them, select the moment when the platform machine resource utilization rate is the highest and determine the req at that moment (i,t) ×count (i,t) , as the maximum Q of the target platform in this link max .

[0083] In one embodiment of the present application, Figure 7 The seventh flowchart of the resource allocation method according to the embodiment of the present application is shown, including:

[0084] Step 702: When the length of the recommendation queue is greater than or equal to a preset length, the recommendation result is output according to the preset length.

[0085] The preset length may be determined according to the target service category.

[0086] In this embodiment, in some special scenarios, if the number of candidate recommendation results is too large, it may cause disturbance to the terminal user. For example, in the scenario of promoting advertising content, users prefer to know the latest advertising content and do not need to recommend all relevant advertising content. At this time, when the length of the recommendation queue is greater than or equal to the preset length, it means that there are too many candidate recommendation results in the recommendation queue, and there may be invalid recommendations. The length of the recommendation queue can be further truncated by the preset length. Then, on the basis of ensuring the recommendation effect, the target platform resources required for the business request are saved, more allocable resources are provided for other business requests, and resource allocation is further optimized to maximize distribution benefits.

[0087] In one embodiment of the present application, Figure 8 As shown, the resource allocation device 800 includes: a determination module 802, which is used to determine the recommended queue length of the recommended service request based on the resource amount of the target platform when receiving the recommended service request of the target platform; and an output module 804, which is used to output the recommendation result according to the recommended queue length.

[0088] In this embodiment, the target platform resources are fully considered, the corresponding recommended queue length (the number of candidates for the recommended queue) is configured for the recommended service request, and the recommended result is output according to the recommended queue length to complete the feedback of the recommended service request. Furthermore, when the machine resources of the target platform are limited, the length of the recommended queue with low attribute parameters is reduced, that is, the resources required for the recommended queue are reduced, so that as many resources as possible are allocated to the recommended queue with high attribute parameters. Without increasing the number of machines, the allocation between the overall traffic quality and the platform resources is considered globally, the recommended queue allocation is optimized, and the platform recommendation efficiency is maximized.

[0089] Optionally, the resource allocation device 800 also includes: an acquisition module (not shown in the figure), the acquisition module is used to obtain multiple business categories of the target platform; the determination module 802 is also used to determine the target business category to which the data sample of the recommended business request belongs, and the multiple business categories include the target business category; according to the amount of resources of the target platform, the initial queue length of the recommended business request is determined; according to the initial queue length and the length coefficient corresponding to the target business category, the recommended queue length is determined.

[0090] Optionally, the determination module 802 is further configured to determine a first distance between the data sample and the cluster center of each business category cluster; and take the business category corresponding to the minimum distance value among the multiple first distances as the target business category.

[0091] Optionally, the acquisition module is also used to acquire historical business data of the target platform; the resource allocation device 800 also includes: a clustering module (not shown in the figure), which is used to cluster the historical business data and determine multiple business categories of the target platform.

[0092] Optionally, the resource allocation device 800 also includes: a sorting module (not shown in the figure), which is used to sort multiple business categories according to preset sorting rules; a setting module (not shown in the figure), which is used to set the length coefficient corresponding to each business category in the multiple business categories according to the arrangement order of the multiple business categories, the initial length coefficient and the preset offset; wherein the initial length coefficient and the preset offset are determined according to the preset sorting rules.

[0093] Optionally, the resource allocation device 800 further includes: a correction module (not shown in the figure), which is used to correct the length coefficient according to the difference between the resource amount and the upper limit of the resource amount of the target platform when the target service category belongs to a preset service category.

[0094] Optionally, the output module 804 is further configured to output the recommendation result according to a preset length when the length of the recommendation queue is greater than or equal to a preset length.

[0095] In this embodiment, each module of the resource allocation device 800 implements the steps of the resource allocation method in any embodiment of the first aspect when performing its respective function. Therefore, the resource allocation device 800 also includes all the beneficial effects of the resource allocation method in any embodiment of the first aspect, which will not be repeated here.

[0096] The resource allocation device in the embodiment of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV) or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0097] The application management device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other operating systems, which are not specifically limited in the embodiment of the present application.

[0098] In one embodiment of the present application, the present application embodiment provides an electronic device, including the second aspect embodiment provides a resource allocation device. Therefore, the electronic device includes all the beneficial effects of the resource allocation device provided in any of the above embodiments, which will not be repeated here.

[0099] In one embodiment of the present application, Fig. 9 As shown, an electronic device 900 is provided, including: a processor 904, a memory 902, and a program or instruction stored in the memory 902 and running on the processor 904. When the program or instruction is executed by the processor 904, the steps of the resource allocation method provided in any of the above embodiments are implemented. Therefore, the electronic device 900 includes all the beneficial effects of the resource allocation method provided in any of the above embodiments, which will not be repeated here.

[0100] It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0101] Fig.10 The hardware structure diagram of an electronic device 1000 for implementing an embodiment of the present application is shown in the following figure. The electronic device 1000 includes, but is not limited to, a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and a processor 1010.

[0102] Those skilled in the art will appreciate that the electronic device 1000 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 1010 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Fig.10 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0103] The processor 1010 is used to determine the recommended queue length of the recommended service request according to the resource amount of the target platform when receiving the recommended service request of the target platform; the radio frequency unit 1001 is used to output the recommendation result according to the recommended queue length.

[0104] In this embodiment, the target platform resources are fully considered, the corresponding recommended queue length (the number of candidates for the recommended queue) is configured for the recommended service request, and the recommended result is output according to the recommended queue length to complete the feedback of the recommended service request. Furthermore, when the machine resources of the target platform are limited, the length of the recommended queue with low attribute parameters is reduced, that is, the resources required for the recommended queue are reduced, so that as many resources as possible are allocated to the recommended queue with high attribute parameters. Without increasing the number of machines, the allocation between the overall traffic quality and the platform resources is considered globally, the recommended queue allocation is optimized, and the platform recommendation efficiency is maximized.

[0105] Furthermore, processor 1010 is also used to obtain multiple business categories of the target platform; determine the target business category to which the data sample of the recommended business request belongs, and the multiple business categories include the target business category; determine the initial queue length of the recommended business request based on the resource amount of the target platform; determine the recommended queue length based on the initial queue length and the length coefficient corresponding to the target business category.

[0106] Furthermore, the processor 1010 is further configured to determine a first distance between the data sample and a cluster center of each cluster of the service category; and take the service category corresponding to the minimum distance value among the multiple first distances as the target service category.

[0107] Furthermore, the processor 1010 is further configured to obtain historical business data of the target platform; perform clustering processing on the historical business data, and determine multiple business categories of the target platform.

[0108] Furthermore, the processor 1010 is also used to sort multiple business categories according to a preset sorting rule; set the length coefficient corresponding to each business category in the multiple business categories according to the arrangement order of the multiple business categories, the initial length coefficient and the preset offset; wherein the initial length coefficient and the preset offset are determined according to the preset sorting rule.

[0109] Furthermore, the processor 1010 is further configured to correct the length coefficient according to a difference between the resource amount and an upper limit value of the resource amount of the target platform when the target service category belongs to a preset service category.

[0110] Furthermore, the RF unit 1001 is further configured to output a recommendation result according to a preset length when the recommended queue length is greater than or equal to a preset length.

[0111] It should be understood that in the embodiment of the present application, the input unit 1004 may include a graphics processor (Graphics Processing Unit, GPU) 1041 and a microphone 1042, and the graphics processor 1041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode. The display unit 1006 may include a display panel 1061, and the display panel 1061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 1007 includes a touch panel 1071 and other input devices 1072. The touch panel 1071 is also called a touch screen. The touch panel 1071 may include two parts: a touch detection device and a touch controller. Other input devices 1072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here. The memory 1009 can be used to store software programs and various data, including but not limited to applications and operating systems. The processor 1010 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the modem processor may not be integrated into the processor 1010.

[0112] In one embodiment of the present application, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the resource allocation method provided in any of the above embodiments are implemented.

[0113] In this embodiment, reading the storage medium can implement each process of the resource allocation method provided in the embodiment of the present application and can achieve the same technical effect. To avoid repetition, it will not be described here.

[0114] The processor is the processor in the communication device in the above embodiment. The read storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0115] An embodiment of the present application also provides a chip, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned resource allocation method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0116] It should be understood that the chip mentioned in the embodiments of the present application may also be referred to as a system-level chip, a system chip, a chip system, or a system-on-chip chip, etc.

[0117] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes 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 one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0119] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A resource allocation method, characterized in that: include: When the target platform receives the recommended service request, determining the recommended queue length of the recommended service request according to the resource amount of the target platform; Outputting the recommendation result according to the recommended queue length; The determining, according to the amount of resources of the target platform, the recommended queue length of the recommended service request includes: Acquire multiple business categories of the target platform; Determine a target service category to which the data sample of the recommended service request belongs, wherein the multiple service categories include the target service category; Determining an initial queue length of the recommended service request according to the resource amount of the target platform; The recommended queue length is determined according to the initial queue length and a length coefficient corresponding to the target service category, wherein the larger the attribute parameter of the target service category is, the larger the length coefficient is.

2. The resource allocation method according to claim 1, characterized in that: The determining the target service category to which the data sample of the recommended service request belongs includes: Determining a first distance between the data sample and a cluster center of a cluster of each business category; The service category corresponding to the minimum distance value among the multiple first distances is used as the target service category.

3. The resource allocation method according to claim 1, characterized in that: The acquiring of multiple business categories of the target platform includes: Acquire historical business data of the target platform; Clustering is performed on the historical business data to determine the multiple business categories of the target platform.

4. The resource allocation method according to claim 1, characterized in that: Also includes: Sorting the multiple business categories according to a preset sorting rule; According to the arrangement order of the multiple service categories, the initial length coefficient and the preset offset, setting the length coefficient corresponding to each service category in the multiple service categories; The initial length coefficient and the preset offset are determined according to the preset sorting rule.

5. A resource allocation device, characterized in that: include: A determination module, configured to determine, when a target platform receives a recommended service request, a recommended queue length of the recommended service request according to the amount of resources of the target platform; An output module, used to output a recommendation result according to the recommendation queue length; An acquisition module, used for acquiring multiple business categories of the target platform; The determination module is further configured to determine a target service category to which the data sample of the recommended service request belongs, wherein the multiple service categories include the target service category; Determining an initial queue length of the recommended service request according to the resource amount of the target platform; The recommended queue length is determined according to the initial queue length and a length coefficient corresponding to the target service category, wherein the larger the attribute parameter of the target service category is, the larger the length coefficient is.

6. The resource allocation device according to claim 5, characterized in that: The determination module is further used to determine a first distance between the data sample and a cluster center of a cluster of each business category; The service category corresponding to the minimum distance value among the multiple first distances is used as the target service category.

7. The resource allocation device according to claim 5, characterized in that: The acquisition module is further used to acquire the historical business data of the target platform; The resource allocation device also includes: A clustering module is used to perform clustering processing on the historical business data to determine the multiple business categories of the target platform.

8. The resource allocation device according to claim 5, characterized in that: Also includes: A sorting module, used to sort the multiple business categories according to a preset sorting rule; A setting module, used to set the length coefficient corresponding to each service category in the multiple service categories according to the arrangement order of the multiple service categories, the initial length coefficient and the preset offset; The initial length coefficient and the preset offset are determined according to the preset sorting rule.

Images