Method and device for determining evaluation parameters, electronic device and storage medium

Abstract

The present disclosure provides a method and device for determining an evaluation parameter, electronic equipment and storage medium, relates to the technical field of data processing, and particularly relates to the technical field of big data and artificial intelligence. The specific implementation scheme is as follows: obtaining an access data set obtained by accessing fragmented data; the fragmented data is obtained by dividing business data in a business data set based on a historical evaluation parameter, and the historical evaluation parameter is used to measure the access characteristics of the business data in the business data set; determining a target common feature between target access data in the access data set; selecting an evaluation parameter associated with the target common feature from a plurality of preset evaluation parameters as a target evaluation parameter; and adjusting the historical evaluation parameter based on the target evaluation parameter. According to the present disclosure, the adjustment of the evaluation parameter can be realized.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to the fields of big data and artificial intelligence. Background Technology

[0002] In related technologies, evaluation parameters can be used to assess the hotness or coldness of data, but these parameters are usually fixed. However, in practical applications, the hotness or coldness of data can change over time, and fixed evaluation parameters cannot accurately assess the hotness or coldness of data. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, and storage medium for determining evaluation parameters.

[0004] According to one aspect of this disclosure, a method for determining evaluation parameters is provided, comprising:

[0005] Obtain the access dataset obtained by accessing the sharded data; wherein the sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, and the historical evaluation parameters are used to measure the access characteristics of the business data in the business dataset;

[0006] Determine the common characteristics of the target access data among the target access data in the access dataset;

[0007] From a set of preset evaluation parameters, select the evaluation parameter that is associated with the common features of the target as the target evaluation parameter;

[0008] The historical evaluation parameters are adjusted based on the target evaluation parameters.

[0009] According to another aspect of this disclosure, an apparatus for determining evaluation parameters is provided, comprising:

[0010] The acquisition unit is used to acquire the access dataset obtained by accessing the sharded data; wherein the sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, and the historical evaluation parameters are used to measure the access characteristics of the business data in the business dataset.

[0011] A determining unit is used to determine the common features among the target access data in the access dataset;

[0012] The selection unit is used to select, from a plurality of preset evaluation parameters, an evaluation parameter associated with the common features of the target as the target evaluation parameter;

[0013] An adjustment unit is used to adjust the historical evaluation parameters based on the target evaluation parameters.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to said at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods disclosed above.

[0016] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided that stores computer instructions for causing a computer to perform the methods described above.

[0017] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the methods described above.

[0018] According to this disclosure, the evaluation parameters can be adjusted to achieve an accurate evaluation of the access characteristics of business data.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0020] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0021] Figure 1 This is a schematic diagram of an embodiment of the method for determining evaluation parameters according to this disclosure;

[0022] Figure 2 This is a schematic diagram of another embodiment of the method for determining evaluation parameters according to this disclosure;

[0023] Figure 3 This is a schematic diagram of an embodiment of obtaining sharded data based on historical evaluation parameters according to this disclosure;

[0024] Figure 4 This is a schematic diagram illustrating the determination of the first evaluation parameter based on prior information according to this disclosure;

[0025] Figure 5 This is a schematic diagram illustrating the method for determining evaluation parameters in a search system according to this disclosure;

[0026] Figure 6 This is a diagram illustrating the evaluation parameters used to measure the access characteristics of business data based on online business, in accordance with this disclosure.

[0027] Figure 7This is a diagram illustrating the grouping and segmentation within the business data disclosed herein;

[0028] Figure 8 This is a schematic diagram of the structural composition of the device for determining the evaluation parameters according to this disclosure;

[0029] Figure 9 This is a block diagram of an electronic device used to implement an embodiment of the method for determining evaluation parameters of this disclosure. Detailed Implementation

[0030] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0031] The terms "first," "second," and "third," etc., used in the embodiments, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as including a series of steps or units. A method, system, product, or apparatus is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0032] This disclosure provides a method for determining evaluation parameters. This method can be applied to electronic devices, including but not limited to fixed devices and / or mobile devices. For example, fixed devices include, but are not limited to, servers, which can be cloud servers or regular servers. Mobile devices include, but are not limited to, one or more of mobile phones, desktop computers, or tablet computers. Figure 1 As shown, the method for determining this evaluation parameter includes:

[0033] S101: Obtain the access dataset obtained by accessing the sharded data; wherein the sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, which are used to measure the access characteristics of the business data;

[0034] Historical evaluation parameters and target evaluation parameters measure the access characteristics of business data in a business dataset from different dimensions, and can both be regarded as evaluation parameters used to measure the access characteristics of business data. Access characteristics refer to the frequency of business data access, which can be represented by hot or cold data. Generally speaking, data that is frequently accessed (high access frequency or many accesses) is called hot data, and data that is not accessed or has a low access frequency or few accesses is called cold data.

[0035] Segmented data refers to business data that can be accessed by users, such as through searching or querying, and is stored in the search system. When a user searches or queries business data from the search system, the searched or queried business data is recorded; this searched or queried business data is called access data. An access dataset is a collection of all business data that has been searched or queried within a recorded period. The access dataset can be obtained by reading or accessing the recorded collection of business data.

[0036] It is understandable that the sharded data itself is business data, and the access data in the access dataset obtained by accessing the business data is the accessed business data.

[0037] S102: Determine the common features of the target access data among the target access data in the access dataset;

[0038] S103: Select the evaluation parameter associated with the common characteristics of the target from a set of preset evaluation parameters as the target evaluation parameter;

[0039] Multiple evaluation parameters are used to evaluate business data from different dimensions. Historical evaluation parameters and target evaluation parameters are one or more of the preset evaluation parameters.

[0040] S104: Adjust the historical evaluation parameter based on the target evaluation parameter.

[0041] In steps S101-S104, an access dataset is acquired, and the target common features among the target access data in the access dataset are determined. From a set of preset evaluation parameters, an evaluation parameter associated with the target common features is selected as the target evaluation parameter. Historical evaluation parameters are then adjusted based on the target evaluation parameter. It can be understood that historical evaluation parameters are generated based on the user's historical access to business data. Over time, the user's access to business data may change; for example, historically cold data may become hot data, and historically hot data may become cold data. The solution of this embodiment can achieve timely and automatic adjustment of the evaluation parameters for measuring the access characteristics of business data based on access patterns (access dataset). For example, adjusting historical evaluation parameters based on the target evaluation parameter can achieve accurate evaluation of the access characteristics of business data.

[0042] For example, taking product data as business data, and product price as the historical evaluation parameter, it's clear that historically, evaluating product popularity based on price was relatively accurate. However, over time, most users searching for products may be more concerned with sales volume. Therefore, evaluating product popularity based solely on price becomes less accurate, necessitating adjustments to the evaluation parameters used to measure the access characteristics of business data. For instance, shifting the evaluation parameter from product price (which can serve as a historical evaluation parameter) to product sales volume (which can serve as a target evaluation parameter) makes evaluating product popularity more accurate. This allows for timely adjustments to the evaluation parameters, adapting to actual user search behavior.

[0043] In this embodiment of the disclosure, the scheme for adjusting historical evaluation parameters based on target evaluation parameters can be implemented through one of the following methods:

[0044] The first implementation method involves replacing historical evaluation parameters with target evaluation parameters to adjust historical evaluation parameters based on the target evaluation parameters. This method is a direct replacement solution, enabling simple and easy adjustment of historical evaluation parameters, which can accurately assess the access characteristics of business data.

[0045] The second implementation method, such as Figure 2 As shown, when the target evaluation parameters are inconsistent with the historical evaluation parameters, the historical evaluation parameters are replaced with the target evaluation parameters (S204).

[0046] It is understandable that in practical applications, the target evaluation parameter selected from multiple preset evaluation parameters and associated with the common characteristics of the target may be consistent with or inconsistent with historical evaluation parameters. In the case of inconsistency, the historical evaluation parameter is replaced with the target evaluation parameter; in the case of consistency, no replacement is performed. In this way, precise adjustments can be achieved, avoiding unnecessary cost increases caused by adjustments that were not necessary.

[0047] exist Figure 2 For the implementation of S201 to S203, please refer to the relevant descriptions of S101 to S103, which will not be repeated here.

[0048] In this embodiment of the disclosure, both the historical evaluation parameter and the target evaluation parameter can be a single evaluation parameter from a plurality of preset evaluation parameters. At least one of the historical evaluation parameter and the target evaluation parameter can be a set of two or more evaluation parameters from a plurality of preset evaluation parameters. In this case, the historical evaluation parameter and the target evaluation parameter are considered inconsistent if there is only one different evaluation parameter, and they are considered consistent only if all evaluation parameters are the same.

[0049] The third implementation involves replacing the historical evaluation parameters with the target evaluation parameters when the target evaluation parameters are inconsistent with the historical evaluation parameters and the first reference value is lower than the second reference value. Here, the first reference value represents the estimated cost of adjusting the historical evaluation parameters based on the target evaluation parameters; the second reference value represents the estimated cost of partitioning the business data based on the historical evaluation parameters.

[0050] The costs included in the above implementation methods include at least one of resource processing costs and economic costs. Among them, resource processing costs include, but are not limited to, the resource processing burden caused by partitioning business data and the resource processing burden caused by adjusting historical evaluation parameters.

[0051] The data shards obtained by dividing business data based on historical evaluation parameters need to be stored in storage units. Specifically, storage units can be memory devices. It is understood that the more data shards there are, the more memory needs to be purchased or deployed in the search system, resulting in greater economic or resource deployment costs. Economic costs include, but are not limited to, the aforementioned economic and resource deployment costs.

[0052] If we consider the estimated cost of adjusting historical evaluation parameters based on target evaluation parameters as the first estimated cost, and the estimated cost of dividing business data based on historical evaluation parameters as the second estimated cost, then the third implementation method is equivalent to adjusting the evaluation parameters when the target evaluation parameters and historical evaluation parameters are inconsistent and the first estimated cost is lower than the second estimated cost, thus saving costs.

[0053] Furthermore, it allows for adjustments to historical evaluation parameters based on the target evaluation parameters, even when the first reference value is lower than the second reference value, without needing to consider whether the target evaluation parameters and historical evaluation parameters are consistent. Further, if the first reference value is lower than the second reference value, the historical evaluation parameters can be replaced with the target evaluation parameters. This effectively saves on the comparison process and costs associated with consistency checks.

[0054] The fourth implementation method involves replacing the historical evaluation parameters with the target evaluation parameters when the target evaluation parameters are inconsistent with the historical evaluation parameters and the difference between the third reference quantity and the fourth reference quantity meets a predetermined difference condition. Here, the third reference quantity represents the proportion of access data with the target common characteristics in the access dataset relative to all access data in the access dataset; the fourth reference quantity represents the proportion of business data with the target common characteristics in the business dataset relative to all business data in the business dataset.

[0055] In the fourth implementation, the business data in the access dataset refers to the data actually accessed by users, such as through searches or queries. From the user's perspective, the data searched or queried by the user is the accessed data. Compared to the unaccessed business data in the business dataset, the accessed business data is considered "hot" data. Access data in the access dataset that shares the target characteristics can be considered as business data with high access frequency or number of visits among the accessed business data. If the difference between the (first) proportion of such business data relative to all access data in the access dataset and the (second) proportion of business data with the target common characteristics relative to all business data in the business dataset reaches a predetermined difference condition, such as the difference between the first and second proportions reaching 20% ​​(both the first and second proportions are expressed as percentages), then the historical evaluation parameters originally used to measure the access characteristics of the business data are no longer able to accurately measure the actual hotness or coldness of the business data. The evaluation parameters used to measure the access characteristics of the business data need to be adjusted or updated to reflect the actual access situation of the business data by users.

[0056] It's understandable that the larger the first proportion and the smaller the second proportion, the greater the difference between the two proportions. A larger first proportion indicates that access data with the target common characteristics in the access dataset is accessed more frequently or multiple times in actual access scenarios. A smaller second proportion indicates that the amount of business data with the target common characteristics in the business dataset is less relative to all business data in the business dataset. The greater the difference between the two proportions, the greater the degree of fluctuation in the popularity of business data in the business dataset; business data that originally had a small proportion in the business dataset is frequently accessed in the access dataset. Thus, the historical evaluation parameters are less accurate in measuring the access characteristics of business data, requiring adjustment. In this case, if the historical evaluation parameters used to measure the access characteristics of business data are not adjusted, the actual popularity of business data cannot be accurately measured.

[0057] Furthermore, without considering whether the target evaluation parameters and historical evaluation parameters are consistent, historical evaluation parameters can be adjusted based on the target evaluation parameters as long as the difference between the third and fourth reference quantities meets predetermined difference conditions. Further, if the difference between the third and fourth reference quantities meets predetermined difference conditions, the historical evaluation parameters are replaced with the target evaluation parameters. This enables timely and automatic adjustment of evaluation parameters used to measure the access characteristics of business data.

[0058] The fifth implementation method involves replacing the historical evaluation parameters with the target evaluation parameters when the target evaluation parameters are inconsistent with the historical evaluation parameters, the first reference value is lower than the second reference value, and the difference between the third and fourth reference values ​​meets the predetermined difference conditions.

[0059] The fifth implementation method can be seen as a combination of the third and fourth implementation methods, which takes into account both cost savings and the actual access of users to business data, thereby enabling accurate adjustment of the evaluation parameters used to measure the access characteristics of business data.

[0060] Furthermore, regardless of whether the target evaluation parameters and historical evaluation parameters are consistent, and provided that the first reference value is lower than the second reference value and the difference between the third and fourth reference values ​​meets a predetermined difference condition, a scheme can be implemented by replacing the historical evaluation parameters with the target evaluation parameters to adjust the historical evaluation parameters based on the target evaluation parameters. This achieves timely and automatic adjustment of the evaluation parameters.

[0061] In the third to fifth implementation methods described above, in addition to considering whether the target evaluation parameters and historical evaluation parameters are consistent, it is also necessary to combine the above reference quantities to determine whether to adjust the historical evaluation parameters based on the target evaluation parameters. This ensures the accuracy of the evaluation parameter adjustment.

[0062] In this embodiment of the disclosure, the scheme for determining the common features among the target access data in the access dataset can be implemented by at least one of the following two implementation schemes:

[0063] Implementation scheme a: Determine the common features among the accessed data in the accessed dataset; obtain the number of accessed data with the same common features in the accessed dataset; based on this number, determine the target accessed data from the accessed dataset and determine the target common features from the common features.

[0064] Taking product data as an example, the common features of the access data include at least one of the product price value and product sales volume value. The number of access data sets with the same product price value and / or product sales volume value is counted, and the target access data and target common features are determined based on this number.

[0065] Determining the target access data and target common features in the access dataset based on the number of access data with the same common features in the access dataset can ensure the accuracy of the determination of the target access data and target common features, thereby enabling the accurate selection of target evaluation parameters.

[0066] In implementation scheme a, the scheme of determining the target access data from the access dataset and the target common features from the common features based on the quantity can be implemented through the following process: when the quantity meets the threshold condition, the access data corresponding to the quantity is taken as the target access data; from the common features, the common features of the target access data are selected as the target common features.

[0067] Taking the access data as product data and the common feature as the product price of 9.9 yuan as an example, if the number of access data with the same product price (9.9 yuan) in the statistically analyzed access dataset reaches a set first quantity threshold, and / or the ratio between this number and the total number of access data in the access dataset reaches a set first ratio threshold, the access data used to calculate this quantity will be used as the target access data. The common feature of this target access data is the target common feature. In this example, the access data in the access dataset with a product price of 9.9 yuan is the target access data. These target access data are all products with a price of 9.9 yuan, so the target common feature is the product price of 9.9 yuan. Among the preset evaluation parameters such as product price and product sales, the evaluation parameter associated with the product price of 9.9 yuan is the product price evaluation parameter, so the product price evaluation parameter can be used as the target evaluation parameter.

[0068] The threshold condition includes reaching at least one of a first quantity threshold and a first ratio threshold.

[0069] It's understandable that if the quantity meets the threshold condition, it means that there are a large number of target access data with the same product price value in the access dataset. The access dataset is a collection of business data that users search or query, and the target access data in the access dataset is the business data that is frequently accessed by users. Using the business data that is frequently accessed by users as the target access data, and using the common features of the target access data as the target common features, is adapted to the actual access situation of users, which can provide a good guarantee for accurately selecting target evaluation parameters.

[0070] Implementation scheme b: Obtain access parameters of the access data in the access dataset; select access data whose access parameters meet preset access conditions as target access data; determine the common features among the target access data; and use the determined common features as the target common features.

[0071] The access parameter is at least one of access frequency and access count. During user access to fragmented data, the access count of the accessed business data is recorded, and / or the access frequency of the accessed business data is calculated and recorded. When needed, the access parameters of the accessed data in the access dataset are obtained by reading the recorded information. Access data with a high access frequency (e.g., an access frequency greater than a preset access frequency threshold) and / or a high access count (e.g., an access count greater than a preset access count threshold) are selected as target access data. Common characteristics among the target access data can be used as target common characteristics.

[0072] For example, if the target access data consists entirely of business data where the sales volume of a product is greater than 100, then the common characteristic of the target is that the sales volume of a product is greater than 100. Among the preset evaluation parameters such as product price and product sales volume, the evaluation parameter associated with the sales volume of a product being greater than 100 is the product sales volume evaluation parameter, so the product sales volume evaluation parameter can be used as the target evaluation parameter.

[0073] It can be understood that the access data that meets the preset access conditions is the data that is accessed most frequently by users in the access dataset, and the target common feature is the feature that is common to the data that is accessed most frequently in the access dataset. Determining the target access data and the target common feature in the access dataset based on the access parameters of the access data can ensure the accuracy of the determination of the target access data and the target common feature, and thus enable the accurate selection of target evaluation parameters.

[0074] In this embodiment of the disclosure, such as Figure 3 As shown, the scheme for dividing the business data in the business dataset into shards based on historical evaluation parameters can be obtained in the following way:

[0075] S301: Group business data based on historical evaluation parameters to obtain grouped data;

[0076] S302: Based on the target attributes of the grouped data, divide the grouped data to obtain fragmented data.

[0077] Grouping the business data yields at least two data groups, each containing at least one business data item from the business dataset. To avoid data skew caused by grouping—that is, to prevent an imbalance in the amount of business data included in different groups (some groups having too much, some too little)—each data group needs to be further subdivided based on its target attributes, such as intra-group sharding. Each data group can be divided into two or more shards. Each shard is stored using a corresponding storage unit.

[0078] This approach of grouping first and then sharding avoids data skew and keeps the amount of data in each shard within a controllable range, making it easier to reasonably expand the amount of data within each shard within that controllable range.

[0079] The business data in the business dataset of this disclosure embodiment is any business data that is accessible to users, such as search or query data, like product data, store data, review data, medical data, etc.

[0080] In real-world search scenarios, business data in a search system may or may not be searched, resulting in "hot" or "cold" characteristics. The technical solution of this disclosure is an adaptive evaluation parameter determination scheme that can efficiently and automatically determine the evaluation parameters used to measure the access characteristics of business data and adjust the evaluation parameters in a timely manner, thereby reducing the cost of the machine on which the search system is located and optimizing the performance of the machine.

[0081] The following is combined with Figures 4 to 7 As shown, the technical solution of this disclosure will be further explained.

[0082] (a) Determine historical evaluation parameters.

[0083] Considering that the technical solution of this solution can be applied to the following three major search scenarios, the historical evaluation parameters are determined according to different search scenarios.

[0084] The embodiments disclosed herein can be applied to the following three types of search scenarios:

[0085] The first type of search scenario involves clearly defined hot and cold data categories for business data. Examples include store searches and site searches. In these scenarios, the accessed business data possesses attributes such as store ID and site ID, which distinguish between hot and cold data. For instance, popular store IDs and popular site IDs are considered "hot" data.

[0086] The second type of search scenario involves situations where the distinction between hot and cold data for business data is generally clear. In these scenarios, users' search needs are relatively well-defined. For example, in medical searches or product searches, the content that users have specific needs for and are searching for is considered hot data. For instance, in a product search scenario, searches for "jeans" or "high-waisted pants" are hot data. The hot and cold data for business data can be distinguished based on user search needs and data attributes such as the type or category of the products being searched.

[0087] For example, in medical searches, the search terms "cold" and "anti-inflammatory" are hot data. Business data can be differentiated into hot and cold data based on user search needs and data attributes such as disease type or symptoms.

[0088] The third type of search scenario involves situations where it is difficult or impossible to clearly distinguish between popular and unpopular business data. For example, in the search scenarios of learning apps, such as question-solving apps, it is impossible to know whether the business data is popular or unpopular.

[0089] It can be understood that the collection of business data for each type of search scenario is the business dataset. For example, in the product search scenario, the collection of different product data is the business dataset. In the store search scenario, the collection of different store IDs is the business dataset.

[0090] For both the first and second type of search scenarios, historical evaluation parameters can be determined based on the attributes of the business data. Taking the product search scenario as an example, the business dataset is a collection of different product data. Product data includes several product attributes such as product ID, product price, product sales volume, and product popularity. The preset evaluation parameters for the product search scenario include at least two of the following: product price, product sales volume, and product popularity.

[0091] The evaluation parameter determination scheme provided in this disclosure can be considered as a cyclical scheme. The historical evaluation parameters used when the scheme is first executed can be determined based on human experience. When the historical evaluation parameters are replaced with the target evaluation parameters according to the scheme shown in this disclosure, the target evaluation parameters from the previous execution are used as the historical evaluation parameters for the next execution of the scheme, and the target evaluation parameters selected in the next execution are used as the historical evaluation parameters for the next execution of the scheme, and so on.

[0092] Combination Figure 4 As shown, the initial execution of this solution is the prior process. In the prior process, based on the user's search requirements for business data in the search system, the first evaluation parameter can be determined according to the prior situation. For example, based on experience, the first evaluation parameter can be selected from multiple preset evaluation parameters. The data is then grouped according to the first evaluation parameter, and the grouped data is fragmented to obtain fragmented data. The number, type, and specifications of storage units used to store the fragmented data are also determined. Figure 4 Please refer to the following description for details of the proposed solution.

[0093] Taking the first implementation of this solution as an example, calculations using empirical formulas show that evaluating products based on price can effectively assess the popularity of product data. Product price is chosen as the first evaluation parameter; for ease of description, it is considered a historical evaluation parameter in this solution, illustrating that accurate assessment of product popularity can be achieved using price as the primary dimension.

[0094] (ii) Group the business data in the business dataset according to historical evaluation parameters.

[0095] Taking product data as an example, a hash partitioning algorithm is used to group the product data according to the product price, resulting in grouped data. For example, products priced between 1 and 10 yuan are assigned to the first group, products priced between 11 and 20 yuan are assigned to the second group, and so on, to obtain the grouped data. The data within each group is the grouped data.

[0096] The aforementioned grouping method is merely a specific example; any other reasonable grouping method falls within the scope of this disclosure.

[0097] As can be seen from the foregoing, for the third type of search scenario, when the solution shown in this disclosure is executed for the first time, since the historical evaluation parameters in this scenario cannot be clearly defined, the business data such as medical data in this scenario can be divided into the same group and wait for subsequent sharding processing.

[0098] (iii) Based on the target attributes of the grouped data, the grouped data is divided into fragmented data.

[0099] The target attribute of grouped data can be the primary key of the grouped data. In application scenarios where the business data is product data, the primary key of the product data is the product ID (identifier). Using a hash algorithm, the data within each group is divided according to the product ID, such as by sharding. For example, each group of data can be divided into two or more shards. The two or more shards within the same group can serve user access requests through sharding. A shard is a collection of services that manage the corresponding shard data. The corresponding business data served by each shard is the shard data.

[0100] Among them, the hash algorithm is a type of hash algorithm that uses hash operations to achieve sharding and the partitioning of sharded data.

[0101] For example, the first group includes 10 (groups) of data. According to the aforementioned partitioning scheme, the partitioning result is that the first to third business data in the 10 data are grouped together, the fourth to sixth data are grouped together, and the seventh to tenth data are grouped together.

[0102] The aforementioned fragments can be stored in storage units such as memory or hard disks. Figure 4 Determining the type of storage unit used to carry the fragments can be seen as choosing between memory (MEMS) and hard disk drives (SSDs). Determining the specifications can be seen as choosing the capacity of memory or hard disk drive that can accommodate the storage of the fragment data.

[0103] The choice between RAM and hard disk as the storage unit for the data slices depends on the economic or resource costs associated with purchasing or deploying RAM or hard disks in the search system. The capacity chosen depends on the amount of data sliced. Typically, the capacity of the RAM or hard disk must be equal to or greater than the capacity of the data slices stored on that RAM or hard disk.

[0104] In layman's terms, based on the aforementioned technical solution, and using historical evaluation parameters to obtain sharded data, such as... Figure 5 As shown, shards 1 to n are hosted on storage units. Shard 1 is hosted on an SSD, while shards 2 and n are hosted on MEM storage units. A shard can be viewed as a service or a collection of services, which can be used to serve users' access to sharded data.

[0105] Users can use the search system to access, search, or query fragmented data located on various storage units. When a user accesses the system, they input a search expression for the content they want to search for. The search system then invokes a request component, which in turn calls the service corresponding to the searched content. The invoked service then retrieves the searched content from the corresponding storage unit.

[0106] from Figure 5 As can be seen, to avoid data congestion caused by the same shard serving multiple access requests simultaneously, such as user search or query requests, each shard includes multiple services (service replicas). Different services or service replicas are used to serve different access requests.

[0107] It is understandable that the more user access requests a shard receives, the more frequently or numerous the shard data stored on the corresponding storage unit is accessed, and thus the more service replicas that shard have.

[0108] like Figure 5 As shown, shards 1 to n are assumed to be divided according to the historical evaluation parameter - product price, and each shard is stored in a corresponding storage unit. The shard data stored in each storage unit is used for access. During subsequent user access, it is found that users mostly access data for products with high sales volume, such as products with sales volume greater than 100. In this case, using the product price as the evaluation parameter to assess the popularity of product data is undoubtedly inaccurate, and the evaluation parameter needs to be adjusted. See the solution shown below for details.

[0109] (iv) Collect the business data accessed in the search system to obtain the access dataset.

[0110] It is understandable that in a search system, the business data that a user wants to search can be obtained through the search expression entered by the user, and the business data accessed by the user within a certain period of time can be recorded to obtain an access dataset.

[0111] The scheme described in (iv) above is a process for collecting and accessing data. See also Figure 6 As shown, online business refers to the aforementioned three types of search business, and the data collection process can be carried out offline. Furthermore, subsequent solution (V) can also be conducted offline, which can avoid data collection blockages and improve collection efficiency. This avoids conflicts with the online activation process and improves data mining efficiency.

[0112] (v) Explore evaluation parameters that can better measure the access characteristics of business data.

[0113] This solution can be achieved by identifying the common characteristics among the target access data in the access set and selecting evaluation parameters associated with these common characteristics from a set of preset evaluation parameters. Please refer to the aforementioned explanations for details.

[0114] For example, taking access data as product data, the historical evaluation parameter assesses the popularity of product data from the perspective of product price. As time goes on, most users may be more concerned with the sales volume of the product itself when searching for products, such as searching for products with high sales volume.

[0115] Taking the common characteristic of accessed data that the product sales volume is greater than 100 as an example,

[0116] If the number of access data entries with the same product sales value (all sales values ​​are greater than 100) in the statistically analyzed access dataset reaches a set second quantity threshold, and / or the ratio between this number and the total number of access data entries in the access dataset reaches a set second ratio threshold, then the access data used to calculate this quantity will be used as the target access data. The common characteristics of this target access data are called the target common characteristics.

[0117] In the aforementioned example, the access data in the access dataset where the product sales volume is greater than 100 is the target access data. Since all of these target access data are products with sales volumes greater than 100, the common target feature is that the product sales volume is greater than 100. Among the preset evaluation parameters such as product price and product sales volume, the evaluation parameter associated with the common target feature of product sales volume greater than 100 is the product sales volume evaluation parameter. Therefore, the product sales volume evaluation parameter can be used as the target evaluation parameter.

[0118] (vi) Perform the redistribution and rebalancing process of business data.

[0119] In simple terms, this disclosure replaces the historical evaluation parameters with the target evaluation parameters when the target evaluation parameters are inconsistent with the historical evaluation parameters, so as to achieve timely adjustment of the evaluation parameters used to measure the access characteristics of business data and thus achieve accurate evaluation of the access characteristics of business data.

[0120] For example, when the evaluation parameters are adjusted from historical evaluation parameters (such as product price) to target evaluation parameters (such as product sales volume), the product data in the business dataset is regrouped according to the target evaluation parameters to obtain target group data (new group data). Then, according to the new target attributes of the target group data, the target group data is re-divided to obtain multiple target fragment data (new fragment data). Figure 5As shown in slot1, slot2, and slot3, these represent m new partitions created from multiple target data partitions. Each new partition serves as a user's access point to the target data partitions.

[0121] A hash partitioning algorithm is used to regroup product data according to sales volume. For example, products with sales volume less than 100 are assigned to the first group, products with sales volume greater than 100 but less than 1000 are assigned to the second group, products with sales volume greater than 1000 but less than 2000 are assigned to the third group, and so on, to obtain each new group. The data located in each new group is the target group data.

[0122] In application scenarios where the business data is product data, the new target attribute can be the product ID. A hash algorithm is used to divide the data within each target group (the new group) based on the product ID, such as re-sharding. For example, instead of the original shards 1 to n, it becomes shards 1 to m, where n and m are both positive integers and are not equal.

[0123] In this case, the original fragments 1 to n are replaced by fragments 1 to m, and each fragment in fragments 1 to n serves the target fragment data after the re-division.

[0124] Whether the storage unit for each of the fragments 1 to n is a hard disk or a memory depends on the specific circumstances. The capacity of the hard disk or memory should be sufficient to store the corresponding target fragment data and support the corresponding fragments.

[0125] The aforementioned redistribution of business data includes the regrouping and resharding of business data. The redistribution process is performed online. Business data rebalancing includes, but is not limited to, adjustments to the number of shards, storage unit specifications and types, and the number of service replicas for each shard. This rebalancing process is performed offline. The rebalancing process may also include readjusting the search system to a reasonable state after shard adjustments.

[0126] Figure 7 This is a diagram illustrating the grouping of business data and the fragmentation within each group. Figure 7In this framework, the process of grouping business data and sharding within groups can be performed based on the current evaluation parameters (historical evaluation parameters or target evaluation parameters). Here, `field` represents all business data in the business dataset, and `F1(field)` represents grouping all business data according to the current evaluation parameters, resulting in `n` groups. The business data within each group is called the group data. `key` represents the primary key of the business data, such as the ID of product data. `slot = F2(group, key)` means sharding the data within each group according to the primary key, resulting in shard slots within each group and one or more shard data within each group, such as the four shards (slot1 to slot4) in group 1 and the four shards (slot1 to slot4) in group 2.

[0127] Figure 7 Data skew constraint in this context refers to the need to shard the data within each group after grouping the business data. This is primarily because grouping alone can lead to an uneven distribution of data across groups, with some groups having too much data and others too little, resulting in data skew. Sharding the data within each group ensures that the size of the sharded business data remains within a controllable range. In online applications, users retrieve sharded data, not grouped data, resulting in consistent search times, a better user experience, and easier, manageable expansion of the data volume within each shard.

[0128] The aforementioned scheme involves replacing historical evaluation parameters with target evaluation parameters when the target evaluation parameters differ from historical evaluation parameters. As another alternative scheme, when the target evaluation parameters differ from historical evaluation parameters, see [link to relevant documentation]. Figure 6 As shown, the costs incurred in replacing historical evaluation parameters with target evaluation parameters (online costs) and the costs incurred in partitioning business data based on historical evaluation parameters also need to be considered. That is, the evaluation parameters should be adjusted only if the first estimated cost is lower than the second estimated cost, in order to save costs. After adjusting the evaluation parameters, the business data should be redistributed and rebalanced to achieve accurate evaluation of the access characteristics of the business data.

[0129] For other alternatives to the above, please refer to the relevant explanations above; repeated details will not be elaborated upon.

[0130] It is understood that the disclosed scheme is a cyclical or iterative scheme. The aforementioned target evaluation parameters can be used as historical evaluation parameters for the next execution of this scheme, and the target evaluation parameters selected in the next execution can be used as historical evaluation parameters for the next execution of this scheme, and so on in a cyclical or iterative manner. Each cycle or iteration can be carried out at least according to the schemes shown in (iv)-(vi) above.

[0131] In layman's terms, the scheme for determining the evaluation parameters disclosed herein can be viewed as an adaptive mining scheme for evaluation parameters used to measure the hot and cold characteristics of business data. This scheme has at least the following advantages:

[0132] First, by grouping and fragmenting business data, and collecting access data, assessment parameters for accurately measuring the current access frequency of business data can be adaptively identified or mined. These assessment parameters can be adaptively adjusted as the actual access patterns of business data change, thus achieving an accurate assessment of the access characteristics of the business data.

[0133] Second, grouping and then sharding business data effectively solves the data skew problem. After grouping the business data, the primary key of the grouped data is sharded to obtain each shard and each shard's data. This ensures that the size skew of the shard data served by each shard is within a certain range, i.e., within a controllable range, and has no impact on the horizontal expansion of the scale of the stored shard data.

[0134] III. Cost Optimization. During online application (for users to query or search sharded data), evaluation parameters that measure the hotness or coldness of data are continuously mined. Based on cost comparison results, the most cost-optimized approach is selected to determine whether to adjust the evaluation parameters. This mining process is transparent to the business data access process, neither affecting the mining of evaluation parameters nor the normal access to business data.

[0135] This disclosed adaptive mining scheme for evaluation parameters used to measure the hot / cold characteristics of business data solves the problem in related technologies where evaluation parameters cannot be automatically adjusted or updated. It also addresses the issue of evaluation parameters not being updated or adjusted in a timely manner when the hot / cold nature of business data changes. Furthermore, it solves the problem of data scalability caused by data skew by grouping before sharding.

[0136] This disclosed adaptive mining scheme for evaluation parameters used to measure the hot / cold characteristics of business data employs a grouping and intra-group sharding approach. A hot data feedback mechanism is introduced based on the collected access dataset. This feedback mechanism enables accurate mining of evaluation parameters that accurately measure the hot / cold characteristics of business data. The data in the access dataset consists of data requested by users, equivalent to other business data in the business dataset, and can be considered hot data. Mining evaluation parameters based on hot data ensures the accuracy of the mining, thus allowing for the use of accurate evaluation parameters to assess the access characteristics of business data.

[0137] This disclosure also provides an embodiment of an apparatus for determining evaluation parameters, such as... Figure 8 As shown, it includes:

[0138] The acquisition unit 801 is used to acquire the access dataset obtained by accessing the sharded data; wherein the sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, and the historical evaluation parameters are used to measure the access characteristics of the business data in the business dataset.

[0139] The determining unit 802 is used to determine the common features of the target access data among the target access data in the access dataset;

[0140] Selection unit 803 is used to select, from a plurality of preset evaluation parameters, an evaluation parameter associated with the common features of the target as the target evaluation parameter;

[0141] The adjustment unit 804 is used to adjust the historical evaluation parameter based on the target evaluation parameter.

[0142] The adjustment unit 804 is also used for

[0143] If the first reference value is lower than the second reference value, the historical evaluation parameter is adjusted based on the target evaluation parameter;

[0144] The first reference quantity represents the estimated cost of adjusting the historical evaluation parameter based on the target evaluation parameter; the second reference quantity represents the estimated cost of dividing the business data based on the historical evaluation parameter.

[0145] The adjustment unit 804 is also used for

[0146] If the difference between the third and fourth reference values ​​meets the predetermined difference condition, the historical evaluation parameter is adjusted based on the target evaluation parameter.

[0147] The third reference quantity is characterized as the proportion of access data with the common characteristics of the target in the access dataset relative to all access data in the access dataset; the fourth reference quantity is characterized as the proportion of business data with the common characteristics of the target in the business dataset relative to all business data in the business dataset.

[0148] The adjustment unit 804 is also used for

[0149] If the target evaluation parameter is inconsistent with the historical evaluation parameter, the historical evaluation parameter shall be replaced with the target evaluation parameter.

[0150] The determining unit 802 is also used for

[0151] Identify the common characteristics among the accessed data in this accessed dataset;

[0152] Obtain the number of access data points in the access dataset that share the same common characteristics;

[0153] Based on this quantity, the target access data is determined from the access dataset, and the target common features are determined from the common features.

[0154] The determining unit 802 is also used for

[0155] If the quantity meets the threshold condition, the access data corresponding to that quantity shall be used as the target access data;

[0156] From these shared features, select the shared features of the target access data as the target shared features.

[0157] The determining unit 802 is also used for

[0158] Obtain the access parameters for accessing data in the access dataset;

[0159] Select the access data whose access parameters meet the preset access conditions as the target access data;

[0160] Identify the common characteristics among the target accessed data;

[0161] The common feature among the target access data is taken as the target common feature.

[0162] The device also includes a partitioning unit ( Figure 8 (Not shown in the image) is used for

[0163] Based on this historical evaluation parameter, the business data is grouped to obtain grouped data;

[0164] Based on the target attributes of the grouped data, the grouped data is divided to obtain the fragmented data.

[0165] It should be noted that the device for determining evaluation parameters in this embodiment of the present disclosure is similar in principle to the aforementioned method for determining evaluation parameters. Therefore, the implementation process and principle of determining evaluation parameters can be found in the description of the implementation process and principle of the aforementioned method, and repeated details will not be repeated.

[0166] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0167] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0168] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the feature filtering method of this disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the feature filtering method of this disclosure described and / or claimed herein.

[0169] like Figure 9 As shown, device 900 includes a computing unit 901, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 902 or a computer program loaded from storage unit 908 into random access memory (RAM) 903. RAM 903 may also store various programs and data required for the operation of device 900. The computing unit 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / output (I / O) interface 905 is also connected to bus 904.

[0170] Multiple components in device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, etc.; output unit 907, such as various types of monitors, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0171] The computing unit 901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as one or more steps of a method for determining evaluation parameters. For example, in some embodiments, the method for determining evaluation parameters may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program may be loaded and / or installed on device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by the computing unit 901, one or more steps of the method for determining evaluation parameters described above may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the method for determining evaluation parameters by any other suitable means (e.g., by means of firmware).

[0172] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0173] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0174] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0175] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0176] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0177] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0178] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0179] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining evaluation parameters, comprising: Obtain the access dataset obtained by accessing the sharded data; The sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, which are used to measure the access characteristics of the business data in the business dataset; wherein, the access dataset is a collection of all business data that has been searched or queried within a certain period of time. Determine the common characteristics of the target access data among the target access data in the access dataset; From a set of preset evaluation parameters, select the evaluation parameter that is associated with the common features of the target as the target evaluation parameter; The historical evaluation parameters are adjusted based on the target evaluation parameters; The adjustment of the historical evaluation parameters based on the target evaluation parameters includes: If the first reference value is lower than the second reference value, the historical evaluation parameters are adjusted based on the target evaluation parameters; Wherein, the first reference quantity represents the estimated cost of adjusting the historical evaluation parameters based on the target evaluation parameters; the second reference quantity represents the estimated cost of segmenting the business data based on the historical evaluation parameters; If the target evaluation parameter is inconsistent with the historical evaluation parameter, the historical evaluation parameter is replaced with the target evaluation parameter; wherein, the historical evaluation parameter and the target evaluation parameter are access characteristics of business data in the business dataset that are measured from different dimensions.

2. The method according to claim 1, wherein, The adjustment of the historical evaluation parameters based on the target evaluation parameters further includes: If the difference between the third and fourth reference values ​​meets a predetermined difference condition, the historical evaluation parameters are adjusted based on the target evaluation parameters. The third reference quantity is characterized as the proportion of access data with the target common feature in the access dataset relative to all access data in the access dataset; the fourth reference quantity is characterized as the proportion of business data with the target common feature in the business dataset relative to all business data in the business dataset.

3. The method according to claim 1 or 2, wherein, Determining the common features among target access data in the access dataset includes: Determine the common characteristics among the accessed data in the accessed dataset; Obtain the number of access data points in the access dataset that share the same common characteristics; Based on the quantity, the target access data is determined from the access dataset, and the target common features are determined from the common features.

4. The method according to claim 3, wherein, The step of determining the target access data from the access dataset and determining the target common features from the common features based on the quantity includes: If the quantity meets the threshold condition, the access data corresponding to the quantity is taken as the target access data; From the common features, select the common features of the target access data as the target common features.

5. The method according to claim 1 or 2, wherein, Determining the common features among target access data in the access dataset includes: Obtain the access parameters for accessing data in the access dataset; Select the access data whose access parameters meet the preset access conditions as the target access data; Identify the common characteristics among the target accessed data; The common features among the target access data are taken as the target common features.

6. The method according to claim 1 or 2, further comprising: The business data is grouped based on the historical evaluation parameters to obtain grouped data. Based on the target attributes of the grouped data, the grouped data is divided to obtain the fragmented data.

7. An apparatus for determining evaluation parameters, comprising: The acquisition unit is used to acquire the access dataset obtained by accessing the fragmented data; The sharded data is obtained by dividing the business data in the business dataset based on historical evaluation parameters, which are used to measure the access characteristics of the business data in the business dataset; wherein, the access dataset is a collection of all business data that has been searched or queried within a certain period of time. The determining unit is used to determine the common features of the target access data among the target access data in the access dataset; The selection unit is used to select, from a plurality of preset evaluation parameters, an evaluation parameter associated with the common features of the target as the target evaluation parameter; An adjustment unit is used to adjust the historical evaluation parameters based on the target evaluation parameters; The adjustment unit is also used for If the first reference value is lower than the second reference value, the historical evaluation parameters are adjusted based on the target evaluation parameters; Wherein, the first reference quantity represents the estimated cost of adjusting the historical evaluation parameters based on the target evaluation parameters; the second reference quantity represents the estimated cost of dividing the business data based on the historical evaluation parameters; when the target evaluation parameters and the historical evaluation parameters are inconsistent, the historical evaluation parameters are replaced with the target evaluation parameters; wherein, the historical evaluation parameters and the target evaluation parameters are access characteristics of the business data in the business dataset measured from different dimensions.

8. The apparatus according to claim 7, wherein, The adjustment unit is also used for If the difference between the third and fourth reference values ​​meets a predetermined difference condition, the historical evaluation parameters are adjusted based on the target evaluation parameters. The third reference quantity is characterized as the proportion of access data with the target common feature in the access dataset relative to all access data in the access dataset; the fourth reference quantity is characterized as the proportion of business data with the target common feature in the business dataset relative to all business data in the business dataset.

9. The apparatus according to claim 7 or 8, wherein, The determining unit is also used for Determine the common characteristics among the accessed data in the accessed dataset; Obtain the number of access data points in the access dataset that share the same common characteristics; Based on the quantity, the target access data is determined from the access dataset, and the target common features are determined from the common features.

10. The apparatus according to claim 9, wherein, The determining unit is also used for If the quantity meets the threshold condition, the access data corresponding to the quantity is taken as the target access data; From the common features, select the common features of the target access data as the target common features.

11. The apparatus according to claim 7 or 8, wherein, The determining unit is also used for Obtain the access parameters for accessing data in the access dataset; Select the access data whose access parameters meet the preset access conditions as the target access data; Identify the common characteristics among the target accessed data; The common features among the target access data are taken as the target common features.

12. The apparatus according to claim 7 or 8, further comprising a dividing unit for... The business data is grouped based on the historical evaluation parameters to obtain grouped data. Based on the target attributes of the grouped data, the grouped data is divided to obtain the fragmented data.

13. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

15. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images