Circuit granularity splitting method, system and equipment, medium and program product
Through the decision tree of line particle size and dynamic threshold control, the order head aggregation problem caused by line particle size solidification is solved, and the dynamic splitting and recommendation of line particle size is realized, which improves user experience and saves operating costs.
Patent Information
- Application Number
- CN202510616990.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-14
- Publication Date
- 2025-08-15
AI Technical Summary
In the prior art, the display of line particles is solidified through data analysis and business operation experience, resulting in the phenomenon of "head aggregation" in orders, and the separation and dynamic display of line particles cannot be achieved.
The decision tree with line granularity is adopted, and the index scores of the parent node and child nodes are obtained, the target level of line granularity is obtained based on the comparison results, and the decision tree is constructed based on the user search terms and multi-dimensional splitting criteria are designed. Dynamic threshold control and memorized search are used to achieve dynamic splitting of line granularity.
The dynamic splitting of line granularity is realized, avoiding the head effects of traffic and orders, alleviating the homogeneity of line recall sets, and saving manual operation costs.
Smart Images
Figure CN120492732A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of line information processing, and in particular to a line granularity splitting method, system, device, medium and program product. Background Art
[0002] Search systems are widely used in major apps, and the quality of search results directly impacts user experience and the revenue of suppliers and platforms. In the travel search scenario, routes are one of the core elements of basic product information. From a geographic information level, routes have multiple levels of granularity, such as country → province → region → prefecture-level city → county-level city → POI. Exhausting all granularities will lead to redundant product display information, poor user experience, and increased business operation pressure. Usually, the granularity of routes is solidified through data analysis and business operation experience for display. However, this solidified granularity cannot accurately capture the needs of different users, and will lead to "head-clustering" of orders. It is also impossible to split the granularity of routes and dynamically display the granularity levels. Summary of the Invention
[0003] The technical problem to be solved by the present disclosure is to overcome the defects in the prior art of displaying line granularity by solidifying it through data analysis and business operation experience, which leads to the phenomenon of "head aggregation" of orders, and the inability to split the line granularity and dynamically display the level of line granularity splitting. A method, system, equipment, medium and program product for splitting the line granularity are provided.
[0004] The present disclosure solves the above technical problems through the following technical solutions:
[0005] A first aspect of the present disclosure provides a method for splitting a line granularity, the method comprising:
[0006] Get the decision tree of line granularity;
[0007] Obtaining a parent node of the line granularity from the decision tree;
[0008] Splitting out the child nodes corresponding to the parent node;
[0009] Get the index score of the parent node and the index score of the child node;
[0010] A target level of line granularity is obtained based on a comparison result of the index score of the parent node and the index score of the child node.
[0011] Preferably, the step of obtaining a decision tree of line granularity includes:
[0012] Get the user's search term;
[0013] Obtaining a line recall set related to the search term;
[0014] A decision tree of line granularity is constructed based on the line recall set.
[0015] Preferably, the step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node includes:
[0016] In response to the indicator score of the child node being less than the indicator score of the parent node, and determining that the child node is a leaf node, the leaf node is used as a target level of the line granularity.
[0017] Preferably, the step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node further comprises:
[0018] In response to the indicator score of the child node being less than the indicator score of the parent node, and determining that the child node is not a leaf node, the child node is split, and the child node is used as the parent node to loop through the steps of obtaining the indicator score of the parent node and the indicator score of the child node; and obtaining the target level of the line granularity based on the comparison result of the indicator score of the parent node and the indicator score of the child node.
[0019] Preferably, the step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node further comprises:
[0020] In response to the indicator score of the child node being not less than the indicator score of the parent node, the parent node is used as a target level of the line granularity.
[0021] Preferably, the index score includes a clustering score and a similarity score.
[0022] A second aspect of the present disclosure provides a line granularity splitting system, the splitting system comprising:
[0023] A first acquisition module is used to obtain a decision tree of line granularity;
[0024] A second acquisition module is used to obtain the parent node of the line granularity from the decision tree;
[0025] A splitting module, used to split out child nodes corresponding to the parent node;
[0026] The third acquisition module is used to obtain the index score of the parent node and the index score of the child node;
[0027] The fourth acquisition module is configured to acquire a target level of line granularity based on a comparison result of the indicator score of the parent node and the indicator score of the child node.
[0028] Preferably, the first acquisition module includes:
[0029] A first acquisition unit is used to acquire a user's search term;
[0030] a second acquiring unit, configured to acquire a line recall set related to the search term;
[0031] A construction unit is used to construct a decision tree of line granularity based on the line recall set.
[0032] Preferably, the fourth acquisition module is configured to, in response to the indicator score of the child node being smaller than the indicator score of the parent node and determining that the child node is a leaf node, use the leaf node as the target level of the line granularity.
[0033] Preferably, the fourth acquisition module is also used to split the child node in response to the indicator score of the child node being less than the indicator score of the parent node, and judging that the child node is not a leaf node, and cyclically call the third acquisition module and the fourth acquisition module with the child node as the parent node.
[0034] Preferably, the fourth acquisition module is further configured to, in response to the indicator score of the child node being not less than the indicator score of the parent node, use the parent node as the target level of the line granularity.
[0035] Preferably, the index score includes a clustering score and a similarity score.
[0036] A third aspect of the present disclosure provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and used to run on the processor, wherein when the processor executes the computer program, the line granularity splitting method described in the first aspect is implemented.
[0037] A fourth aspect of the present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for splitting the line granularity described in the first aspect is implemented.
[0038] A fifth aspect of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the line granularity splitting method as described in the first aspect.
[0039] On the basis of conforming to the common sense in this field, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present disclosure.
[0040] The positive progress of this disclosure is:
[0041] The present disclosure realizes the splitting of line granularity by combining the indicator scores of parent nodes and child nodes based on the decision tree of line granularity. It can avoid the head effect of traffic and orders while recommending the target level of line granularity, alleviate the homogeneity of line recall sets, provide a reference basis for line granularity splitting, and save manual operation costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 This is a flowchart of the line granularity splitting method provided in Example 1 of the present disclosure.
[0043] Figure 2 Schematic diagram of the hierarchical structure of the decision tree provided for Examples 1 and 2 of the present disclosure.
[0044] Figure 3 Schematic diagram of the decision tree structure provided for Examples 1 and 2 of the present disclosure.
[0045] Figure 4 A schematic diagram of the modules of the line granularity splitting system provided in Example 2 of the present disclosure.
[0046] Figure 5 This is a structural diagram of an electronic device for implementing the circuit granularity splitting method according to embodiment 3 of the present disclosure. DETAILED DESCRIPTION
[0047] The present disclosure is further illustrated below by way of examples, but the present disclosure is not limited to the scope of the examples.
[0048] In the embodiments of the present disclosure, prefixes such as "first" and "second" are used only to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. In the embodiments of the present disclosure, the use of prefixes such as ordinal numbers to distinguish description objects does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitations should be constituted due to the use of such prefixes. In addition, in the description of this embodiment, unless otherwise specified, the meaning of "plurality" is two or more.
[0049] In the embodiments of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0050] Example 1
[0051] Figure 1 This is a flow chart of a method for splitting line granularity provided in Example 1 of the present disclosure, such as Figure 1 As shown, the splitting methods include:
[0052] S1. Obtaining a decision tree for line granularity;
[0053] In this embodiment, a decision tree is a tree-structured supervised learning method that constructs classification / regression rules by splitting features. Common splitting criteria include using information gain (ID3), gain ratio (C4.5), or the Gini coefficient (CART) to assess the quality of feature splits. The decision tree growth strategy recursively selects the optimal features until a stopping condition is met. This structure is designed to mimic the human decision-making process, making it intuitive and interpretable, and capable of handling both numerical and categorical data. They are widely used in various fields, including healthcare, finance, and marketing, for tasks such as diagnosis, risk assessment, and customer segmentation.
[0054] In this embodiment, Figure 2 is the hierarchical structure of the decision tree, Figure 3 This is a schematic diagram of the decision tree structure. In addition, the levels of the decision tree can be added and deleted according to the current status of the line granularity level that needs to be determined. There is a hierarchical geographical relationship between the levels.
[0055] S2. Obtain the parent node of the line granularity from the decision tree;
[0056] S3, split out the child nodes corresponding to the parent node;
[0057] In this embodiment, the parent node is the node before splitting, and the child node is the node obtained after splitting the parent node.
[0058] S4. Obtain the index score of the parent node and the index score of the child node;
[0059] In an optional embodiment, the index score includes a clustering score and a similarity score.
[0060] In this embodiment, the aggregation score is used to measure the head effect of the line. A large aggregation score indicates that the line head effect is obvious and needs to be split into finer granularity. The similarity score is used to characterize the similarity between lines. A large similarity score indicates serious homogeneity and needs to be aggregated to a coarser granularity.
[0061] Specifically, the aggregation score is characterized based on the entropy of the line-level order ratio. The expression of the aggregation score is shown in formula (1):
[0062] (1)
[0063] in, represents the aggregation score; The entropy representing the proportion of orders at the line level;
[0064] Pord Indicates the order distribution of the route; i indicates the i-th route; N indicates that there are N routes in total;
[0065] Furthermore, The expression of is shown in formula (2):
[0066]
[0067] in, represents the order distribution of the i-th route.
[0068] The similarity score is based on the Jaccard similarity weighted by the line heat. The expression of the similarity score is shown in formula (3):
[0069]
[0070] in, represents the similarity score; j represents the jth route; poi represents the scenic spot in the route; represents the scenic spots in the jth route, represents the scenic spot in the i-th route; Indicates the popularity of attractions in the route.
[0071] Furthermore, the aggregation score and the similarity score are fused to obtain the fusion score. The expression of the fusion score is shown in formula (4):
[0072] (4)
[0073] in, represents the fusion score; Indicates the aggregation score of a certain granularity; Indicates the similarity score of a certain granularity; Indicates a certain granularity; and are constants. In addition, and Used to balance the clustering score and similarity score.
[0074] Furthermore, the expression of node splitting decision is shown in formula (5):
[0075] (5)
[0076] in, Indicates the split decision of the current node; Indicates the current node;
[0077] It should be noted that when calculating the metric score S1 of the current node (current geographical level) without splitting into child nodes (next geographical level), and then calculating the status metric score S2 when the current node splits into child nodes, if S2 < S1, the splitting criterion is met, and the current node can split into child nodes.
[0078] S5. Obtain the target level of the line granularity based on the comparison result of the metric scores of the parent node and the child nodes.
[0079] In this embodiment, in different search scenarios, the line name needs to be split into different granularities. The invention combines user cognition, the actual situation of the supply chain, and the performance of platform orders and traffic to dynamically recommend the granularity of line splitting required in different scenarios.
[0080] This embodiment realizes the splitting of the line granularity by combining the decision tree of the line granularity with the metric scores of the parent node and the child nodes. It can avoid the head effect of traffic and orders while recommending the target level of the line granularity, alleviate the homogenization of the line recall set, provide a reference basis for line granularity splitting, and save the labor operation cost.
[0081] In an optional embodiment, S1 includes:
[0082] Obtain the user's search term;
[0083] Obtain the line recall set related to the search term;
[0084] Construct a decision tree of line granularity based on the line recall set.
[0085] In this embodiment, a decision tree with geographical levels is constructed based on the line recall set related to the search term, then a multi-dimensional line granularity splitting criterion is designed, and finally the global optimal level (e.g., the target level) is controlled by a dynamic threshold and memoization search.
[0086] In an optional embodiment, S5 includes:
[0087] In response to the metric score of the child node being less than that of the parent node and determining that the child node is a leaf node, take the leaf node as the target level of the line granularity.
[0088] In an optional embodiment, S5 further includes:
[0089] In response to the metric score of the child node being less than that of the parent node and determining that the child node is not a leaf node, split the child node and take the child node as the parent node to loop and execute steps S4 - S5.
[0090] In this embodiment, if the index score of the child node is less than the index score of the parent node, the splitting criterion is met, and it is further determined whether the child node is a leaf node. If it is a leaf node, the leaf node is used as the target level of the line granularity (that is, as the optimal target level of the line granularity); if the child node is not a leaf node, the child node is split, and the child node is used as the parent node to loop through steps S4-S5.
[0091] It should be noted that the leaf node is the bottom-level node, that is, the node that cannot be split any further; the target level is the optimal level.
[0092] In an optional embodiment, S5 further includes:
[0093] In response to the indicator score of the child node being not less than the indicator score of the parent node, the parent node is used as a target level of the line granularity.
[0094] This embodiment combines the algorithm characteristics of the decision tree and the requirements for route granularity splitting to construct a decision tree based on the geographical hierarchy. By designing two quantitative indicators, the aggregation score and the similarity score, as the node splitting criteria, dynamic threshold control and memorization search are used to achieve the global optimality. The aggregation score is used to measure the head effect of the route. A large aggregation score indicates that the route head effect is obvious and needs to be split into finer granularity; the similarity score is used to characterize the similarity between routes. A large similarity score indicates serious homogeneity and needs to be aggregated to a coarser granularity. Finally, the aggregation score and the similarity score are balanced by α and β. Specifically, the route granularity splitting level is recommended by different search terms, and the route pass rate reaches 88%. It is applied to the route level judgment of the group tour list page, alleviating the problem of homogeneity of the recalled routes; it provides a reference basis for the splitting of route granularity and saves manual operating costs.
[0095] Example 2
[0096] Corresponding to the aforementioned embodiment of a method for splitting a circuit at a granularity, the present disclosure also provides an embodiment of a system for splitting a circuit at a granularity.
[0097] Figure 4 A schematic diagram of a module of a line granularity splitting system provided in Example 2 of the present disclosure, such as Figure 4 As shown, the split system includes:
[0098] A first acquisition module 21 is used to obtain a decision tree of line granularity;
[0099] In this embodiment, a decision tree is a tree-structured supervised learning method that constructs classification / regression rules by splitting features. Common splitting criteria include using information gain (ID3), gain ratio (C4.5), or the Gini coefficient (CART) to assess the quality of feature splits. The decision tree growth strategy recursively selects the optimal features until a stopping condition is met. This structure is designed to mimic the human decision-making process, making it intuitive and interpretable, and capable of handling both numerical and categorical data. They are widely used in various fields, including healthcare, finance, and marketing, for tasks such as diagnosis, risk assessment, and customer segmentation.
[0100] In this embodiment, Figure 2 is the hierarchical structure of the decision tree, Figure 3 This is a schematic diagram of the decision tree structure. In addition, the levels of the decision tree can be added and deleted according to the current status of the line granularity level that needs to be determined. There is a hierarchical geographical relationship between the levels.
[0101] A second acquisition module 22 is used to obtain the parent node of the line granularity from the decision tree;
[0102] Splitting module 23, used to split out child nodes corresponding to parent nodes;
[0103] In this embodiment, the parent node is the node before splitting, and the child node is the node obtained after splitting the parent node.
[0104] The third acquisition module 24 is used to obtain the index score of the parent node and the index score of the child node;
[0105] In an optional embodiment, the index score includes a clustering score and a similarity score.
[0106] In this embodiment, the aggregation score is used to measure the head effect of the line. A large aggregation score indicates that the line head effect is obvious and needs to be split into finer granularity. The similarity score is used to characterize the similarity between lines. A large similarity score indicates serious homogeneity and needs to be aggregated to a coarser granularity.
[0107] Specifically, the concentration score is characterized based on the entropy of the line-level order ratio. The expression of the concentration score is shown in Formula (1) in Example 1. Further, Its expression is as shown in formula (2) in Embodiment 1; the similarity score is the Jaccard similarity weighted by line popularity, and the expression of the similarity score is as shown in formula (3) in Embodiment 1; further, after fusing the aggregation score and the similarity score, a fusion score is obtained, and the expression of the fusion score is as shown in formula (4) in Embodiment 1; further, the expression of the node splitting decision is as shown in formula (5) in Embodiment 1; it should be noted that when calculating the index score when the current node (current geographical level) is not split into child nodes (next geographical level), it is S1, and then calculate the state index score S2 when the current node splits out child nodes. If S2 < S1, that is, the splitting criterion is met, and the current node can split out child nodes.
[0108] The fourth acquisition module 25 is configured to obtain the target level of the line granularity based on the comparison result between the index score of the parent node and the index score of the child node.
[0109] In this embodiment, in different search scenarios, the line name needs to be split into different granularities. The invention combines user cognition, the actual situation of the supply chain, and the performance of platform orders and traffic to dynamically recommend the granularity that the line needs to be split into in different scenarios.
[0110] This embodiment realizes the splitting of the line granularity based on the decision tree of the line granularity in combination with the index scores of the parent node and the child node, can avoid the head effect of traffic and orders while recommending the target level of the line granularity, alleviate the homogenization of the line recall set, provide a reference basis for the splitting of the line granularity, and save the labor operation cost.
[0111] In an optional embodiment, the first acquisition module includes:
[0112] The first acquisition unit is configured to acquire the search term of the user;
[0113] The second acquisition unit is configured to acquire the line recall set related to the search term;
[0114] The construction unit is configured to construct a decision tree of the line granularity based on the line recall set.
[0115] In this embodiment, a decision tree with geographical levels is constructed based on the line recall set related to the search term, then a multi-dimensional line granularity splitting criterion is designed, and finally the global optimal level (for example, the target level) is controlled by a dynamic threshold and memory search.
[0116] In an optional embodiment, the fourth acquisition module is configured to, in response to the index score of the child node being less than the index score of the parent node and determining that the child node is a leaf node, use the leaf node as the target level of the line granularity.
[0117] In an optional embodiment, the fourth acquisition module is also used to split the child node in response to the indicator score of the child node being less than the indicator score of the parent node and determining that the child node is not a leaf node, and cyclically call the third acquisition module and the fourth acquisition module with the child node as the parent node.
[0118] In this embodiment, if the index score of the child node is less than the index score of the parent node, the splitting criterion is met, and it is further determined whether the child node is a leaf node. If it is a leaf node, the leaf node is used as the target level of the line granularity (that is, as the optimal target level of the line granularity); if the child node is not a leaf node, the child node is split, and the child node is used as the parent node to cyclically call the third acquisition module and the fourth acquisition module.
[0119] It should be noted that the leaf node is the bottom-level node, that is, the node that cannot be split any further; the target level is the optimal level.
[0120] In an optional embodiment, the fourth acquisition module is further configured to use the parent node as the target level of the line granularity in response to the indicator score of the child node being not less than the indicator score of the parent node.
[0121] This embodiment combines the algorithm characteristics of the decision tree and the requirements for route granularity splitting to construct a decision tree based on the geographical hierarchy. By designing two quantitative indicators, the aggregation score and the similarity score, as the node splitting criteria, dynamic threshold control and memorization search are used to achieve the global optimality. The aggregation score is used to measure the head effect of the route. A large aggregation score indicates that the route head effect is obvious and needs to be split into finer granularity; the similarity score is used to characterize the similarity between routes. A large similarity score indicates serious homogeneity and needs to be aggregated to a coarser granularity. Finally, the aggregation score and the similarity score are balanced by α and β. Specifically, the route granularity splitting level is recommended by different search terms, and the route pass rate reaches 88%. It is applied to the route level judgment of the group tour list page, alleviating the problem of homogeneity of the recalled routes; it provides a reference basis for the splitting of route granularity and saves manual operating costs.
[0122] Since the system embodiments generally correspond to the method embodiments, reference will be made to the description of the method embodiments for relevant details. The system embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components of the units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the disclosed solution.
[0123] Example 3
[0124] Figure 5This is a structural diagram of an electronic device shown in Example 3 of the present disclosure. The electronic device includes a memory, a processor, and a computer program stored in the memory and used to run on the processor. When the processor executes the computer program, it implements the line granularity splitting method described in any of the above embodiments. Figure 5 The electronic device 90 shown is only an example and should not limit the functionality and scope of use of the embodiments of the present disclosure.
[0125] like Figure 5 As shown, the electronic device 90 may be a general-purpose computing device, such as a server device. Components of the electronic device 90 may include, but are not limited to, the at least one processor 91, the at least one memory 92, and a bus 93 connecting different system components (including the memory 92 and the processor 91).
[0126] The bus 93 includes a data bus, an address bus, and a control bus.
[0127] The memory 92 may include a volatile memory, such as a random access memory (RAM) 921 and / or a cache memory 922 , and may further include a read-only memory (ROM) 923 .
[0128] The memory 92 may also include a program tool 925 (or utility) having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.
[0129] The processor 91 executes various functional applications and data processing by running the computer program stored in the memory 92, such as the line granularity splitting method provided in any of the above embodiments.
[0130] The electronic device 90 can also communicate with one or more external devices 94 (e.g., keyboards, pointing devices, etc.). Such communication can be performed through an input / output (I / O) interface 95. In addition, the electronic device 90 can also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 96. Figure 5 As shown, the network adapter 96 communicates with other modules of the electronic device 90 via the bus 93. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device 90, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.
[0131] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.
[0132] Example 4
[0133] Embodiment 4 of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the line granularity splitting method provided in any of the above embodiments.
[0134] The readable storage medium may include, but is not limited to, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device, or any suitable combination thereof.
[0135] Example 5
[0136] Embodiment 5 of the present disclosure further provides a computer program product, including a computer program, which implements any of the above-mentioned line granularity splitting methods when executed by a processor.
[0137] The program code for executing the computer program product of the present disclosure may be written in any combination of one or more programming languages, and the program code may be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on the remote device.
[0138] While specific embodiments of the present disclosure have been described above, those skilled in the art will appreciate that these are merely illustrative and that the scope of protection of the present disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present disclosure, and such changes and modifications are intended to fall within the scope of protection of the present disclosure.
Claims
1. A method for splitting line granularity, characterized in that: The splitting method comprises: Get the decision tree of line granularity; Obtaining a parent node of the line granularity from the decision tree; Splitting out the child nodes corresponding to the parent node; Get the index score of the parent node and the index score of the child node; A target level of line granularity is obtained based on a comparison result of the index score of the parent node and the index score of the child node.
2. The method for splitting line granularity according to claim 1, characterized in that: The step of obtaining a decision tree for line granularity includes: Get the user's search term; Obtaining a line recall set related to the search term; A decision tree of line granularity is constructed based on the line recall set.
3. The method for splitting line granularity according to claim 1, characterized in that: The step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node includes: In response to the indicator score of the child node being less than the indicator score of the parent node, and determining that the child node is a leaf node, the leaf node is used as a target level of the line granularity.
4. The method for splitting line granularity according to claim 1, characterized in that: The step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node further includes: In response to the indicator score of the child node being less than the indicator score of the parent node, and determining that the child node is not a leaf node, the child node is split, and the child node is used as the parent node to loop through the steps of obtaining the indicator score of the parent node and the indicator score of the child node; and obtaining the target level of the line granularity based on the comparison result of the indicator score of the parent node and the indicator score of the child node.
5. The method for splitting line granularity according to claim 1, characterized in that: The step of obtaining a target level of line granularity based on a comparison result of the index score of the parent node and the index score of the child node further includes: In response to the indicator score of the child node being not less than the indicator score of the parent node, the parent node is used as a target level of the line granularity.
6. The circuit granularity splitting method according to claim 1, characterized in that: The index scores include clustering scores and similarity scores.
7. A line granularity splitting system, characterized in that: The splitting system comprises: A first acquisition module is used to obtain a decision tree of line granularity; A second acquisition module is used to obtain the parent node of the line granularity from the decision tree; A splitting module, used to split out child nodes corresponding to the parent node; The third acquisition module is used to obtain the index score of the parent node and the index score of the child node; The fourth acquisition module is configured to acquire a target level of line granularity based on a comparison result of the indicator score of the parent node and the indicator score of the child node.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and configured to run on the processor, wherein: When the processor executes the computer program, the circuit granularity splitting method according to any one of claims 1 to 6 is implemented.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the circuit granularity splitting method according to any one of claims 1 to 6 is implemented.
10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the circuit granularity splitting method according to any one of claims 1 to 6 is implemented.