阅读说明：本技术 一种基于图的多点路径查询方法和装置 (Multipoint path query method and device based on graph ) 是由 郝彤 于 2019-02-22 设计创作，主要内容包括：本申请提供了一种基于图的多点路径查询方法,接收到查询请求时,针对V中的每个起始顶点扩展M层,将每个起始顶点分别记录到一个数组中；将V中的两两起始顶点对应的扩展信息求交集之后,再将所有交集内的扩展信息求并集；并获取并集中前N条扩展信息作为初步扩展信息；根据缓存中的邻接表将初步扩展信息对应的每个顶点再扩展L-M层,并获取对应的扩展信息作为二次扩展信息；合并初步扩展信息和二次扩展信息,过滤掉非起始顶点对应的扩展信息作为待返回扩展信息；获取待返回扩展信息中的前N条扩展信息,并将所述获取的前N条扩展信息中的路径信息响应给发送查询请求的设备。该方法能够提高查询效率。(The application provides a multipoint path query method based on a graph, wherein when a query request is received, M layers are expanded aiming at each initial vertex in a V, and each initial vertex is respectively recorded into an array; solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information; according to an adjacency list in the cache, re-expanding each vertex corresponding to the primary expansion information by an L-M layer, and acquiring corresponding expansion information as secondary expansion information; combining the primary extension information and the secondary extension information, and filtering extension information corresponding to the non-initial vertex as extension information to be returned; acquiring the first N pieces of extended information in the extended information to be returned, and responding the path information in the acquired first N pieces of extended information to the equipment sending the query request. The method can improve the query efficiency.)

1. A multipoint path query method based on a graph is characterized by comprising the following steps:

when a query request is received, acquiring a set V of initial vertexes, an expanded layer number L and a returned path number N carried by the request;

expanding M layers aiming at each initial vertex in the V, and respectively recording the expansion information corresponding to each initial vertex into an array; caching the passed vertexes and edges in the expansion process by using an adjacency list; wherein M is less than L;

solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections;

arranging the combined and concentrated extension information according to a preset rule; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information;

according to an adjacency list in the cache, re-expanding each vertex corresponding to the primary expansion information by an L-M layer, and acquiring corresponding expansion information as secondary expansion information;

combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to be used as extension information to be returned;

acquiring the first N pieces of extended information in the extended information to be returned, and responding the path information in the acquired first N pieces of extended information to the equipment sending the query request.

2. The method of claim 1,

the preset rule is as follows:

the layer number is from small to large, and the vertex IDs are from small to large;

or, in random order;

or, the number of layers is from small to large, with the vertices in the order of the starting vertices in V.

3. The method of claim 1,

the extended information includes: layer ID, vertex ID, and path information; the path information is a set of edges passed through in the expansion process, and one edge is represented by using a starting vertex and an ending vertex of a vertex.

4. The method of claim 1,

and M is L/2, and when L/2 is not an integer, the L/2 is rounded up to be taken as M.

5. The method of claim 1, wherein intersecting the extended information corresponding to two starting vertices in V comprises:

and if the same vertex ID exists in the path information in the extended information corresponding to every two initial vertices, determining that the extended information is the extended information in the intersection.

6. The method according to any one of claims 1-5, wherein the method further comprises:

presetting the range of the number of query initial vertexes, the range of the expanded layer times and the range of the number of returned paths;

returning a response that cannot be queried to the device that sent the query request when any of the following does not satisfy the configured scope:

the number of vertices in the set of starting vertices V, the number of layers extended L, and the number of paths returned N.

7. A graph-based multipoint path querying device, comprising: the device comprises a receiving unit, an acquisition unit, a first extension unit, a second extension unit and a response unit;

the receiving unit is used for receiving a query request;

the acquiring unit is configured to acquire, when the receiving unit receives the query request, a set V of starting vertices, an expanded layer number L, and a number N of returned paths that are carried by the request;

the first expansion unit is used for expanding M layers aiming at each initial vertex in the V in the acquisition unit and respectively recording expansion information corresponding to each initial vertex into an array; caching the passed vertexes and edges in the expansion process by using an adjacency list; solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections; arranging the combined and concentrated extension information according to a preset rule; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information; wherein M is less than L;

the second expansion unit is used for re-expanding each vertex corresponding to the preliminary expansion information by an L-M layer according to the adjacency list cached in the cache by the first expansion unit, and acquiring corresponding expansion information as secondary expansion information; combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to be used as extension information to be returned; acquiring the first N pieces of extended information in the extended information to be returned;

and the response unit is used for responding the path information in the first N pieces of expansion information acquired by the second expansion unit to the equipment sending the query request.

8. The apparatus of claim 7,

the preset rule is as follows:

the layer number is from small to large, and the vertex IDs are from small to large;

or, in random order;

or, the number of layers is from small to large, with the vertices in the order of the starting vertices in V.

9. The apparatus of claim 7,

the extended information includes: layer ID, vertex ID, and path information; the path information is a set of edges passed through in the expansion process, and one edge is represented by using a starting vertex and an ending vertex of a vertex.

10. The apparatus of claim 7,

and M is L/2, and when L/2 is not an integer, the L/2 is rounded up to be taken as M.

11. The apparatus of claim 7,

the first extension unit is specifically configured to, when intersection is obtained for the extension information corresponding to every two starting vertices in V, determine that the extension information is the extension information in the intersection if the same vertex ID exists in the path information in the extension information corresponding to every two starting vertices.

12. The apparatus according to any one of claims 7 to 11,

the acquiring unit is further configured to pre-configure a number range of query start vertices, a range of expanded layer times, and a range of returned path numbers; when any one of the acquired following items does not meet the configured scope, returning a response which cannot be queried to the equipment sending the query request: the number of vertices in the set of starting vertices V, the number of layers extended L, and the number of paths returned N.

13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-6 when executing the program.

14. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the method of any one of claims 1 to 6.

Technical Field

The present invention relates to the field of information processing technologies, and in particular, to a multipoint path query method and apparatus based on a graph.

Background

The Graph (Graph) is an object for representing the relationship between objects, and is a basic research object of Graph theory. The graph is an ordered binary set (V, E), where V is called the set of vertices and E is called the set of edges. The element of E is represented by (x, y), where x, y ∈ V.

It makes sense to calculate the relationships between multiple vertices in a graph, such as: calculating the relationship of two persons in the network, calculating the investment relationship among a plurality of companies, and the like.

The traditional calculation method comprises the following steps: and expanding towards the outer layer by taking one node as a center until the node is expanded to a destination node or a defined hierarchy is reached.

When a vertex expands to the outer layer, along with the increase of the expansion calculation level, the calculation nodes form exponential increase, so the calculation mode not only has low efficiency, but also has larger consumption on the memory.

Such as: assuming that each vertex in the graph has a direct relationship with 100 vertices, if the 2-level relationship of the vertex is to be traversed, the load of the data of the 2 nd power of 100, namely 10000 vertices, is involved. If a 6-level relationship is computed, 1 trillion vertex data is loaded.

Disclosure of Invention

In view of this, the present application provides a multipoint path query method and apparatus based on a graph, which can improve query efficiency.

In order to solve the technical problem, the technical scheme of the application is realized as follows:

a multipoint path query method based on a graph comprises the following steps:

when a query request is received, acquiring a set V of initial vertexes, an expanded layer number L and a returned path number N carried by the request;

expanding M layers aiming at each initial vertex in the V, and respectively recording the expansion information corresponding to each initial vertex into an array; caching the passed vertexes and edges in the expansion process by using an adjacency list; wherein M is less than L;

solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections;

arranging the combined and concentrated extension information according to a preset rule; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information;

according to an adjacency list in the cache, re-expanding each vertex corresponding to the primary expansion information by an L-M layer, and acquiring corresponding expansion information as secondary expansion information;

combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to be used as extension information to be returned;

acquiring the first N pieces of extended information in the extended information to be returned, and responding the path information in the acquired first N pieces of extended information to the equipment sending the query request.

A graph-based multipoint path querying device, the device comprising: the device comprises a receiving unit, an acquisition unit, a first extension unit, a second extension unit and a response unit;

the receiving unit is used for receiving a query request;

the acquiring unit is configured to acquire, when the receiving unit receives the query request, a set V of starting vertices, an expanded layer number L, and a number N of returned paths that are carried by the request;

the first expansion unit is used for expanding M layers aiming at each initial vertex in the V in the acquisition unit and respectively recording expansion information corresponding to each initial vertex into an array; caching the passed vertexes and edges in the expansion process by using an adjacency list; solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections; arranging the combined and concentrated extension information according to a preset rule; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information; wherein M is less than L;

the second expansion unit is used for re-expanding each vertex corresponding to the preliminary expansion information by an L-M layer according to the adjacency list cached in the cache by the first expansion unit, and acquiring corresponding expansion information as secondary expansion information; combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to be used as extension information to be returned; acquiring the first N pieces of extended information in the extended information to be returned;

and the response unit is used for responding the path information in the first N pieces of expansion information acquired by the second expansion unit to the equipment sending the query request.

An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the graph-based multipoint path querying method when executing the program.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the graph-based multipoint path querying method.

According to the technical scheme, the initial vertex is expanded in the M level mode, then the vertexes in the expanded information intersection are obtained to expand the residual L-M level, and query information is fed back according to the number of the return paths. After the front M levels are expanded, intersection processing of the expanded information is carried out, and data information can be converged; when the rest L-M levels are expanded, the cache data generated by the expansion of the previous M levels are used, and the query efficiency is improved.

Drawings

FIG. 1 is a schematic diagram of a multipoint path query process based on a graph in an embodiment of the present application;

FIG. 2 is a diagram illustrating vertex relationships in a graph according to an embodiment;

fig. 3 is a schematic structural diagram of an apparatus applied to the above-described technology in the embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more clearly apparent, the technical solutions of the present invention are described in detail below with reference to the accompanying drawings and examples.

The embodiment of the application provides a multipoint path query method based on a graph, which comprises the steps of firstly carrying out M-level expansion on an initial vertex, then obtaining the vertexes in an expanded information intersection to carry out residual L-M-level expansion, and feeding back query information according to the number of return paths. After the front M levels are expanded, intersection processing of the expanded information is carried out, and data information can be converged; when the rest L-M levels are expanded, the cache data generated by the expansion of the previous M levels are used, and the query efficiency is improved.

For convenience of description in the embodiments of the present application, a main body that executes a query request and responds to a query result is referred to as a query device. The querying device only has the following functions, and the specific existence form is not limited.

The following describes in detail a multipoint graph-based route query procedure in an embodiment of the present application with reference to the drawings.

Referring to fig. 1, fig. 1 is a schematic diagram of a multipoint path query process based on a graph in the embodiment of the present application. The method comprises the following specific steps:

step 101, when receiving a query request, acquiring a set V of initial vertices, an expanded layer number L, and a returned path number N carried by the request.

In the embodiment of the present application, during specific implementation, the number range of the query start vertex, the range of the expanded layer number, and the range of the number of the returned paths need to be configured in advance to check whether the query parameter carried by the query request is legal.

For example, the number of the starting vertices ranges from 1 to 100, the number of the expanded layers ranges from 1 to 10, and the number of the returned paths ranges from 1 to 10000, but in a specific implementation, the number is not limited to the parameter range given in the example, and may be set according to actual needs.

Returning a response that cannot be queried to the device that sent the query request when any of the following does not satisfy the configured scope:

the number of vertices in the set of starting vertices V, the number of layers extended L, and the number of paths returned N.

That is to say, if the number of vertices in the set V of starting vertices, the number of expanded layers L, and the number N of returned paths do not satisfy the configured range, it is determined that the query parameter is not legal, and it is further determined that the corresponding path information cannot be responded, and at this time, a response that cannot be queried is returned to the device that sent the query request.

If the number of vertices in V is 50, the parameter for determining the number of start vertices is legal, and if the expanded layer number L is 12, the parameter for determining the expanded layer number is illegal.

When all three parameters are determined to be legal, step 102 is performed.

The query request also usually carries an expansion mode, and the expansion mode usually consists of three types: edge out, edge in and two-way; the bidirectional fingers comprise an edge outgoing direction and an edge incoming direction, and during specific query, query is carried out according to an expansion mode carried in a query request;

and if the expansion mode is not carried in the query request, the expansion is carried out by default in a bidirectional mode.

102, expanding M layers aiming at each initial vertex in V, and respectively recording expansion information corresponding to each initial vertex into an array; and caching the passed vertexes and edges in the expansion process by using an adjacency list.

M in the embodiment of the application is smaller than L, and in particular implementation, M can be half of L, that is, M can be L/2, and when L/2 is not an integer, L/2 is rounded up to be M.

If L is 4, M is 2, if M is 5, M is 3.

The extended information in the embodiment of the present application includes: layer ID, vertex ID, and path information; wherein, the layer ID indicates that the current extension is to the layer number; the vertex ID represents the current vertex ID after expansion; the path information is a set of edges passed through in the expansion process, and one edge is represented by using a start vertex and an end vertex of a vertex.

The adjacency list comprises a vertex and an edge corresponding to the vertex.

Referring to FIG. 2, FIG. 2 is a schematic diagram illustrating vertex relationships in a graph according to an embodiment. Fig. 2 includes 12 vertices, and the query request carries V (vertex 1, vertex 11), L is 4, and N is 5 as an example.

Expand in a bidirectional manner as an example: expanding L/2 to 2 layers aiming at the vertex 1 and the vertex 2, and respectively recording the expansion information obtained in the process of expanding the starting vertex 1 and the starting vertex 11 into two arrays (array 1 and array 2);

storing the corresponding extended information of the vertex 1 and the vertex 2 by the array 1 and the array 2 as follows:

the array 1 contains the following contents: {12(1,2),13(1,3),14(1,4),25(1,2)(5,2),26(1,3)(3,6),26(1,4)(6,4)}.

The content of the array 2 is: {112(11,12),26(11,12)(12,6)}.

Still taking fig. 2 as an example, the contents of the adjacency list generated during the previous 2-layer expansion are shown in table 1, and table 1 is the contents included in the adjacency list in the embodiment of the present application.

TABLE 1

And 103, solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all the intersections.

Since the number of the starting vertices in V in the embodiment of the present application is only two, the result after the intersection is taken for the two arrays is the union result in step 103.

The intersection determination here is to determine extended information corresponding to the same vertex in the path information in the extended information.

If the same vertex ID exists in the path information in the extended information corresponding to every two initial vertices, determining the extended information as the extended information in the intersection; otherwise, determining that the corresponding extension information is not extension information in the intersection, and not remaining in the union result.

If the same vertex in the path information is vertex 6 in array 1 and array 2, the union result in step 103 is:

{2 6(1,3)(3,6)，2 6(1,4)(6,4)，2 6(11,12)(12,6)}。

when the extension information in all the intersection sets is empty, namely the union set is empty, namely the extension information of any node does not exist, responding to the equipment which sends the query request that the path information is not queried; when the extended information in all the intersections is not empty, that is, the union is not empty, step 104 is executed.

104, arranging the merged extended information according to a preset rule; and acquiring and concentrating the first N pieces of extension information as preliminary extension information.

The preset rule is as follows:

the layer number is from small to large, and the vertex IDs are from small to large;

or, in random order;

or, the number of layers is from small to large, with the vertices in the order of the starting vertices in V.

The three preset rules are provided, and in practical application, the preset rules may not be limited, and the user sets the preset rules according to a specific application scenario to perform sorting, and if the number of layers is from small to large, and the vertexes are sorted according to the preset rule of the order of the initial vertexes in the V, the vertex sorting in the V in the request information is considered.

In the special case where the layers are the same and the vertex IDs are the same in the extended information in the union result, the extended information may be randomly arranged.

In the above example, N is 5, 3 pieces of extension information in the union result are all retained, and if the number of the extension information exceeds 5, the 6 th and the extension information after the 6 th are all filtered out.

And 105, re-expanding each vertex corresponding to the primary expansion information by an L-M layer according to the adjacency list in the cache, and acquiring corresponding expansion information as secondary expansion information.

The preliminary extension information after filtering the union result is as follows:

{2 6(1,3)(3,6)，2 6(1,4)(6,4)，2 6(11,12)(12,6)}。

it is possible or only vertex 6 in the preliminary extension information is used, and then the L-M layer is re-extended from vertex 6 to 4-2 layer, that is, the layer 3 and layer 4 extensions are performed.

And in the subsequent layer expansion process, the adjacency list generated in the previous M layer expansion process can be used for expansion, so that the expansion efficiency can be greatly improved.

The vertex 6 is expanded in the 3 rd layer and the 4 th layer, the generated expanded information is recorded in the array 3 as secondary expanded information, and the data 3 includes:

{4 1(11,12)(12,6)(3,6)(1,3)，4 1(11,12)(12,6)(6,4)(1,4)，3 3(11,12)(12,6)(3,6)，3 4(11,12)(12,6)(6,4)，3 12(1,3)(3,6)(12,6)，3 12(1,4)(6,4)(12,6)，4 11(1,3)(3,6)(12,6)(11,12)，4 11(1,4)(6,4)(12,6)(11,12)}。

and 106, combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to serve as the extension information to be returned.

The merging result after merging the primary extended information and the secondary extended information is as follows:

{2 6(1,3)(3,6)，2 6(1,4)(6,4)，2 6(11,12)(12,6)，4 1(11,12)(12,6)(3,6)(1,3)，4 1(11,12)(12,6)(6,4)(1,4)，3 3(11,12)(12,6)(3,6)，3 4(11,12)(12,6)(6,4)，3 12(1,3)(3,6)(12,6)，3 12(1,4)(6,4)(12,6)，4 11(1,3)(3,6)(12,6)(11,12)，4 11(1,4)(6,4)(12,6)(11,12)}。

after filtering out the extension information corresponding to the non-initial vertex, the reserved extension information is as follows:

{4 1(11,12)(12,6)(3,6)(1,3)，4 1(11,12)(12,6)(6,4)(1,4)，4 11(1,3)(3,6)(12,6)(11,12)，4 11(1,4)(6,4)(12,6)(11,12)}。

that is, only the information corresponding to the vertices having vertex IDs 1 and 11 is retained.

The preset rule is as follows:

the layer number is from small to large, and the vertex IDs are from small to large;

or, in random order;

or, the number of layers is from small to large, with the vertices in the order of the starting vertices in V.

After the layers are arranged in the order from small to large and the vertex IDs are arranged in the order from small to large, the layers cannot be further arranged, if the vertexes represent the extension information with the same layer number, the front and back orders are random, and the extension information retained after the ordering can be in the following mode according to the rule:

{4 1(11,12)(12,6)(3,6)(1,3)，4 1(11,12)(12,6)(6,4)(1,4)，4 11(1,3)(3,6)(12,6)(11,12)，4 11(1,4)(6,4)(12,6)(11,12)}。

and taking the extension information in the array as the extension information to be returned.

And 107, acquiring the first N pieces of extended information in the extended information to be returned, and responding the path information in the acquired first N pieces of extended information to the equipment sending the query request.

Still taking N as 5 as an example, all the path information in the extended information to be returned is responded to the device sending the query request, and the specific path information is as follows:

(11,12)(12,6)(3,6)(1,3)，(11,12)(12,6)(6,4)(1,4)，(1,3)(3,6)(12,6)(11,12)，(1,4)(6,4)(12,6)(11,12)}。

at this point, the query is completed and the query result is fed back to the device sending the query request.

Based on the same inventive concept, the embodiment of the application also provides a multipoint path inquiry device based on the graph. Referring to fig. 3, fig. 3 is a schematic structural diagram of an apparatus applied to the above technology in the embodiment of the present application. The device includes: a receiving unit 301, an acquiring unit 302, a first expanding unit 303, a second expanding unit 304, and a responding unit 305;

a receiving unit 301, configured to receive a query request;

an obtaining unit 302, configured to obtain, when the receiving unit 301 receives the query request, a set V of start vertices, an expanded layer number L, and a returned path number N that are carried by the request;

a first expanding unit 303, configured to expand M layers for each starting vertex in the V in the obtaining unit, and record expansion information corresponding to each starting vertex in an array respectively; caching the passed vertexes and edges in the expansion process by using an adjacency list; solving an intersection of the extended information corresponding to every two initial vertexes in the V, and then solving a union of the extended information in all intersections; arranging the combined and concentrated extension information according to a preset rule; acquiring and concentrating the first N pieces of expansion information as preliminary expansion information; wherein M is less than N;

a second expansion unit 304, configured to re-expand each vertex corresponding to the preliminary expansion information by an L-M layer according to the adjacency list cached in the cache by the first expansion unit 303, and acquire corresponding expansion information as secondary expansion information; combining the primary extension information and the secondary extension information, filtering out extension information corresponding to the non-initial vertex, and arranging the extension information according to a preset rule to be used as extension information to be returned; acquiring the first N pieces of extended information in the extended information to be returned;

a responding unit 305, configured to respond to the path information in the first N pieces of extension information acquired by the second extension unit 304 to the device sending the query request.

Preferably, the first and second liquid crystal films are made of a polymer,

the preset rule is as follows:

the layer number is from small to large, and the vertex IDs are from small to large;

or, in random order;

or, the number of layers is from small to large, with the vertices in the order of the starting vertices in V.

Preferably, the first and second liquid crystal films are made of a polymer,

the extended information includes: layer ID, vertex ID, and path information; the path information is a set of edges passed through in the expansion process, and one edge is represented by using a starting vertex and an ending vertex of a vertex.

Preferably, the first and second liquid crystal films are made of a polymer,

and M is L/2, and when L/2 is not an integer, the L/2 is rounded up to be taken as M.

Preferably, the first and second liquid crystal films are made of a polymer,

the first extension unit 303 is specifically configured to, when intersection is obtained for the extension information corresponding to every two starting vertices in V, determine that the extension information is the extension information in the intersection if the same vertex ID exists in the path information in the extension information corresponding to every two starting vertices.

Preferably, the first and second liquid crystal films are made of a polymer,

an obtaining unit 302, further configured to pre-configure a number range of query start vertices, a range of expanded layer times, and a range of returned path numbers; when any one of the acquired following items does not meet the configured scope, returning a response which cannot be queried to the equipment sending the query request: the number of vertices in the set of starting vertices V, the number of layers extended L, and the number of paths returned N.

The units of the above embodiments may be integrated into one body, or may be separately deployed; may be combined into one unit or further divided into a plurality of sub-units.

In addition, an electronic device is further provided in an embodiment of the present application, and includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the graph-based multi-point path query method.

In addition, a computer-readable storage medium is provided in an embodiment of the present application, and a computer program is stored thereon, and when being executed by a processor, the computer program implements the steps of the graph-based multipoint path querying method.

In summary, according to the method and the device, M-level expansion is performed on the initial vertex in advance, then, the vertex in the expanded information intersection is obtained to perform the remaining L-M-level expansion, and query information is fed back according to the number of return paths. After the front M levels are expanded, intersection processing of the expanded information is carried out, and data information can be converged; when the rest L-M levels are expanded, the cache data generated by the expansion of the previous M levels are used, and the query efficiency is improved.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.