WO2023124253A1

WO2023124253A1 - Graph search method and apparatus, device and storage medium

Info

Publication number: WO2023124253A1
Application number: PCT/CN2022/119014
Authority: WO
Inventors: 崖飞虎; 周可; 邸帅; 卢道和
Original assignee: 深圳前海微众银行股份有限公司
Priority date: 2021-12-28
Filing date: 2022-09-15
Publication date: 2023-07-06
Also published as: CN114385864A

Abstract

Provided are a graph search method and apparatus, a device, and a storage medium. The method comprises: receiving a selected search mode and a search vertex input at a search entry of a front-end page (S201), the search vertex being a point in a directed acyclic graph; determining an adaptation algorithm corresponding to the search mode, and executing the adaptation algorithm to search on the basis of the search vertex to obtain a search result (S202); displaying the search result by means of a visualization plug-in (S203).

Description

A graph search method, device, equipment and storage medium

Cross References to Related Applications

This application claims the priority of the Chinese patent application with the application number 202111627096.0 and the application title "A Graph Search Method, Device, Equipment, and Storage Medium" submitted to the China Patent Office on December 28, 2021, the entire contents of which are incorporated by reference incorporated in this application.

technical field

The embodiment of the present application relates to the technical field of financial technology (Fintech) data processing, and relates to but not limited to a graph search method, device, device and storage medium.

Background technique

With the development of computer computing, more and more technologies are applied in the financial field, and the traditional financial industry is gradually transforming into Fintech. However, due to the security and real-time requirements of the financial industry, higher requirements are placed on technology. requirements.

In the field of financial technology, enterprises have various graph search scenarios, such as data kinship search, application kinship search, etc. Usually, the amount of graph model data in an enterprise is very large. At present, the complex search of graphs can only be performed through the back-end offline algorithm. To achieve, there is no real-time return and visual graphic display solution.

Contents of the invention

The embodiment of the present application provides a graph search method, device, equipment and storage medium to solve the problem in the related art that the complex search of graphs can only be realized through the back-end offline algorithm, and there is no real-time return and visual graph display solution.

The technical scheme of the embodiment of the application is realized in this way:

An embodiment of the present application provides a graph search method, including:

Receive a search vertex and a selected search mode input at the search entry of the front-end page; wherein, the search vertex is a point in a directed acyclic graph;

determining an adaptation algorithm corresponding to the search mode, and executing the adaptation algorithm to perform a search based on the search vertex to obtain a search result;

The search results are displayed through a visual plug-in.

An embodiment of the present application provides a graph search device, including:

The display module is used to display the search entry of the front-end page;

A receiving module, configured to receive a search vertex input at the search entry and a selected search mode; wherein the search vertex is a point in a directed acyclic graph;

A processing module, configured to determine an adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm to perform a search based on the search vertex to obtain a search result;

The display module is configured to display the search results through a visual plug-in.

The memory is used to store executable instructions; the processor is used to implement the above method when executing the executable instructions stored in the memory.

An embodiment of the present application provides a storage medium, which stores executable instructions, and is used to cause a processor to implement the above method when executed.

The embodiment of the present application has the following beneficial effects:

By receiving the search vertex and the selected search mode input in the search entry of the front-end page; wherein, the search vertex is a point in a directed acyclic graph; determine the adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm based on the search The vertices are searched to get the search results; the search results are displayed through the visualization plug-in. It solves the problem that the complex search of graphs in related technologies can only be realized through the back-end offline algorithm, and there is no real-time return and visual graphic display solution, and supports online analytical processing (OLAP) complex scene search of ultra-large-scale data sets And a complete set of solutions for visual display of various graph models.

Description of drawings

FIG. 1 is a schematic diagram of an optional architecture of a server provided in an embodiment of the present application;

Fig. 2 is a schematic flow diagram of a graph search method provided in an embodiment of the present application;

FIG. 3 is a schematic diagram of the architecture of the graph visualization analysis system provided by the embodiment of the present application;

Fig. 4 is the second schematic flow diagram of the graph search method provided by the embodiment of the present application;

Fig. 5 is a schematic diagram of the directed acyclic graph G and the interval label graph of vertices provided by the embodiment of the present application;

Fig. 6 is a schematic diagram of the directed acyclic graph G and the forward and reverse breadth layers and the forward and reverse topology layers of the vertices provided by the embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the application clearer, the application will be further described in detail below in conjunction with the accompanying drawings. All other embodiments obtained under the premise of creative labor belong to the scope of protection of this application.

In the following description, references to "some embodiments" describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or a different subset of all possible embodiments, and Can be combined with each other without conflict. Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meaning as commonly understood by those skilled in the technical field of the embodiments of the present application. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.

The following describes the exemplary application of the graph search device provided by the embodiment of the present application. The graph search device provided by the embodiment of the present application can be implemented as a notebook computer, a tablet computer, a desktop computer, a mobile device (for example, a mobile phone, a portable music player, Personal digital assistants, special messaging devices, portable game devices), intelligent robots and other terminals with screen display functions can also be implemented as servers. Next, an exemplary application when the graph search device is implemented as a server will be explained.

Referring to FIG. 1, FIG. 1 is a schematic structural diagram of a server 100 provided by an embodiment of the present application. The server 100 shown in FIG. Various components in the server 100 are coupled together through the bus system 140 . It can be understood that the bus system 140 is used to realize connection and communication between these components. In addition to the data bus, the bus system 140 also includes a power bus, a control bus and a status signal bus. However, the various buses are labeled as bus system 140 in FIG. 1 for clarity of illustration.

Processor 110 can be a kind of integrated circuit chip, has signal processing capability, such as general-purpose processor, digital signal processor (DSP, Digital Signal Processor), or other programmable logic device, discrete gate or transistor logic device, discrete hardware Components, etc., wherein the general-purpose processor can be a microprocessor or any conventional processor, etc.

User interface 130 includes one or more output devices 131 that enable presentation of media content, including one or more speakers and/or one or more visual displays. The user interface 130 also includes one or more input devices 132, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

Memory 150 may be removable, non-removable or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disk drives, and the like. Memory 150 optionally includes one or more storage devices located physically remote from processor 110 . Memory 150 includes volatile memory or nonvolatile memory, and may include both volatile and nonvolatile memory. The non-volatile memory can be read-only memory (Read Only Memory, ROM), and the volatile memory can be random access memory (Random Access Memory, RAM). The memory 150 described in the embodiment of the present application is intended to include any suitable type of memory. In some embodiments, the memory 150 is capable of storing data to support various operations, examples of which include programs, modules, and data structures, or subsets or supersets thereof, as exemplified below.

Operating system 151, including system programs for processing various basic system services and performing hardware-related tasks, such as framework layer, core library layer, driver layer, etc., for implementing various basic services and processing hardware-based tasks;

Network communication module 152 for reaching other computing devices via one or more (wired or wireless) network interfaces 120, exemplary network interfaces 120 include: Bluetooth, Wireless Compatibility Authentication (Wi-Fi), and Universal Serial Bus (Universal Serial Bus, USB), etc.;

The input processing module 153 is configured to detect one or more user inputs or interactions from one or more of the input devices 132 and translate the detected inputs or interactions.

In some embodiments, the device provided by the embodiment of the present application can be realized by software. FIG. 1 shows a graph search device 154 stored in the memory 150. The graph search device 154 can be a graph search The device, which can be software in the form of programs and plug-ins, includes the following software modules: a display module 1541, a receiving module 1542, and a processing module 1543. These modules are logical, so any combination or further split. The function of each module will be explained below.

In other embodiments, the device provided in the embodiment of the present application may be implemented in hardware. As an example, the device provided in the embodiment of the present application may be a processor in the form of a hardware decoding processor, which is programmed to execute the In the graph search method provided by the embodiment, for example, the processor in the form of a hardware decoding processor may adopt one or more Application Specific Integrated Circuits (Application Specific Integrated Circuit, ASIC), DSP, Programmable Logic Device (Programmable Logic Device, PLD), Complex Programmable Logic Device (Complex Programmable Logic Device, CPLD), Field Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other electronic components.

The graph search method provided by the embodiment of the present application will be described below in conjunction with the exemplary application and implementation of the server 100 provided in the embodiment of the present application. Referring to FIG. 2, FIG. 2 is an optional flowchart of the graph search method provided by the embodiment of the present application, which will be described in conjunction with the steps shown in FIG. 2.

Step S201, receiving the search vertex and the selected search mode entered in the search entry of the front page.

Among them, the search vertex is a point in the directed acyclic graph.

In the embodiment of the present application, a directed acyclic graph (Directed acyclic graph, DAG) is a data set composed of vertices and edges, the vertices represent an entity, and the edges represent the relationship between entities. Here, use G=(V, E) represents a directed acyclic graph for short, V represents the set of vertices in graph G, and E represents the set of edges in graph G.

In the embodiment of the present application, during the search process, the number of input search vertices is greater than or equal to 1. When the number of search vertices is multiple, the search mode includes the mode corresponding to the reachability analysis and the mode corresponding to the path search between two points; when the number of search vertices is single, the search mode includes a single The search fixed-point is the pattern corresponding to the K-step search of a specific vertex. Among them, accessibility refers to: any two vertices u and v in the graph, if there is a path starting from u that can reach v, it means that vertex u can reach vertex v. The path refers to: a path from vertex u to vertex v is a vertex sequence (u, u1,...,v) starting from vertex u, and the vertex sequence ends at vertex v.

In this embodiment of the application, during the search process, the graph database used for querying and storing the graph model can be an open source distributed graph database, such as nebulagraph, which supports millisecond-level query of very large-scale data sets. When searching at the search entry of the front-end page, it can be realized through the search engine of the front-end page. The search engine can be a distributed full-text search engine, such as elasticsearch, which supports word segmentation, indexing and fast retrieval of massive data.

In practical applications, enterprises have various graph search scenarios, including but not limited to data kinship search in data traceability and application kinship search in application traceability. Faced with the huge amount of graph model data in an enterprise, this embodiment of the application provides a search and visualization platform, which presents front-end pages to users, and users can perform various searches on graphs on the platform.

Step S202, determine the adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm to search based on the search vertex, and obtain the search result.

In the embodiment of the present application, the corresponding adaptation algorithm is different if the selected search mode is different; the adaptation algorithm in the present application is a search method adopted in the process of searching for the input search fixed point. It should be noted that in the embodiment of this application, the adaptation algorithm is an algorithm design for the graph model in the graph database, which involves the operation of the graph model; Disassembly-based graph semantic operations, such as viewing vertex attributes, obtaining adjacent vertices, obtaining edges between two points, these most basic graph semantic operations, that is to say, in the search process, the adaptation algorithm will The operation is decomposed into basic graph semantic atomic operations, thereby reducing the complexity of graph operations; using elasticsearch to search, indexing all algorithm features, and using the index only requires one-time retrieval to obtain search results, which improves the overall stability of the service sex.

In some embodiments of the present application, when executing the adaptation algorithm to search based on the search vertex, the breadth-first traversal method (Breath First Search, BFS) can be used to process the vertices according to the hierarchy, and the vertices closest to the starting point are visited first, in order of distance Traverse the vertices in the graph in turn.

Step S203, displaying search results through a visualization plug-in.

In the embodiment of the present application, the visualization plug-in includes but is not limited to the CYSOSCAPE.JS visualization plug-in. The visualization plug-in can flexibly assemble the visualization structure packaged in the backend into various graph visualizations, and perform rendering and color matching. In some embodiments, the visualization plug-in can flexibly assemble the visualization structure into various graphs in the following way: take the input search vertex as the root node, and then expand outward layer by layer to build the entire complete graph relationship link; specifically The set of all vertices of the structure and the edges between any two vertices. Then, the visualization plug-in takes the input search vertex as the root node, and constructs a complete graph layer by layer outward.

In practical applications, after users can search various graphs on the platform, further, the platform can return the search results in real time through the visualization plug-in, and display the search results visually. It can be seen that the graph search method provided by this application is applicable to various OLAP search scenarios of graphs, such as reachability analysis, two-point link search, and k-step search. Various search conditions, and then obtain the results of visual display of various graph models.

It can be seen from this that this application can support users to complete various types of graph relationship data (blood relationship data, risk control data, etc.) Configure different conditional queries in different search scenarios to realize visual search.

The graph search method provided by the embodiment of the present application receives the search vertex and the selected search mode input in the search entry of the front page; wherein, the search vertex is a point in a directed acyclic graph; and the appropriate search mode corresponding to the search mode is determined. Algorithm matching, and execution of the matching algorithm to search based on the search vertex to obtain the search results; display the search results through the visualization plug-in; solve the problem that the complex search of the graph in the related technology can only be realized through the back-end offline algorithm, without a real-time return and visualization For the problem of graphic display solutions, it supports a complete set of solutions for OLAP complex scene search and visual display of various graph models.

Further, in a graph search scenario, this application proposes a schematic diagram of a graph search architecture, as shown in Figure 3, the graph search architecture can also be a graph visualization analysis system, and the graph visualization analysis system includes front-end WEB services There are two parts, and the graph analysis engine. Among them, the front-end WEB service includes the following components: the front-end search entry and the CYSOSCAPE.JS visualization plug-in. Here, the front-end search entry is used as the search entry of the front-end page, which supports search modes such as graph accessibility, path between two points, and specific vertex K-step search, and various search conditions can be configured. The CYSOSCAPE.JS visualization plug-in, as a visualization plug-in for the front-end diagram, can flexibly assemble the visualization structure packaged in the back-end into various visualization graphics for rendering and color matching. Among them, the graph analysis engine includes the following components: application programming interface (Application Programming Interface, API), standardized template, adapter and policy factory, algorithm library and graph atomic semantic operation layer. Here, the graph analysis engine corresponding to the API provides graph analysis functions for various scenarios through spring boot microservices. The standardized template is a standardized template for graph analysis, which defines the standardized steps from the API to the final packaging process. The standardization steps are: add root node -> adapt algorithm strategy -> execute search algorithm -> adapt packaging strategy -> execute packaging method -> return packaging structure. Adapters and policy factories can adapt different algorithm strategies and packaging strategies according to various search modes and conditions input by the front end. As a self-developed high-performance search algorithm integrated algorithm library, the algorithm library supports reachability analysis, path search between two points, and K-step search of specific vertices and other search modes. Elasticsearch is used to index various features of vertices, such as degree, label interval, breadth layer, and topology layer. The graph atomic semantic operation layer refers to decomposing various algorithms in the algorithm library into the smallest atomic operations on semantics, such as viewing vertex attributes, obtaining adjacent vertices, and obtaining edges between two points. The storage and query of graph models use nebulagraph graph database.

Further, in combination with the above graph search architecture shown in Figure 3 and Figure 4, the graph search process of this application is divided into three stages for description:

The first stage, search mode and algorithm relationship: various search modes input by the front-end determine the specific algorithm for back-end adaptation, that is, the above-mentioned adaptation algorithm. For example, the front-end searches for whether two application vertices are related by blood, and the algorithm adapter will automatically match them. Reachability analysis strategy, calling the reachability analysis algorithm.

The second stage, the relationship between search conditions and pruning strategies: the search conditions configured on the front end will be configured into different condition categories according to the search mode, and the search results will be pruned; for example, during the K-step search process, search for a table 10 degrees upstream Dependency table, and configure a time interval condition, when using BFS search, pruning will be performed according to the time condition. Here, pruning refers to removing unqualified vertices and edges during graph traversal to improve the time and space efficiency of the traversal algorithm.

The third stage, returning the structure: the search results need to be packaged into a json structure of vertices and edges, the vertex set is the unnecessary set of vertices on all paths, and the edge is the directed edge set between two adjacent vertices, CYSOSCAPE. The JS plug-in will automatically assemble and render the returned structure; the specific assembly process is to add the input node to the root node, and then expand outward from the edge layer upstream and downstream of the root node to build a complete graph relationship link.

The following is a further description of the graph search process of this application:

In other embodiments of the present application, the search vertex includes vertex u and vertex v. In step S202, the adaptation algorithm is executed to search based on the search vertex, and the search result is obtained, which can be realized through the following steps:

A11. Execute the adaptation algorithm to judge whether the reachability condition between vertex u and vertex v is satisfied based on the recursive calling method;

A12. If the reachability condition is satisfied between the vertex u and the vertex v, perform a search based on the relationship between the out-degree outDeg(u,G) of the vertex u and the in-degree inDeg(v,G) of the vertex v, and obtain the search result. Among them, G represents a directed acyclic graph.

In the embodiment of the present application, the degree of a vertex refers to the number of adjacent vertices of the vertex. In a directed graph, the degree of a vertex is the sum of the out-degree and in-degree of the vertex. Use inDeg(u,G) to mark the in-degree of u, that is, the number of vertices pointing to u, and use outDeg(u,G) to mark the out-degree of u, that is, the number of vertices pointed to by u.

In other embodiments of the present application, the determination in A11 whether the reachability condition is satisfied between the vertex u and the vertex v can be realized through the following steps:

A111, judging the interval label relationship between vertex u and vertex v;

A112, judging the relationship between the forward breadth layers and reverse breadth layers of vertex u and vertex v;

A113, judging the relationship between the forward and reverse topological layers of vertex u and vertex v;

A114. Based on at least one of the interval label relationship, the forward breadth layer number and reverse breadth layer number relationship, and the forward topology layer number and reverse topology layer number relationship, determine whether the reachability condition between the vertex u and the vertex v is satisfied.

In some scenarios, as shown in Figure 5, each vertex in graph G has two interval labels, for example, for vertex v there is the first interval label of vertex v

and the second interval label of vertex v

in,

and

in

Indicates the starting value of the i-th interval,

Identifies the end value of the i-th interval, where i=1,2. the first interval label of vertex v

It is the label obtained when the child vertex of vertex v is traversed forward in the pre-order traversal of vertex v, and the starting value of the interval

For the current vertex, include the minimum value of the start value and the end value of the child vertex

Access the sequential value of v for subsequent traversals; the second interval label

It is the label obtained by traversing the child vertices in reverse order when traversing the vertex v for the second time, similarly

It is the minimum value of the starting value among the child vertices contained in the current vertex,

Ordered value for accessing vertex v for subsequent traversals. Figure 5

is the first value in the first bracket,

is the second value in the first square bracket; for example, for vertex 1

it's 1,

is 10.

is the first value in the second bracket,

is the second value in the second square bracket; for example, for vertex 1 corresponding

it's 1,

is 10.

Further, as shown in Figure 6, for the directed graph G shown in Figure 5 above, define the number of forward breadth layers and reverse breadth layers of vertex v, the number of forward breadth layers of each vertex in Figure 5 is Figure 6 The value before the comma before the middle separator, and the number of reverse breadth layers is the value after the comma, for example, the number of forward breadth layers for vertex 1 is 1, and the number of reverse breadth layers is 6. The specific definition of the number of positive breadth layers is as follows:

Among them, inN(v, G) identifies the in-degree of vertex v, and the number of forward breadth layers of a vertex with an in-degree of 0 is 1, otherwise the number of forward breadth layers is equal to the number of forward breadth layers of all in-degree vertices of the vertex + 1 The minimum value of , that is, the number of forward breadth layers of vertex v is equal to the layer value of the parent node of v+1 when performing breadth-first traversal on the graph G.

The reverse breadth layer rebre(v) of vertex v is to reverse all the directed edges of graph G, and then perform the same calculation to obtain the reverse breadth layer of all vertices.

Referring to Figure 6, for the directed graph G shown in Figure 5 above, define the number of forward topological layers and reverse topological layers of vertex v, and the number of forward topological layers of each vertex in Figure 5 is the segmentation graph in Figure 6 After the value before the comma, the number of reverse topology layers is the value after the comma, for example, the number of forward topology layers for vertex 1 is 1, and the number of reverse topology layers is 1. The specific definition of forward topology layers is as follows:

Among them, the number of forward topological layers of a vertex with an in-degree of 0 is 1, otherwise the number of forward topological layers is equal to the maximum value after the number of forward topological layers of all in-degree vertices of the vertex + 1, that is, the number of forward topological layers of vertex v is equal to The traversal order value when performing pre-order traversal on vertices with in-degree 0;

The specific definition of reverse topology layers is as follows:

Among them, the reverse topology layer number of the vertex whose out-degree is 0 is the length of the longest path in the current graph, and the reverse topology layer number of other vertices v is the minimum value after the reverse topology layer number of all out-degree vertices of v -1.

In other embodiments of the present application, A111 judges the interval label relationship between vertex u and vertex v, which can be realized through the following steps: judge the first interval label of vertex v

Is it the first interval label of vertex u

within , and the second interval label of vertex v

Whether in the second interval label of vertex u

within;

Among them, the first interval label of each vertex is the label obtained when the child vertex of each vertex is traversed forward in the preorder traversal of each vertex, and the second interval label of each vertex is the second interval label for each vertex. When a vertex is traversed in pre-order, the label is traversed according to the reverse order of the child vertices;

Among them, whether the reachability condition is satisfied between the vertex u and the vertex v, including: if

exist

within and

exist

Within, it is determined that the reachability condition is satisfied between the vertex u and the vertex v.

That is to say, in the embodiment of this application, for any two vertices u and v in a given directed acyclic graph G, if

or

Then u->v is unreachable.

In other embodiments of the present application, A112 judges the relationship between the number of forward breadth layers and the number of reverse breadth layers of vertex u and vertex v, which can be realized through the following steps:

Determine whether the number of forward breadth layers bre(v) of vertex v is greater than the number of forward breadth layers bre(u) of vertex u;

Determine whether the reverse breadth layer rebre(u) of vertex u is greater than the reverse breadth layer rebre(v) of vertex v;

Among them, whether the reachability condition is satisfied between the vertex u and the vertex v, including: if bre(v)≤bre(u) and rebre(u)≤rebre(v), determine whether the reachability condition is satisfied between the vertex u and the vertex v .

That is to say, in this embodiment of the application, for any two vertices u and v in a given directed acyclic graph G, bre(v)-bre(u)>0 or rebre(u)-rebre(v )>0, then u->v is unreachable.

In other embodiments of the present application, A113 judges the relationship between the forward and reverse topological layers of vertex u and vertex v, which can be realized through the following steps:

Determine whether the forward topological layer topo(u) of vertex u is smaller than the forward topological layer topo (v) of vertex v;

Determine whether the number of reverse topology layers retopo(u) of vertex u is smaller than the number of reverse topology layers retopo(v) of vertex v;

Wherein, whether the reachability condition is satisfied between the vertex u and the vertex v includes: topo(u)<topo(v) or retopo(u)<retopo(v), and it is determined that the reachability condition is satisfied between the vertex u and the vertex v.

That is to say, in the embodiment of this application, for any two vertices u and v in a given directed acyclic graph G, if topo(u)>=topo(v) or retopo(u)>=retopo(v ), then u->v is unreachable.

In other embodiments of the present application, when the search pattern includes the pattern corresponding to the reachability analysis, A11 executes the adaptation algorithm based on the recursive call method to judge whether the reachability condition is satisfied between the vertex u and the vertex v, including the following steps:

Determine whether the vertex u _i and vertex v _i input by the recursive call are equal; among them, vertex u _i belongs to the set of child nodes of vertex u and vertex u, and vertex v _i belongs to the set of parent nodes of v and v;

Here, if u _i =v _i , then vertex u and vertex v are reachable, and the recursion terminates, where u _i belongs to vertex u and the child node set of vertex u, and v _i belongs to vertex v and the parent node set of vertex v.

However, if the vertex u _i is not equal to the vertex v _i , judge whether the reachability condition between the vertex u and the vertex v is satisfied based on the recursive calling method.

In the process of judging whether the reachability conditions are met, there are the following three scenarios, all of which are confirmed to be unreachable:

Determine the interval label of vertex v

and

Whether it is within the interval label of vertex u, if not, it is unreachable, stop traversing.

Determine the relationship between the forward and reverse breadth layers of vertices v and u. If bre(v)-bre(u)>0 or rebre(u)-rebre(v)>0, it is unreachable and the traversal is stopped.

Determine the relationship between the forward and reverse topological layers of vertices v and u. If topo(u)≥topo(v) or retopo(u)≥retopo(v), it is unreachable and the traversal is stopped.

Further, in the case of the pattern corresponding to the reachability analysis, a two-way search method is used to perform the search between the out-degree outDeg(u,G) of vertex u and the in-degree inDeg(v,G) of vertex v in A12 The relationship executes the search and obtains the search results, including the following steps:

If outDeg(u,G)≤inDeg(v,G), perform a recursive call on each vertex u _i and vertex v in the set of adjacent child nodes of vertex u to obtain the search result;

If outDeg(u,G)>inDeg(v,G), perform a recursive call on each vertex v _i in all parent node sets of vertex u and vertex v, and obtain the search result.

In practical applications, when the search pattern includes the pattern corresponding to the reachability analysis, the reachability algorithm is implemented, and the offline calculation job is updated every day. Features of the algorithm: use the spark engine to calculate the interval labels of all vertices of the graph model

and

And the number of forward breadth layers bre(v), reverse breadth layers rebre(v), forward topology layers topo(v), and reverse topology layers retopo(v) of all labels; then upsert all vertex features to elasticsarch indexing.

The Elasticsearch vertex feature index structure is:

Using online OLAP analysis, execute the recursive call of the bidirectional search and pruning algorithm, decompose the operation of the algorithm on the graph into basic atomic semantic layer operations, and call the nebulagraph client to perform graph semantic atomic query; the algorithm is based on interval labels, breadth layers, and topology When the number of layers is pruned and the two-way search direction is selected according to the degree of the vertex, only one search is performed on elasticsearch, and all the eigenvalues in the above table of the target vertex are searched out and calculated. This application greatly reduces the delay of OLAP analysis of large-scale graph data sets by integrating the two-way pruning search algorithm. In the implementation process of the adaptation algorithm, all the operations on the graph are dismantled into atomic semantic operations of the graph, and through daily offline batch feature engineering calculations, the features of each dimension are indexed into the high-performance distributed search engine, which ensures the realization of the algorithm efficient and stable.

The steps of the following path search algorithm between two points are the same as those of the reachability algorithm, except that the complete link needs to be recorded through the linked list during the traversal process, that is, input vertex u and vertex v, and search for all links between vertex u and vertex v road.

In other embodiments of the present application, the search mode includes a mode corresponding to path search between two points, and A11 performs an adaptation algorithm based on a recursive calling method to judge whether the reachability condition between the vertex u and the vertex v is met, including the following steps:

Create linked list v1 and linked list v2; among them, linked list v1 adds vertex u as the head node, and linked list v2 adds vertex v as the head node;

If u _i =v _i , then vertex u and vertex v are reachable, and the recursion terminates, where u _i belongs to vertex u and the child node set of vertex u, and v _i belongs to vertex v and the parent node set of vertex v. Reverse the linked list of link2, remove the tail node from the linked list of link1, splice with the linked list of link2, and return the set of linked lists after splicing.

Further, in the case of searching for the corresponding mode between two points, a two-way search method is used to execute the relationship between the out-degree outDeg(u,G) of vertex u based on vertex u and the in-degree inDeg(v,G) of vertex v in A12. Execute the search for the relationship and get the search results, including the following steps:

If outDeg(u,G)≤inDeg(v,G), perform a recursive call on each vertex u _i and vertex v in the adjacent child node set of vertex u, clone the current out-degree vertex list and add u _i as the tail node , get the search results of all links between vertex u and vertex v;

If outDeg(u,G)>inDeg(v,G), perform a recursive call on each vertex v _i in all parent node sets of vertex u and vertex v, clone the current in-degree vertex list and add v _i as the tail node, Get the search results of all links between vertex u and vertex v.

In other embodiments of the present application, the search mode includes the corresponding mode of K-step search, and breadth-first search is adopted. When breadth-first search is used to search the adjacent hierarchical nodes of a vertex, it can be searched out at one time using the batch search mode; combined with the search conditions, each search When an adjacent fixed point comes out, use the search condition to prune, which can greatly speed up the search and get the search results quickly.

The following continues to describe the exemplary structure in which the graph search device 154 provided by the embodiment of the present application is implemented as a software module. In some embodiments, as shown in FIG. 1 , the software module stored in the graph search device 154 of the memory 150 may be a server The graph search apparatus in 100, comprising:

Display module 1541, used to display the search entry of the front-end page;

The receiving module 1542 is used to receive the search vertex input at the search entry and the selected search mode; wherein, the search vertex is a point in a directed acyclic graph;

The processing module 1543 is used to determine the adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm to search based on the search vertex, and obtain the search result;

The display module 1541 is configured to display search results through a visual plug-in.

In some embodiments, the search vertex includes vertex u and vertex v, and the processing module 1543 is used to execute the adaptation algorithm to judge whether the reachability condition between vertex u and vertex v is satisfied based on the recursive calling method; Satisfy the reachability condition, perform a search based on the relationship between the out-degree outDeg(u,G) of vertex u and the in-degree inDeg(v,G) of vertex v, and obtain the search results, where G represents a directed acyclic graph .

In some embodiments, the processing module 1543 is used to judge the interval label relationship between vertex u and vertex v; Topological layer number and reverse topological layer number relationship; based on interval label relationship, forward breadth layer number and reverse breadth layer number relationship, and at least one of forward topological layer number and reverse topological layer number relationship, determine vertex u and vertex Whether the reachability condition is satisfied between v.

In some embodiments, the processing module 1543 is used to determine the first interval label of the vertex v

Is it the first interval label of vertex u

within , and the second interval label of vertex v

Whether in the second interval label of vertex u

Within; where, the first interval label of each vertex is the forward direction of the child vertex of each vertex when the pre-order traversal is performed on each vertex;

The label obtained during the traversal, the second interval label of each vertex is the label obtained by traversing the reverse order of the child vertex during the second preorder traversal of each vertex; among them, whether the relationship between vertex u and vertex v satisfies Accessible conditions, including: if

exist

within and

exist

In some embodiments, the processing module 1543 is used to determine whether the difference between the forward breadth level bre(v) of the vertex v minus the forward breadth level bre(u) of the vertex u is greater than 0;

Determine whether the difference between the number of reverse breadth layers rebre(u) of vertex u minus the number of reverse breadth layers rebre(v) of vertex v is greater than 0;

Among them, whether the reachability condition is satisfied between vertex u and vertex v, including: if bre(v)-bre(u)<0 and rebre(u)-rebre(v)<0, determine the distance between vertex u and vertex v The reachability condition is met.

In some embodiments, the processing module 1543 is configured to determine whether the forward topology layer topo(u) of the vertex u is smaller than the forward topology layer topo(v) of the vertex v;

In some embodiments, the search mode includes a mode corresponding to the reachability analysis, and the processing module 1543 is used to determine whether the vertex u _i and the vertex v _i input by the recursive call are equal; wherein, the vertex u _i belongs to the vertex u and the vertex u A set of child nodes, vertex v _i belongs to the set of parent nodes of v and v; if vertex u _i is not equal to vertex v _i , judge whether the reachability condition between vertex u and vertex v is satisfied based on the recursive calling method.

In some embodiments, the processing module 1543 is configured to perform a recursive call on each vertex u _i and vertex v in the set of adjacent child nodes of vertex u if outDeg(u, G)≤inDeg(v, G), to obtain search results;

In some embodiments, the search mode includes a mode corresponding to path search between two points, and the processing module 15 to 43 is used to create a linked list v1 and a linked list v2; wherein, the linked list v1 adds a vertex u as a head node, and the linked list v2 adds a vertex v as a head node Node;

If the vertex u _i is not equal to the vertex v _i , judge whether the reachability condition between the vertex u and the vertex v is satisfied based on the recursive calling method.

In some embodiments, the processing module 1543 is configured to perform recursive calls on each vertex u _i and vertex v in the set of adjacent child nodes of vertex u if outDeg(u, G)≤inDeg(v, G), and clone The current out-degree vertex list and add u _i as the tail node to get the search results of all links between vertex u and vertex v;

The graph search device provided by the embodiment of the present application receives the search vertex and the selected search mode inputted in the search entry of the front page; wherein, the search vertex is a point in a directed acyclic graph; and the appropriate search mode corresponding to the search mode is determined. Algorithm matching, and execution of the matching algorithm to search based on the search vertex to obtain the search results; display the search results through the visualization plug-in; solve the problem that the complex search of the graph in the related technology can only be realized through the back-end offline algorithm, without a real-time return and visualization For the problem of graphic display solutions, it supports a complete set of solutions for OLAP complex scene search and visual display of various graph models.

It should be noted that the description of the device in the embodiment of the present application is similar to the description of the above-mentioned method embodiment, and has similar beneficial effects to the method embodiment, so details are not repeated here. For the technical details not disclosed in the device embodiment of this application, please refer to the description of the method embodiment of this application for understanding.

The embodiment of the present application provides a storage medium storing executable instructions, and the executable instruction is stored therein. When the executable instruction is executed by the processor, it will cause the processor to execute the method provided in the embodiment of the present application, for example, as shown in FIG. 2 shows the method.

The storage medium provided by the embodiment of the present application receives the search vertex and the selected search mode input in the search entry of the front page; wherein, the search vertex is a point in a directed acyclic graph; and the matching corresponding to the search mode is determined Algorithm, and execute the adaptation algorithm to search based on the search vertex to obtain the search results; display the search results through the visualization plug-in; solve the problem that the complex search of the graph in the related technology can only be realized through the back-end offline algorithm, there is no real-time return and visual graph For the problem of displaying solutions, it supports a complete set of solutions for OLAP complex scene search and visual display of various graph models.

In some embodiments, the storage medium can be a computer-readable storage medium, for example, a ferroelectric memory (FRAM, Ferromagnetic Random Access Memory), a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read Only Memory), Erasable Programmable Read Only Memory (EPROM, Erasable Programmable Read Only Memory), Electrically Erasable Programmable Read Only Memory (EEPROM, Electrically Erasable Programmable Read Only Memory), flash memory, magnetic surface memory, optical disc, Or memory such as CD-ROM (Compact Disk-Read Only Memory); It can also be various devices including one or any combination of the above-mentioned memories.

In some embodiments, executable instructions may take the form of programs, software, software modules, scripts, or code written in any form of programming language, including compiled or interpreted languages, or declarative or procedural languages, and its Can be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

As an example, executable instructions may, but do not necessarily correspond to files in a file system, may be stored as part of a file that holds other programs or data, for example, in Hyper Text Markup Language (HTML, Hyper Text Markup Language) in one or more scripts in a document, in a single file dedicated to the program in question, or in multiple cooperating files (for example, a file that stores one or more modules, subroutines, or code sections )middle. As an example, executable instructions may be deployed to be executed on one computing device, or on multiple computing devices located at one site, or alternatively, on multiple computing devices distributed across multiple sites and interconnected by a communication network. to execute.

The above descriptions are merely examples of the present application, and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Industrial Applicability

The embodiment of the present application provides a graph search method, device, device and storage medium, by receiving the search vertex and the selected search mode input in the search entry of the front-end page; wherein, the search vertex is a directed acyclic graph point; determine the adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm to search based on the search vertex, and obtain the search result; display the search result through a visualization plug-in. It solves the problem that the complex search of graphs in related technologies can only be realized through the back-end offline algorithm, and there is no real-time return and visual graphic display solution, and supports online analytical processing (OLAP) complex scenarios of ultra-large-scale data sets A complete set of solutions for searching and visual display of various graph models.

Claims

A graph search method, comprising:

Receive a search vertex and a selected search mode input at the search entry of the front-end page; wherein, the search vertex is a point in a directed acyclic graph;

determining an adaptation algorithm corresponding to the search mode, and executing the adaptation algorithm to perform a search based on the search vertex to obtain a search result;

The search results are displayed through a visual plug-in.
The method according to claim 1, wherein the search vertex includes vertex u and vertex v, and performing the adaptation algorithm to search based on the search vertex to obtain a search result includes:

Executing the adaptation algorithm to determine whether the reachability condition between the vertex u and the vertex v is satisfied based on a recursive calling method;

If the reachability condition is satisfied between the vertex u and the vertex v, based on the out-degree outDeg(u, G) of the vertex u and the in-degree inDeg(v, G) of the vertex v A relation performs a search to obtain the search result, where G represents the directed acyclic graph.
The method according to claim 2, wherein said judging whether a reachability condition is satisfied between the vertex u and the vertex v comprises:

judging the interval label relationship between the vertex u and the vertex v;

judging the relationship between the forward breadth layers and the reverse breadth layers of the vertex u and the vertex v;

judging the relationship between the forward and reverse topological layers of the vertex u and the vertex v;

Determine the vertex u and the vertex v based on at least one of the interval label relationship, the forward breadth layer number and reverse breadth layer number relationship, and the forward topology layer number and reverse topology layer number relationship Whether the reachability condition is satisfied.
The method according to claim 3, wherein the judging the interval label relationship between the vertex u and the vertex v includes;

Determine the first interval label of the vertex v
Is the first interval label in the vertex u
within , and the second interval label of the vertex v
is the second interval label of the vertex u
Within; wherein, the first interval label of each vertex is the label obtained when the child vertex of each vertex is traversed forward when the pre-order traversal is performed on each vertex, and the second interval label of each vertex An interval label is the label obtained by traversing the reverse order of the child vertex when the pre-order traversal is performed on each vertex for the second time;

Wherein, whether the reachability condition is satisfied between the vertex u and the vertex v includes: if the
in the
within and as stated
in the
Within, it is determined that the reachability condition is satisfied between the vertex u and the vertex v.
The method according to claim 3 or 4, wherein the judging the relationship between the forward breadth layers and the reverse breadth layers of the vertex u and the vertex v includes:

Judging whether the forward breadth number bre(v) of the vertex v is greater than the forward breadth number bre(u) of the vertex u;

Judging whether the reverse breadth number rebre(u) of the vertex u is greater than the reverse breadth number rebre(v) of the vertex v;

Wherein, whether the reachability condition is satisfied between the vertex u and the vertex v includes: if bre(v)≤bre(u) and rebre(u)≤rebre(v), determine the vertex u and the vertex v The reachability condition is satisfied between vertices v.
The method according to any one of claims 3 to 5, wherein the judging the relationship between the forward topological layer number and the reverse topological layer number of the vertex u and the vertex v includes:

Judging whether the forward topology layer number topo(u) of the vertex u is less than the forward topology layer number topo(v) of the vertex v;

Judging whether the reverse topology layer number retopo(u) of the vertex u is less than the reverse topology layer number retopo(v) of the vertex v;

Wherein, whether the reachability condition is satisfied between the vertex u and the vertex v includes: topo(u)<topo(v) or retopo(u)<retopo(v), and determining the vertex u and the vertex The reachability condition is satisfied between v.
The method according to claim 2, wherein the search mode includes a mode corresponding to reachability analysis, and the execution of the adaptation algorithm judges whether the vertex u and the vertex v meet the requirements based on a recursive calling method. Accessible conditions, including:

Judging whether the vertex u i and vertex v i input by the recursive call are equal; wherein, the vertex u i belongs to the vertex u and the child node set of the vertex u, and the vertex v i belongs to the v and the v set of parent nodes;

If the vertex u i is not equal to the vertex v i , judge whether the reachability condition between the vertex u and the vertex v is satisfied based on a recursive calling manner.
The method according to claim 7, wherein the search is performed based on the relationship between the out-degree outDeg(u, G) of the vertex u and the in-degree inDeg(v, G) of the vertex v, to obtain the The above search results, including:

If outDeg(u, G)≤inDeg(v, G), perform a recursive call on each vertex u i and vertex v in the set of adjacent child nodes of the vertex u to obtain the search result;

If outDeg(u,G)>inDeg(v,G), perform a recursive call on each vertex v i in the set of all parent nodes of the vertex u and vertex v, and obtain the search result.
The method according to claim 2, wherein the search mode includes a mode corresponding to path search between two points, and the execution of the adaptation algorithm judges whether there is a relationship between the vertex u and the vertex v based on a recursive calling method. Accessible conditions are met, including:

Create a linked list v1 and a linked list v2; wherein, the linked list v1 adds the vertex u as a head node, and the linked list v2 adds the vertex v as a head node;

Judging whether the vertex u i and vertex v i input by the recursive call are equal; wherein, the vertex u i belongs to the vertex u and the child node set of the vertex u, and the vertex v i belongs to the v and the v set of parent nodes;

If the vertex u i is not equal to the vertex v i , judge whether the reachability condition between the vertex u and the vertex v is satisfied based on a recursive calling manner.
The method according to claim 9, wherein the search is performed based on the relationship between the out-degree outDeg(u, G) of the vertex u and the in-degree inDeg(v, G) of the vertex v, to obtain the The above search results, including:

If outDeg(u,G)≤inDeg(v,G), perform a recursive call on each vertex u i and vertex v in the set of adjacent child nodes of the vertex u, clone the current out-degree vertex linked list and add u i as Tail node, obtain the search results of all links between the vertex u and the vertex v;

If outDeg(u,G)>inDeg(v,G), perform a recursive call on each vertex v i in the set of all parent nodes of the vertex u and vertex v, clone the current in-degree vertex list and add v i as the tail node, to obtain the search results of all the links between the vertex u and the vertex v.
A graph search device, comprising:

The display module is used to display the search entry of the front-end page;

A receiving module, configured to receive a search vertex input at the search entry and a selected search mode; wherein the search vertex is a point in a directed acyclic graph;

A processing module, configured to determine an adaptation algorithm corresponding to the search mode, and execute the adaptation algorithm to perform a search based on the search vertex to obtain a search result;

The display module is configured to display the search results through a visual plug-in.
A graph search device, comprising:

The memory is used to store executable instructions; the processor is used to implement the method according to any one of claims 1 to 10 when executing the executable instructions stored in the memory.
A storage medium storing executable instructions for causing a processor to implement the method according to any one of claims 1 to 10.