CN112100450A - Graph calculation data segmentation method, terminal device and storage medium - Google Patents
Graph calculation data segmentation method, terminal device and storage medium Download PDFInfo
- Publication number
- CN112100450A CN112100450A CN202010927101.9A CN202010927101A CN112100450A CN 112100450 A CN112100450 A CN 112100450A CN 202010927101 A CN202010927101 A CN 202010927101A CN 112100450 A CN112100450 A CN 112100450A
- Authority
- CN
- China
- Prior art keywords
- node
- graph
- label
- subgraph
- nodes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a graph computation data segmentation method, a terminal device and a storage medium, wherein the method comprises the following steps: s1: dividing an original graph community into weighted graphs by improving a label propagation algorithm; s2: dividing the weighted graph into a plurality of subgraphs, and adjusting nodes in different subgraphs according to the gain value of each node in each subgraph to ensure that the gain value of each node in each subgraph is less than or equal to 0 or the adjustment times is greater than a time threshold; s3: and restoring the subgraph with the weight graph into the data scale corresponding to the original graph. The invention can process the division of large-scale graph data, so that the divided data blocks are more balanced, the coupling degree between the data blocks is lower, the data communication between all working nodes in a parallel computing platform is effectively reduced, the task operation efficiency of a graph processing framework is greatly improved, and the task throughput is increased.
Description
Technical Field
The present invention relates to the field of graph computation, and in particular, to a graph computation data segmentation method, a terminal device, and a storage medium.
Background
After Google proposed the concept of "Knowledge Graph" (Knowledge Graph) in 2012, the Knowledge Graph technology rapidly developed, and Graph computation attracted extensive attention, and with the coming of the internet and big data era in recent years, the demand for rapidly processing and analyzing mass data becomes urgent, such as pagerank computation of mass web data, social relationship analysis in social networks, network literature relationship analysis, and the like. Since the graph computation task is unique in itself: the graph data has strong dependency, so that the distributed processing mode needs to exchange data among various machines frequently; and the graph mining algorithm flow itself is composed of multiple iterative processes. Therefore, mainstream big data platforms are admittedly not suitable for processing graph mining tasks, and big data analysis platforms based on graph calculation become a new method of graph mining and related machine learning algorithms. The MapReduce series based on a MapReduce model, the Pregel series based on a BSP programming model, the GraphLap based on a GAS model, the PowerGraph series and the like are mainly available. Under a distributed environment, a graph partitioning algorithm in a graph computation processing framework directly influences the processing efficiency of the framework, the graph partitioning method provided by the existing distributed graph computation framework does not consider the principle of data locality, and most of the frameworks adopt a simple hash algorithm, so that load balancing can be satisfied by nature, but the hash mode cannot be divided according to the connectivity of the graph, so that the message transmission overhead between the super steps is large. In the traditional graph segmentation algorithm, such as a Kemighan-Lin algorithm, a spectrum analysis method and the like, the complexity of computing time or the complexity of computing space is high, parallelization is difficult to perform, the completion time of a task is influenced to a great extent, and the computing performance and the service quality of the whole platform are further influenced.
Disclosure of Invention
The invention provides a graph computation data segmentation method, terminal equipment and a storage medium, and aims to solve the problems that after the structure of a graph is segmented, the data scale deviation in each sub-graph is large, namely the phenomenon of 'barrel effect' caused by uneven load, the segmented result cannot guarantee strong connectivity inside each sub-graph and low coupling between the sub-graphs, and the segmentation algorithm is too high in time complexity and space complexity and difficult to parallelize.
The specific scheme is as follows:
a graph computation data segmentation method comprises the following steps:
s1: dividing an original graph community into weighted graphs by improving a label propagation algorithm;
the improved label propagation algorithm is to calculate the label confidence weight of each node transmitted to each neighbor node of the node, update the label of each node to the label of the neighbor node corresponding to the maximum label confidence weight in the neighbor node of the node, and then gather the vertexes with the same label value;
s2: dividing the weighted graph into a plurality of subgraphs, and adjusting nodes in different subgraphs according to the gain value of each node in each subgraph to ensure that the gain value of each node in each subgraph is less than or equal to 0 or the adjustment times is greater than a time threshold;
s3: and restoring the subgraph with the weight graph into the data scale corresponding to the original graph.
Further, the calculation formula of the label confidence weight of the node in the improved label propagation algorithm is as follows:
wherein: w is aijRepresenting the label confidence weight, WL, conducted from the current node i to the neighbor node jiLabel confidence weight, WR, representing current node ii-＞jRepresenting the influence of the relationship between the current node i and the neighboring node j,
and the label confidence weight and the cumulative sum of the relation influence of the current node i and all the neighbor nodes thereof are represented, k represents the node serial number, and l represents the total number of all the neighbor nodes of the current node i.
Further, the number of nodes for switching in different subgraphs is equal.
Further, the gain value of a node is the difference value between the sum of the association weights of the node and other nodes which are not in the same subgraph as the node and the sum of the association weights of the node and other nodes which are in the same subgraph as the node.
A graph computation data segmentation terminal device comprises a processor, a memory and a computer program stored in the memory and operable on the processor, wherein the processor executes the computer program to implement the steps of the method of the embodiment of the present invention.
A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to an embodiment of the invention as described above.
By adopting the technical scheme, the method can process the segmentation of large-scale graph data, so that the segmented data blocks are more balanced, the coupling degree between the data blocks is lower, the data communication between all working nodes in a parallel computing platform is effectively reduced, the task operation efficiency of a graph processing framework is greatly improved, and the task throughput is increased.
Drawings
Fig. 1 is a flowchart illustrating a first embodiment of the present invention.
Fig. 2 is a schematic diagram showing an original diagram in this embodiment.
Fig. 3 is a diagram illustrating the effect of community division in this embodiment.
Fig. 4 shows the effect diagram with the weight map in this embodiment.
Fig. 5 is a schematic diagram of the initially divided subgraph in this embodiment.
Fig. 6 is a schematic diagram of the sub-graph in this embodiment after adjustment.
Fig. 7 is a diagram showing the effect of the restoration to the original diagram in this embodiment.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures.
The invention will now be further described with reference to the accompanying drawings and detailed description.
The first embodiment is as follows:
the embodiment of the invention provides a graph calculation data segmentation method, as shown in fig. 1, the method comprises the following steps:
s1: and dividing the original graph community into weighted graphs by improving a label propagation algorithm.
The improved label propagation algorithm carries out community division on the original graph through the improved label propagation algorithm, so that the scale of the original graph can be gradually reduced, and the divided graph presents a weighted contracted graph with a small enough number of top points.
In a traditional Label Propagation Algorithm (LPA), a label with the maximum number of labels in neighbor nodes is selected as a label of a current node, and when the number of the labels with the maximum number is multiple, the label is randomly selected from the multiple labels with the maximum number as the label of the current node, and the selection mode may cause a weak relationship or a label with a small influence to be a decisive factor, and finally causes a label propagation result to show a large uncertainty.
In the embodiment, an improved label propagation algorithm is adopted, label confidence weights and relation influences of fusion nodes of the improved label propagation algorithm are conducted on labels according to the joint importance degree of the improved label propagation algorithm, so that the nodes and the relations with higher importance have representative significance, and the uncertainty and the random countercurrent phenomenon of the label propagation algorithm can be effectively reduced.
The improved label propagation algorithm is to calculate the label confidence weight of each node transmitted to each neighbor node of the node, update the label of each node to the label of the neighbor node corresponding to the maximum label confidence weight in the neighbor node of the node, and then gather the vertexes with the same label value to obtain the weighted graph. The nodes in the weighted graph are aggregation nodes, the relationships are aggregation relationships, and the weights of the aggregation relationships are the sum of the weights of all the relationships connecting the two communities after the communities are divided.
The calculation formula of the label confidence weight of the node is as follows:
wherein: w is aijRepresenting the label confidence weight, WL, conducted from the current node i to the neighbor node jiLabel confidence weight, WR, representing current node ii-＞jRepresenting the influence of the relationship between the current node i and the neighboring node j,
and the label confidence weight and the cumulative sum of the relation influence of the current node i and all the neighbor nodes thereof are represented, k represents the node serial number, and l represents the total number of all the neighbor nodes of the current node i.
Therefore, the weighted graph obtained by the improved label propagation algorithm in this embodiment can ensure that the aggregated nodes and edges have a high degree of cohesion, and the obtained graph presents a state in which the number of vertices is sufficiently small, and also retains the important structure of the original graph as much as possible, so that the result of compression does not cause invalid segmentation at the next stage and increase the complexity of segmentation.
At the beginning of this step, the data of the original graph, as shown in FIG. 2, needs to be imported into the graph database. The subsequent community division effect graph is shown in fig. 3. Fig. 4 shows an effect diagram of a weighted graph formed by aggregating and compressing divided communities, where the weight of the aggregation relation in the weighted graph is the sum of the weights of all the relations connecting the two communities after the community division.
S2: and dividing the weighted graph into a plurality of subgraphs, and adjusting the nodes in different subgraphs according to the gain value of each node in each subgraph, so that the gain value of each node in each subgraph is less than or equal to 0 or the adjustment times are greater than a time threshold, and the subgraph has the best division effect, namely low coupling among different subgraphs and balanced data.
The size of the time threshold can be set by a person skilled in the art according to experience, and is not limited herein.
When the weighted graph is initially divided into a plurality of subgraphs, a random average division method is adopted, and the method is not limited herein. As shown in fig. 4, since the number of nodes is small, the weighted graph is divided into two subgraphs, i.e., P1 ═ a, B, C } and P2 ═ D, E, F }, as shown in fig. 5.
And after the initial segmentation of the subgraphs is completed, respectively calculating the gain value of each node in each subgraph. When the gain value is larger than 0, the segmentation effect of moving the representative node from the subgraph where the representative node is located to another subgraph is better; when the gain value is equal to 0, the segmentation effect representing moving and not moving is the same; when the gain value is less than 0, the segmentation effect representing no movement is better. Therefore, when the gain value is greater than 0, the node is exchanged with the nodes with the gain value greater than 0 in other subgraphs. In this embodiment, there are only two subgraphs P1 and P2, so the gain value of each node in P1 and P2 is calculated, and the gain value of each node in subgraph P1 is calculated as: p (a) ═ 3, p (b) ═ -6, p (c) ═ 0; the gain values of the nodes in the sub-graph P2 are: p (d) ═ 2, p (e) ═ 8, and p (f) ═ 3. The node A and the node F can be exchanged according to the node gain values of the two subgraphs, and the subgraph has a good segmentation effect, so that the node A in the subgraph P1 is exchanged with the node F in the subgraph P2, as shown in FIG. 6.
In this embodiment, during switching, the same number of nodes are selected from different subgraphs for switching, so as to ensure that the number of each subgraph can still be balanced after switching the nodes.
The gain value of the node is the difference value between the sum of the association weights of the node and other nodes which are not in the same subgraph as the node and the sum of the association weights of the node and other nodes which are in the same subgraph as the node.
As shown in fig. 5, in the calculation of the gain value P (a) of the node a, the node a is associated with the nodes D and F in the sub-graph P2, the sum of the associated weights thereof is (2+3) to 5, the node a is associated with the node B of the sub-graph P1, the associated weight thereof is 2, and thus the gain value P (a) of the node a is 5-2 to 3.
The gain values of other nodes calculated according to the method are respectively as follows:
P(B)＝1-(2+5)＝-6
P(C)＝(4+1)-5＝0
P(D)＝(1+2)-5＝-2
P(E)＝1-(4+5)＝-8
P(F)＝(4+3)-4＝3
by the subgraph segmentation and adjustment of the step, the segmented subgraph has stronger connectivity, and the subgraph have lower coupling; meanwhile, the data of each sub-graph is ensured to be more balanced, and the phenomenon of data inclination is avoided. The segmentation method adopted by the application is that on the basis that the relationship weight influences the segmentation effect, the segmentation method is initialized and segmented randomly, then judgment is carried out by calculating the gain value of each node, different sub-graph nodes with the same number are selected for exchange, and finally the final result of better segmentation operation is achieved by multiple exchanges.
S3: and restoring the subgraph with the weight graph to the data scale corresponding to the original graph as shown in fig. 7, thereby completing the whole segmentation process.
The embodiment I of the invention can process the segmentation of large-scale graph data, so that the segmented data blocks are more balanced, the coupling degree between the data blocks is lower, the data communication between all working nodes in a parallel computing platform is effectively reduced, the task operation efficiency of a graph processing framework is greatly improved, and the task throughput is increased.
Example two:
the present invention further provides a graph computation data segmentation terminal device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps in the foregoing method embodiments of the first embodiment of the present invention when executing the computer program.
Further, as an executable solution, the graph computation data segmentation terminal device may be a computing device such as a vehicle-mounted computer and a cloud server. The graph computation data segmentation terminal device may include, but is not limited to, a processor, a memory. It is understood by those skilled in the art that the above-mentioned constituent structure of the graph computation data division terminal device is only an example of the graph computation data division terminal device, and does not constitute a limitation on the graph computation data division terminal device, and may include more or less components than the above, or combine some components, or different components, for example, the graph computation data division terminal device may further include an input output device, a network access device, a bus, and the like, which is not limited by the embodiment of the present invention.
Further, as an executable solution, the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, and the like. The general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being a control center of the graph calculation data division terminal device, and various interfaces and lines are used to connect the respective parts of the entire graph calculation data division terminal device.
The memory may be used to store the computer programs and/or modules, and the processor may implement various functions of the graph computation data segmentation terminal device by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. The memory can mainly comprise a program storage area and a data storage area, wherein the program storage area can store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the mobile phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The invention also provides a computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned method of an embodiment of the invention.
The integrated module/unit of the graph calculation data division terminal device may be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), software distribution medium, and the like.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (7)
1. A graph computation data segmentation method is characterized by comprising the following steps:
s1: dividing an original graph community into weighted graphs by improving a label propagation algorithm;
the improved label propagation algorithm is to calculate the label confidence weight of each node transmitted to each neighbor node of the node, update the label of each node to the label of the neighbor node corresponding to the maximum label confidence weight in the neighbor node of the node, and then gather the vertexes with the same label value;
s2: dividing the weighted graph into a plurality of subgraphs, and adjusting nodes in different subgraphs according to the gain value of each node in each subgraph to ensure that the gain value of each node in each subgraph is less than or equal to 0 or the adjustment times is greater than a time threshold;
s3: and restoring the subgraph with the weight graph into the data scale corresponding to the original graph.
2. The graph computation data segmentation method according to claim 1, characterized in that: the calculation formula of the label confidence weight of the node in the improved label propagation algorithm is as follows:
wherein: w is aijRepresenting the label confidence weight, WL, conducted from the current node i to the neighbor node jiLabel confidence weight, WR, representing current node ii-＞jRepresenting the influence of the relationship between the current node i and the neighboring node j,
and the label confidence weight and the cumulative sum of the relation influence of the current node i and all the neighbor nodes thereof are represented, k represents the node serial number, and l represents the total number of all the neighbor nodes of the current node i.
3. The graph computation data segmentation method according to claim 1, characterized in that: the method for adjusting the nodes in different subgraphs in step S2 includes: and respectively calculating the gain value of each node in each subgraph, and exchanging the node with the gain value larger than O in other subgraphs when the gain value is larger than O.
4. The graph computation data segmentation method according to claim 3, characterized in that: the number of nodes exchanged is equal for different subgraphs.
5. The graph computation data segmentation method according to claim 1, characterized in that: the gain value of the node is the difference value between the sum of the association weights of the node and other nodes which are not in the same subgraph as the node and the sum of the association weights of the node and other nodes which are in the same subgraph as the node.
6. A graph computation data division terminal device, characterized by: comprising a processor, a memory and a computer program stored in the memory and running on the processor, the processor implementing the steps of the method according to any of claims 1 to 5 when executing the computer program.
7. A computer-readable storage medium storing a computer program, characterized in that: the computer program when executed by a processor implementing the steps of the method as claimed in any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010927101.9A CN112100450A (en) | 2020-09-07 | 2020-09-07 | Graph calculation data segmentation method, terminal device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010927101.9A CN112100450A (en) | 2020-09-07 | 2020-09-07 | Graph calculation data segmentation method, terminal device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112100450A true CN112100450A (en) | 2020-12-18 |
Family
ID=73757566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010927101.9A Pending CN112100450A (en) | 2020-09-07 | 2020-09-07 | Graph calculation data segmentation method, terminal device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112100450A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966054A (en) * | 2021-02-07 | 2021-06-15 | 撼地数智(重庆)科技有限公司 | Enterprise graph node relation-based ethnic group division method and computer equipment |
| CN112990332A (en) * | 2021-03-26 | 2021-06-18 | 杭州海康威视数字技术股份有限公司 | Sub-graph scale prediction and distributed training method and device and electronic equipment |
| CN113869904A (en) * | 2021-08-16 | 2021-12-31 | 工银科技有限公司 | Suspicious data identification method, device, electronic equipment, medium and computer program |
| CN114416913A (en) * | 2022-03-28 | 2022-04-29 | 支付宝(杭州)信息技术有限公司 | Method and device for data slicing of knowledge graph |
-
2020
- 2020-09-07 CN CN202010927101.9A patent/CN112100450A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966054A (en) * | 2021-02-07 | 2021-06-15 | 撼地数智(重庆)科技有限公司 | Enterprise graph node relation-based ethnic group division method and computer equipment |
| CN112990332A (en) * | 2021-03-26 | 2021-06-18 | 杭州海康威视数字技术股份有限公司 | Sub-graph scale prediction and distributed training method and device and electronic equipment |
| CN112990332B (en) * | 2021-03-26 | 2023-06-02 | 杭州海康威视数字技术股份有限公司 | Sub-graph scale prediction and distributed training method and device and electronic equipment |
| CN113869904A (en) * | 2021-08-16 | 2021-12-31 | 工银科技有限公司 | Suspicious data identification method, device, electronic equipment, medium and computer program |
| CN114416913A (en) * | 2022-03-28 | 2022-04-29 | 支付宝(杭州)信息技术有限公司 | Method and device for data slicing of knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112100450A (en) | Graph calculation data segmentation method, terminal device and storage medium | |
| CN109033340B (en) | Spark platform-based point cloud K neighborhood searching method and device | |
| CN105069039A (en) | Overlapping community parallel discovery method of memory iteration on basis of spark platform | |
| CN110765320B (en) | Data processing method, device, storage medium and computer equipment | |
| CN103838803A (en) | Social network community discovery method based on node Jaccard similarity | |
| CN112883241B (en) | Supercomputer benchmark test acceleration method based on connected component generation optimization | |
| CN110428386B (en) | Map grid merging method and device, storage medium and electronic device | |
| CN113515672A (en) | Data processing method and device, computer readable medium and electronic equipment | |
| CN110853120B (en) | Network layout method, system and medium based on segmentation drawing method | |
| CN111159577A (en) | Community division method and device, storage medium and electronic device | |
| CN113127697B (en) | Method and system for optimizing graph layout, electronic device and readable storage medium | |
| CN108880872B (en) | Method and device for decomposing topological structure of Internet test bed | |
| CN104951442A (en) | Method and device for determining result vector | |
| EP4012573A1 (en) | Graph reconstruction method and apparatus | |
| CN111444309B (en) | System for learning graph | |
| CN115865607A (en) | Distributed training computing node management method and related device | |
| CN115544307A (en) | Directed graph data feature extraction and expression method and system based on incidence matrix | |
| CN110309367B (en) | Information classification method, information processing method and device | |
| CN116681857A (en) | Space collision detection method and device and electronic equipment | |
| CN113722554A (en) | Data classification method and device and computing equipment | |
| CN106991195B (en) | Distributed subgraph enumeration method | |
| CN113905066A (en) | Networking method of Internet of things, networking device of Internet of things and electronic equipment | |
| CN113761293A (en) | Graph data strong-connectivity component mining method, device, equipment and storage medium | |
| CN113705598A (en) | Data classification method and device and electronic equipment | |
| CN110704693A (en) | Distributed graph calculation system and distributed graph calculation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |