CN111177188A - Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge - Google Patents
Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge Download PDFInfo
- Publication number
- CN111177188A CN111177188A CN201911405552.XA CN201911405552A CN111177188A CN 111177188 A CN111177188 A CN 111177188A CN 201911405552 A CN201911405552 A CN 201911405552A CN 111177188 A CN111177188 A CN 111177188A
- Authority
- CN
- China
- Prior art keywords
- graph
- time
- aggregation
- edge
- time sequence
- 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 OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2452—Query translation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2468—Fuzzy queries
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Fuzzy Systems (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Automation & Control Theory (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a rapid mass time sequence data processing method based on a polymerization edge and a time sequence polymerization edge, which realizes rapid real-time processing of incidence relation based on a graph data structure in a mass data mode. The invention provides an innovative data structure of 'aggregation edge' and 'time sequence aggregation edge' on the basis of incremental flow calculation based on a time window, and is suitable for data modeling of a real-time dynamic graph. The invention also introduces a sequence diagram query language, increases the description semantics of the sequence diagram information, supports the basic query based on points, edges and attributes, and can realize the graph query of the index calculation result in a certain time window by a user, including graph matching and graph filtering. The method is particularly suitable for the fields of marketing based on mass data mining, real-time wind control and the like, and has good timeliness control and high extensibility.
Description
Technical Field
The invention provides a rapid massive time sequence data processing method based on a polymerization edge and a time sequence polymerization edge. The method comprises modeling based on a graph data structure, dynamically building a graph, aggregating edges (Aggregated edges) and Time-series Aggregated edges (Time-series Aggregated edges), and corresponding methods such as graph association relation query and pattern matching on the basis of the Aggregated edges. The method is mainly suitable for the fields of finance, electric power, traffic, internet and the like, and is used for analyzing the association relation existing in the data in real time.
Background
In the fields of financial real-time wind control, accurate marketing and the like, the calculation problems of related incidence relation variables and the like of a certain user, such as 'a consumed merchant in the past 24 hours', 'an opposite party with the accumulated transfer amount of more than 100 ten thousand yuan in the past 180 days', and the like are often involved. And the method also relates to a pattern matching problem based on the association relation, such as that a certain user's account is more than 100 in the past 1 week.
In solving the incidence relation query, a simple query can be queried by table concatenation (join) based on a table structure of a database. In a complex business scenario, when there are many different types of entities and there are many different types of relationships, because the database table is essentially based on a binary relationship, the splicing operation based on the database table is complex, and the query performance may not meet the business requirement. The above technique generally selects a structure for modeling incidence relations on complex business scenarios into graphs (Graph). A graph is a data structure consisting of nodes (Vertex) and edges (Edge). A Graph Database (Graph Database) is a Database that uses Graph data structures for semantic queries, which use nodes, edges, and attributes to represent and store data. When modeling a business scenario, a modeling manner using an attribute map model is generally selected. In the attribute graph, the node representations are entities and the edge representations are relationships. A node or edge may have zero, one, or multiple attributes, and the attribute key of an entity is unique. For example, in a business scenario of a transaction, the attribute graph model takes a user as a node and takes a transaction as an edge. The attributes on the edge may record the details of the transaction (transaction amount, transaction location, etc.). If multiple transactions occur between two users, multiple edges may be established to indicate the relationship.
There are several main defects in the technical scheme of using a general graph database product attribute graph model for modeling to calculate the incidence relation, which are respectively:
1. under the condition of large data volume of the query result set, the response time is long, and the requirement of business on real-time response time cannot be met. Under a real business scene, when an object of an association relation is generated due to transaction in the past 30 days, the query calculation of the association relation cannot be completed on a cluster single node due to huge transaction flow in the past 30 days.
2. The graph database product converts all data information into points, edges and attributes and places the points, edges and attributes into the graph database, so that the method has the advantages that the transaction information in the graph database is complete and convenient to display; but the disadvantage is that the data is too comprehensive, so that a part of time is needed for screening out the data to be calculated during calculation.
3. A larger scale computing device is required. Because the processing mode needs intensive computation based on a graph data structure, a large-scale cluster and a graph-based computation framework middleware are generally required to be built to solve the problem. This requires a relatively large hardware cost and a maintenance cost of the middleware.
4. Current graph database products lack the ability to aggregate based on time series. Under the service scene of inquiring based on time sequence aggregation, aggregation calculation is carried out simultaneously in the inquiring process, and the inquiring delay is caused. Many graph database products cannot even generate query results in real time under the condition of massive data.
Disclosure of Invention
Aiming at the problems of processing mass time sequence data by a current graph database or a graph calculation middleware product, the invention provides a rapid mass time sequence data association relation processing method based on an aggregation edge and a time sequence aggregation edge, and the rapid real-time processing of the association relation based on a graph data structure is realized in a mass data mode.
The invention provides an innovative data structure of 'aggregation edge' and 'time sequence aggregation edge' on the basis of incremental flow calculation based on a time window, and is suitable for data modeling of a real-time dynamic graph. The invention also introduces a sequence diagram query language, increases the description semantics of the sequence diagram information, supports the basic query based on points, edges and attributes, and can realize the graph query of the index calculation result in a certain time window by a user, including graph matching and graph filtering.
The purpose of the invention is realized by the following technical scheme: a rapid massive time sequence data processing method based on an aggregation edge and a time sequence aggregation edge comprises the following steps:
performing polymerization calculation on the time sequence data based on a time window to obtain a polymerization result;
generating a polymerization edge and a time sequence polymerization edge by using the polymerization result, and storing data in a memory according to a data structure of the polymerization edge;
establishing a graph relationship by the aggregation side or the time sequence aggregation side according to the mode of the attribute graph;
and for the generated graph relation, performing quick query based on the aggregation index.
Further, the generating of the aggregated edge comprises: in the process of establishing the associated relationship graph, according to the attribute fields which are selected and defined in advance on the service, the aggregation calculation is carried out in advance, and the result after the aggregation calculation is formed on the edge attribute.
Further, the generating of the time-series aggregation edge comprises: cutting continuous time according to a certain time unit to form a series of time windows with fixed lengths; all the time sequence data are distributed into corresponding time windows according to the value of the appointed time attribute field; and aggregating the data in the time window according to an aggregation algorithm required by the service to obtain an aggregation value corresponding to the time window.
Further, different calculation indexes adopt different aggregation algorithms according to different calculation service contents, such as one or more of counting, summing, averaging, maximum, minimum, variance, standard deviation, collecting and de-duplication collecting; different calculation indexes may be assigned different time window lengths according to their business implications.
Furthermore, data of the aggregation edges are queried through a graph association relation query language, the graph association relation query language increases description semantics of time sequence information, and query based on points, edges and attributes can be achieved.
Further, the user can perform graph query on the index calculation result in a certain time window, wherein the graph query comprises graph matching and graph filtering.
Further, the graph association relation query language supports predicate filtering semantics and fuzzy matching based on indefinite step number edges.
Further, the graph matching specifically includes: given a starting point and a graph pattern to be matched, entity objects satisfying the matching pattern are returned.
Further, the graph filtering specifically comprises: a specified subset of results is found based on the filter criteria on the basis of the graph matching.
Further, in the graph filtering, the filtering condition may specify a time window, and the specified time window may be different from a time window used when the graph is matched.
The invention has the beneficial effects that: the technology of 'aggregation side' and 'time sequence aggregation side' based on the time window aggregation calculation result provided by the invention is very suitable for the fields of marketing, real-time wind control and the like based on massive data mining, and the advantages brought by the technology are self-evident and summarized, and mainly comprise:
1) and (4) good timeliness control. In the calculation process of the incidence relation, traversal based on a graph structure is easily involved. Due to the result of the pre-aggregation calculation, the calculation amount in the graph traversal process is greatly reduced. By aggregating the results, the size of the space in the graph search may be reduced.
2) High scalability. Under the condition of improving the calculation variables, the service scale and the like, the calculation capacity can be improved by simply adding the calculation equipment and the distributed storage memory, so that the controllable time delay of the complex logic calculation is ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic diagram comparing a simple side and a polymeric side in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a time-series aggregation edge implementation in a simple service scenario;
fig. 3 is a schematic diagram of implementation of a time-series aggregation edge in a complex service scenario.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
FIG. 1 shows the process of establishing a graph relationship structure according to the present invention, in contrast to the process of establishing a graph of a general graph database. The innovation point of the invention is that the concept of the innovative 'aggregation edge' is embodied in the generation process of the graph data structure.
The traditional graph structure is a processing mode of 'simple edge' commonly used in the process of drawing. The concept of 'aggregate edge' proposed by the invention is that in the process of drawing construction, aggregate calculation can be carried out in advance according to attribute fields which are selected and defined in advance on business. Such as: in the process of establishing a graph of the incidence relation graph of the transaction transfer, the invention does not simply record the relation of transfer details of two opponents. According to the index requirement of business query, for example, the transfer times are calculated with emphasis on variable calculation, the transfer times can be calculated in advance in an aggregation mode, and an aggregation edge of the result after the aggregation calculation is formed on the edge attribute.
On the basis of the aggregation edge, the concept of the aggregation edge is further expanded by combining the concept of the time slice, so that the concept of the time sequence aggregation edge is led out. "time slicing" refers to cutting a succession of times in units of time (e.g., daily, hourly, past 30 minutes, etc.) to form a series of time windows of fixed length. All data are assigned to corresponding time windows according to the value of some agreed time attribute field (such as transaction time or event occurrence time). The data in the time window is aggregated according to the algorithm of the service requirement, and an aggregation value is calculated for the time window. Different calculation indexes have different aggregation value algorithms according to different calculation service contents. Further, different computation indicators may be assigned different time slice lengths according to their business implications.
The graph model structure and the storage mode of the invention are that the result of the calculation index is stored in the memory of the system by the data structure concept of 'time sequence aggregation edge'. The "time-series aggregation edge" result can be further used for calculating various real-time correlation or calculating the need of real-time decision and the like.
As shown in FIG. 2, an example of an implementation of the "time-sequential aggregate edges" of the present invention is given. Suppose account a runs down transfers to account B in six hours, with the numbers representing the transfer amount; the query index is the total transfer amount of the account A to the account B within a certain several hours; the time window length is defined as 1 hour. Here, the number of data items is 13 and the number of time windows is 6. The invention does not directly store 13 pieces of original data, but only stores the result obtained by aggregation in each time window, and the total number of the results is 6 and is called as an aggregation value. In response to the query, some calculations (in this case additions) are performed using these aggregate values in a pre-agreed algorithm. It can be seen that at the time of query, the computation is much less than with the original data. Thus, the present invention consumes less time and space than other databases. In this way, the information (amount) in the association relationship (transfer) between the entities (accounts) is calculated according to the predefined index (accumulated amount) to obtain the aggregation result of the time window of each data, and the aggregation result is stored in the memory of the system, i.e. a 'time sequence aggregation edge' is formed.
Fig. 3 further illustrates a complex business scenario. The nodes shown in the figure represent different entities in a business scenario. There is a user-based dimension and also an IP address-based dimension. All the indexes and the association relations (such as login frequency of IP _ A, device list appearing in past period of time and the like) have time sequence information, namely, the result required in the period of time can be shown according to different specified time windows. And according to the service requirement, utilizing a time sequence aggregation edge technology to place the aggregation calculation result in a memory for further inquiry and use.
In order to enable a user to conveniently inquire the data of the time sequence aggregation side, the invention also introduces a novel adaptive graph incidence relation inquiry language. The query language grammar resembles the Cypher language. The graph association relation query language supports basic fixed point, edge and attribute query, and also adds description semantics of time sequence information, so that a user can perform graph query on index calculation results in a certain time window, and the graph query comprises graph matching, graph filtering and the like. The invention also supports predicate filtering semantics such as 'all', 'any', and the like, and also supports fuzzy matching based on indefinite step number edges. Therefore, the technical support of the invention to the service scene is greatly enhanced.
The following table shows several examples of query languages:
graph matching in the invention refers to that given starting points and graph patterns needing matching return entity objects meeting the matching patterns. Such as "find all merchants that have been consumed in the past 24 hours with bank cards with account numbers 123 bound", etc. The "account binding bank card" and "using bank card for consumption at merchant" are two association relations, and the graph association relation query statement of the invention can be used for describing the serial association relation and matching the result.
Graph filtering in the present invention refers to a specified subset of results that can be found based on the filtering conditions, on the basis of graph matching. For example, the above example may add the condition "find a subset of merchants with a 3 month cumulative transaction amount greater than ten thousand dollars". The filter criteria may also specify a time window, and this time window may be different (3 months and 24 hours) from the time window used when the graph is matched.
Aggregation computation in the present invention refers to the aggregation computation capability on the "time-series aggregation edges"; it also refers to the calculation or aggregation calculation of the indices of different points, edges with respect to each other. For example, "find account number 123 account X transferred over the last 24 hours of the set S, maximum transfer amount X minus c, where c is the average of the maximum transfer amounts of all elements in S". The "maximum value" in the above example is for an account, and is an aggregation value on the "time-series aggregation edge" among the "time-series aggregation edges", so that the "maximum value" can be directly obtained from the map cube memory data; the average value is an aggregation operation performed on the set S, and the operation is performed in an execution plan of the graph association relation query language of the invention, and is aggregation calculation among different point indexes; "subtract" is then the general calculation of the intermediate results from each other.
The graph query in the invention is to return the result of the above operation to the querier or send the result to other modules as the characteristic for further graph analysis.
In order to compare the performance improvement of the association relation query by the technology of the aggregation edge and the technology of the time sequence aggregation edge, a performance test is also carried out by comparing with the open source software Neo4 j. Neo4j is a high performance open source NoSQL graph database. Since open-source Neo4j does not support horizontal spreading, testing is performed at a single node. The test focuses on examining the graph building efficiency and the query of the service scene.
The common trade pipelining in financial scenarios is used in the test, and the data structure is shown in the following table:
in the performance comparison of mapping, the method is used for loading 1 hundred million transaction running water through a Benchmark tool to perform mapping and evaluating the mapping efficiency. Neo4j completed 1 million data charting, which took 6153.544 s; the invention completes 1 hundred million data mapping, and takes 1026 s. Response times were counted for 90 data and the performance was compared as shown in the following table:
in the scenario of graph relationship query, the following three business scenarios are compared.
A first service scenario: in the past 4 days, starting from any support party card number A, the card number B exists at most 4 layers, the card number B meets the condition that in the past 1 day, the payment amount of a consumption channel which is not web is accumulated for 600 yuan; and the payment amount accumulated by the consumption channel for the web is not equal to 7000 yuan within the past n days (n is 2,3,4, 5); and the cumulative amount of payment is not equal to 10000 yuan for the past n days (n is 2,3,4, 5).
Service scene two: in the last 4 days, the maximum 4 layers are started from any support party card number A, and the accumulated collection amount and the payment amount in the last 2 days of each node in the graph are not more than 90 percent in relation.
And a third service scenario is that in the last 4 days, the card number A of any support party starts from 4 layers at most, adjacent payers and payees exist in the transaction link, and the standard deviation of the transaction amount in the last 2 days is more than 10.0 and the average value is less than 13000 yuan.
From the above test results, the average number of transactions per second is larger and the response time is shorter. The aggregation side and the time sequence aggregation side can accelerate the incidence relation query of mass time sequence data. Furthermore, the present invention also supports horizontal expansion into clusters, as opposed to the open source Neo4j graph database. And the support for the processing capacity of mass data is further promoted by using the performance of the cluster.
The foregoing is only a preferred embodiment of the present invention, and although the present invention has been disclosed in the preferred embodiments, it is not intended to limit the present invention. Those skilled in the art can make numerous possible variations and modifications to the present teachings, or modify equivalent embodiments to equivalent variations, without departing from the scope of the present teachings, using the methods and techniques disclosed above. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention are still within the scope of the protection of the technical solution of the present invention, unless the contents of the technical solution of the present invention are departed.
Claims (10)
1. A rapid massive time sequence data processing method based on an aggregation edge and a time sequence aggregation edge is characterized by comprising the following steps:
performing polymerization calculation on the time sequence data based on a time window to obtain a polymerization result;
generating a polymerization edge and a time sequence polymerization edge by using the polymerization result, and storing data in a memory according to a data structure of the polymerization edge;
establishing a graph relationship by the aggregation side or the time sequence aggregation side according to the mode of the attribute graph;
and for the generated graph relation, performing quick query based on the aggregation index.
2. The method of claim 1, wherein the generating of the aggregated edge comprises: in the process of establishing the associated relationship graph, according to the attribute fields which are selected and defined in advance on the service, the aggregation calculation is carried out in advance, and the result after the aggregation calculation is formed on the edge attribute.
3. The method of claim 1, wherein the generating of the time-series aggregation edge comprises: the continuous time is cut according to a certain time unit to form a series of time windows with fixed length. All time sequence data are distributed into corresponding time windows according to the value of the appointed time attribute field. And aggregating the data in the time window according to an aggregation algorithm required by the service to obtain an aggregation value corresponding to the time window.
4. The method according to claim 3, wherein different calculation indexes adopt different aggregation algorithms according to different calculation service contents; different calculation indexes may be assigned different time window lengths according to their business implications.
5. The method of claim 1, wherein the data of the aggregated edges is queried through a graph association query language that adds description semantics of timing information and supports point, edge, and attribute based queries.
6. The method of claim 5, wherein a user can perform graph queries for metric calculations within a certain time window, the graph queries including graph matching and graph filtering.
7. The method according to claim 6, wherein the graph matching is specifically: given a starting point and a graph pattern to be matched, entity objects satisfying the matching pattern are returned.
8. The method according to claim 6, wherein the graph filtering is specifically: a specified subset of results is found based on the filter criteria on the basis of the graph matching.
9. The method of claim 8, wherein in the graph filtering, the filtering condition may specify a time window, and the specified time window may be different from a time window used when the graph is matched.
10. The method of claim 6, wherein the graph association query language supports predicate filtering semantics and fuzzy matching based on an indefinite number of step edges.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911405552.XA CN111177188A (en) | 2019-12-30 | 2019-12-30 | Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911405552.XA CN111177188A (en) | 2019-12-30 | 2019-12-30 | Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111177188A true CN111177188A (en) | 2020-05-19 |
Family
ID=70655887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911405552.XA Pending CN111177188A (en) | 2019-12-30 | 2019-12-30 | Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111177188A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113515568A (en) * | 2021-07-13 | 2021-10-19 | 北京百度网讯科技有限公司 | Graph relation network construction method, graph neural network model training method and device |
| CN114841806A (en) * | 2022-03-17 | 2022-08-02 | 浙江邦盛科技股份有限公司 | A pattern matching system under real-time transfer link |
| CN114896285A (en) * | 2022-05-26 | 2022-08-12 | 浙江邦盛科技股份有限公司 | Bank flow calculation service real-time index system based on multi-dimensional intermediate state aggregation |
| CN115391428A (en) * | 2022-08-31 | 2022-11-25 | 广东工业大学 | Real-time calculation method for dynamic association relation of mass financial time sequence data |
| CN116089489A (en) * | 2022-11-16 | 2023-05-09 | 武汉众智鸿图科技有限公司 | Sequential data analysis method and system for continuous aggregation |
| CN117149843A (en) * | 2023-07-19 | 2023-12-01 | 浙江大学 | Graph data management method for sequence diagrams |
| CN119782344A (en) * | 2024-12-04 | 2025-04-08 | 支付宝(杭州)信息技术有限公司 | A method and device for optimizing execution of graph query |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090271729A1 (en) * | 2006-10-31 | 2009-10-29 | Disetronic Licensing Ag | Method for processing a chronological sequence of measurements of a time dependent parameter |
| JP2013246729A (en) * | 2012-05-28 | 2013-12-09 | Nec Corp | Similar time series data search system |
| CN103593433A (en) * | 2013-11-12 | 2014-02-19 | 中国科学院信息工程研究所 | Graph data processing method and system for massive time series data |
| CN104915879A (en) * | 2014-03-10 | 2015-09-16 | 华为技术有限公司 | Social relationship mining method and device based on financial data |
| JP2015176210A (en) * | 2014-03-13 | 2015-10-05 | カシオ計算機株式会社 | Data analysis apparatus, data analysis method, and data analysis program |
| CN107579855A (en) * | 2017-09-21 | 2018-01-12 | 桂林电子科技大学 | A hierarchical multi-domain visual security operation and maintenance method based on graph database |
| CN108268589A (en) * | 2017-12-05 | 2018-07-10 | 北京百度网讯科技有限公司 | Aggregate query method, apparatus, computer equipment and the readable medium of time series data |
| CN109101565A (en) * | 2018-07-16 | 2018-12-28 | 浪潮软件集团有限公司 | Graph database-based semantic search implementation method |
| CN109447807A (en) * | 2019-01-25 | 2019-03-08 | 杨城 | The weak center's trust data management system and method for the total score double-strand of digital bank |
| CN110019188A (en) * | 2017-09-15 | 2019-07-16 | 上海诺悦智能科技有限公司 | A kind of suspicious characteristic discovery method based on trade network node |
| CN110046183A (en) * | 2019-04-16 | 2019-07-23 | 北京易沃特科技有限公司 | A kind of time series data polymerization search method, equipment and medium |
-
2019
- 2019-12-30 CN CN201911405552.XA patent/CN111177188A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090271729A1 (en) * | 2006-10-31 | 2009-10-29 | Disetronic Licensing Ag | Method for processing a chronological sequence of measurements of a time dependent parameter |
| JP2013246729A (en) * | 2012-05-28 | 2013-12-09 | Nec Corp | Similar time series data search system |
| CN103593433A (en) * | 2013-11-12 | 2014-02-19 | 中国科学院信息工程研究所 | Graph data processing method and system for massive time series data |
| CN104915879A (en) * | 2014-03-10 | 2015-09-16 | 华为技术有限公司 | Social relationship mining method and device based on financial data |
| JP2015176210A (en) * | 2014-03-13 | 2015-10-05 | カシオ計算機株式会社 | Data analysis apparatus, data analysis method, and data analysis program |
| CN110019188A (en) * | 2017-09-15 | 2019-07-16 | 上海诺悦智能科技有限公司 | A kind of suspicious characteristic discovery method based on trade network node |
| CN107579855A (en) * | 2017-09-21 | 2018-01-12 | 桂林电子科技大学 | A hierarchical multi-domain visual security operation and maintenance method based on graph database |
| CN108268589A (en) * | 2017-12-05 | 2018-07-10 | 北京百度网讯科技有限公司 | Aggregate query method, apparatus, computer equipment and the readable medium of time series data |
| CN109101565A (en) * | 2018-07-16 | 2018-12-28 | 浪潮软件集团有限公司 | Graph database-based semantic search implementation method |
| CN109447807A (en) * | 2019-01-25 | 2019-03-08 | 杨城 | The weak center's trust data management system and method for the total score double-strand of digital bank |
| CN110046183A (en) * | 2019-04-16 | 2019-07-23 | 北京易沃特科技有限公司 | A kind of time series data polymerization search method, equipment and medium |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113515568A (en) * | 2021-07-13 | 2021-10-19 | 北京百度网讯科技有限公司 | Graph relation network construction method, graph neural network model training method and device |
| CN114841806A (en) * | 2022-03-17 | 2022-08-02 | 浙江邦盛科技股份有限公司 | A pattern matching system under real-time transfer link |
| CN114841806B (en) * | 2022-03-17 | 2024-11-29 | 浙江邦盛科技股份有限公司 | Pattern matching system under real-time transfer link |
| CN114896285A (en) * | 2022-05-26 | 2022-08-12 | 浙江邦盛科技股份有限公司 | Bank flow calculation service real-time index system based on multi-dimensional intermediate state aggregation |
| CN114896285B (en) * | 2022-05-26 | 2024-12-20 | 深圳市邦盛实时智能技术有限公司 | A real-time indicator system for bank flow computing business based on multi-dimensional intermediate state aggregation |
| CN115391428A (en) * | 2022-08-31 | 2022-11-25 | 广东工业大学 | Real-time calculation method for dynamic association relation of mass financial time sequence data |
| CN116089489A (en) * | 2022-11-16 | 2023-05-09 | 武汉众智鸿图科技有限公司 | Sequential data analysis method and system for continuous aggregation |
| CN117149843A (en) * | 2023-07-19 | 2023-12-01 | 浙江大学 | Graph data management method for sequence diagrams |
| CN119782344A (en) * | 2024-12-04 | 2025-04-08 | 支付宝(杭州)信息技术有限公司 | A method and device for optimizing execution of graph query |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111177188A (en) | Rapid massive time sequence data processing method based on aggregation edge and time sequence aggregation edge | |
| Ren et al. | On querying historical evolving graph sequences | |
| CN106570778B (en) | A kind of method that data integration based on big data is calculated with line loss analyzing | |
| Zhang et al. | A system for tender price evaluation of construction project based on big data | |
| CN111159184B (en) | Metadata tracing method, device and server | |
| Yujun et al. | Research on a hybrid prediction model for stock price based on long short-term memory and variational mode decomposition | |
| CN113836310B (en) | Knowledge graph driven industrial product supply chain management method and system | |
| CN110766272A (en) | Power business collaborative classification method and system based on ID3 decision tree algorithm | |
| CN108376143A (en) | A kind of novel OLAP precomputations model and the method for generating precomputation result | |
| CN110796331A (en) | Power business collaborative classification method and system based on C4.5 decision tree algorithm | |
| Chang et al. | Data analysis of digital currency networks: Namecoin case study | |
| CN115391428A (en) | Real-time calculation method for dynamic association relation of mass financial time sequence data | |
| CN105405069A (en) | Electricity purchase operating decision analysis and data processing method | |
| Shah et al. | Performance study of time series databases | |
| CN120181518A (en) | A computing power thermal evaluation method and decision-making system based on multidimensional data analysis | |
| WO2023103527A1 (en) | Access frequency prediction method and device | |
| CN106126727A (en) | A kind of big data processing method of commending system | |
| CN105303447A (en) | Method and device for carrying out credit rating through network information | |
| CN114138743A (en) | ETL task automatic configuration method and device based on machine learning | |
| WO2021134318A1 (en) | Rapid mass time-series data processing method based on aggregated edge and time-series aggregated edge | |
| CN114036174B (en) | Data updating method, device, equipment and storage medium | |
| Zhang et al. | Analysis of attention on venture capital: A method of complex network on time series | |
| Wang | Big data mining method of marketing management based on deep trust network model | |
| CN117273953A (en) | Data asset value evaluation method, device and computer equipment | |
| CN111753020A (en) | A method and device for establishing a relationship network model |
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 | ||
| CB02 | Change of applicant information |
Address after: Room ABCD, 17th floor, building D, Paradise Software Park, No.3 xidoumen Road, Xihu District, Hangzhou City, Zhejiang Province, 310012 Applicant after: Zhejiang Bangsheng Technology Co.,Ltd. Address before: Room ABCD, 17th floor, building D, Paradise Software Park, No.3 xidoumen Road, Xihu District, Hangzhou City, Zhejiang Province, 310012 Applicant before: ZHEJIANG BANGSUN TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200519 |
|
| RJ01 | Rejection of invention patent application after publication |