CN117726436A - Knowledge graph-based enterprise credit information link analysis algorithm - Google Patents
Knowledge graph-based enterprise credit information link analysis algorithm Download PDFInfo
- Publication number
- CN117726436A CN117726436A CN202410180336.4A CN202410180336A CN117726436A CN 117726436 A CN117726436 A CN 117726436A CN 202410180336 A CN202410180336 A CN 202410180336A CN 117726436 A CN117726436 A CN 117726436A
- Authority
- CN
- China
- Prior art keywords
- link
- abnormal
- value
- links
- chain
- 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.)
- Granted
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 19
- 230000002159 abnormal effect Effects 0.000 claims abstract description 183
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000012098 association analyses Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 22
- 230000005540 biological transmission Effects 0.000 claims description 14
- 238000011835 investigation Methods 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims 1
- 239000013598 vector Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of knowledge graphs, in particular to an enterprise credit information link analysis algorithm based on the knowledge graph, which comprises the following steps: dividing information links of knowledge maps constructed by information related to different enterprises into different links, and obtaining energy efficiency data of the different links; analyzing each link based on energy efficiency data of different links; if the energy efficiency difference value is smaller than the energy efficiency difference threshold value, generating a link normal signal and marking the link normal signal as an abnormal link; carrying out association analysis on each abnormal chain link to obtain an input end abnormal length value and an output end abnormal length value, and calculating to obtain an abnormal chain link representation value; based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal; the invention can judge the abnormal state and abnormal representation value of each chain link to reflect the quality of each chain link, and can analyze in a mode of comprehensive calculation of each chain link, thereby judging the overall quality level of the information link.
Description
Technical Field
The invention relates to the technical field of knowledge graphs, in particular to an enterprise credit information link analysis algorithm based on the knowledge graph.
Background
An information link refers to a path for transmitting information from a transmitting end to a receiving end, and includes multiple layers such as a link layer, a network layer, a transmission layer, and an application layer. The information link is used for establishing a communication session and realizing the transmission and exchange of information. The data transmitted over the information link may be different types of information such as text, voice, video, etc.;
chinese patent CN112446778A discloses a knowledge-graph-based enterprise credit risk recognition method, which processes enterprise feature fields to obtain structured enterprise feature vectors; extracting the relationship among enterprises from the enterprise data wide table, establishing an inter-enterprise graph network according to the relationship among enterprises, and forming corresponding graph structural features by enterprise nodes and enterprise relationship chains; using a label propagation algorithm to propagate credit risks along an enterprise relationship chain in the inter-enterprise graph network by using a preset black sample to obtain graph risk characteristics; generating an enterprise total feature vector according to the graph structural feature, the graph risk feature and the structured enterprise feature vector, and training a preset risk identification model according to training data to obtain a preset risk model; inputting the predicted feature vector into a preset risk model to obtain a predicted risk probability;
in the prior art, in the working process of the enterprise credit information link, the whole link is mainly calculated and analyzed to judge the operation capability of the link, the energy efficiency of each node or each link in the link cannot be judged, and the influence degree and relation of each link operation capability cannot be judged under the abnormal condition, so that the link cannot be rapidly checked, and the normal operation of the link is ensured.
Disclosure of Invention
The invention aims to provide an enterprise credit information link analysis algorithm based on a knowledge graph, which solves the following technical problems:
in the working process of the enterprise credit information-related link, the whole link is mainly calculated and analyzed to judge the operation capability of the link, the energy efficiency of each node or each link in the link cannot be judged, and the influence degree and relation of each link operation capability under the abnormal condition cannot be judged, so that the link cannot be rapidly checked, and the normal operation of the link is ensured.
The aim of the invention can be achieved by the following technical scheme:
the enterprise credit information link analysis algorithm based on the knowledge graph comprises the following steps:
step 1: dividing the information links of the knowledge graph into different links, and obtaining energy efficiency data of the different links; the energy efficiency data comprises energy efficiency differences of the links;
step 2: analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link;
if the energy efficiency difference value is smaller than the energy efficiency difference threshold value, generating a link normal signal and marking the link normal signal as an abnormal link;
carrying out association analysis on each abnormal chain link to obtain an input end abnormal length value and an output end abnormal length value, and marking the input end abnormal length value and the output end abnormal length value as ZDr and ZDC respectively;
by the formulaCalculating to obtain an abnormal chain link representation value ZBl, wherein a1 and a2 are weight proportion coefficients;
step 3: based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal;
wherein the link quality signal comprises a link quality high signal and a link quality low signal;
step 4: when a low-quality signal of a link is obtained, a link abnormal representation standard value is used as a horizontal dividing line, a midpoint of link operation time is used as a vertical dividing line, the horizontal dividing line and the vertical dividing line are substituted into a coordinate system, the number of coordinate points positioned on the upper side and the lower side of the horizontal dividing line is obtained, the number of coordinate points is marked as the number of coordinate points on the horizontal side and the number of coordinate points on the horizontal side respectively, and the number of coordinate points on the horizontal side is divided by the number of coordinate points on the horizontal side to obtain a horizontal distribution ratio;
the method comprises the steps of obtaining the number of coordinates located on the left side and the right side of a vertical dividing line, marking the number of coordinates as the number of vertical left coordinate points and the number of vertical right coordinate points respectively, dividing the number of vertical right coordinate points by the number of vertical left coordinate points, and obtaining a vertical distribution ratio;
and multiplying the obtained horizontal distribution ratio by the obtained vertical distribution ratio to obtain the abnormal value of the chain link duration.
As a further scheme of the invention: in step 1, dividing the information link into different links, and marking the different links as Li;
and obtaining the processing speeds of different chain links Li, respectively carrying out difference calculation on the processing speeds of the different chain links Li and the corresponding standard processing speeds of the chain links to obtain a chain link processing speed difference value, and dividing the chain link processing speed difference value by the corresponding standard processing speed of the chain links to obtain the energy efficiency difference value of the chain links.
As a further scheme of the invention: in step 2, if the energy efficiency difference value is greater than or equal to the energy efficiency difference threshold value, generating a link abnormal signal, and marking the link abnormal signal as a normal link.
As a further scheme of the invention: in step 2, the process of obtaining the input end abnormal length value ZDr is as follows:
the method comprises the steps of obtaining a transmission route in abnormal links, dividing the transmission route into an input route and an output route, obtaining the number of continuous abnormal links in each branch in the input route, recording the number as an abnormal length value of each branch, adding and summing the abnormal length values of all branches in the input route, and obtaining an input end abnormal length value ZDr.
As a further scheme of the invention: in step 2, the process of obtaining the abnormal length value ZDc at the output end is as follows:
the method comprises the steps of obtaining a transmission route in abnormal links, dividing the transmission route into an input route and an output route, obtaining the number of continuous abnormal links in each branch in the output route, recording the number as an abnormal length value of each branch, adding and summing the abnormal length values of all branches in the output route, and obtaining an output end abnormal length value ZDc.
As a further scheme of the invention: in step 3, obtaining the abnormal expression value of each chain link, and adding and summing the abnormal expression values of all the chain links on the information link to obtain a total abnormal expression value of the link;
comparing the obtained link abnormal expression total value with a link abnormal expression total threshold value;
if the total abnormal link performance value is greater than or equal to the total abnormal link performance threshold, generating a low link quality signal;
and if the total link abnormal performance value is smaller than the total link abnormal performance threshold, generating a link quality high signal.
As a further scheme of the invention: in step 4, a two-dimensional coordinate system is constructed by taking the link running time as an X axis and the abnormal link representation value as a Y axis, the obtained real-time abnormal link representation value is substituted into the two-dimensional coordinate system, and a plurality of abnormal link representation coordinate points are obtained.
As a further scheme of the invention: if the abnormal value of the duration of the link is greater than or equal to the abnormal threshold of the duration of the link, generating a link association influence small signal;
and if the abnormal value of the time length of the link is smaller than the abnormal threshold value of the time length of the link, generating a link association influence large signal.
As a further scheme of the invention: the method is characterized by further comprising the following steps:
step 5: when the link association influence small signal is obtained, all abnormal link expression values ZBl are obtained, the obtained abnormal link expression values ZBl are ordered from large to small, and the sequential fault checking is performed according to the order of the abnormal link expression values ZBl.
As a further scheme of the invention: and when a link association influence large signal is obtained, extracting the link with the maximum abnormal value of the link duration, and carrying out fault investigation on the link.
The invention has the beneficial effects that:
according to the invention, information links for constructing a knowledge graph by using information related to different enterprises are divided into different links, and energy efficiency data of the different links are obtained; analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link; based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal; the invention analyzes the information link in a mode of calculating each link individually, can judge the abnormal state and abnormal representation value of each link to reflect the quality of each link, can also analyze in a mode of calculating each link comprehensively, and can judge the overall quality level of the information link;
the method is based on a low-quality signal of a link and an abnormal link representation value, obtains a duration abnormal value of the link, and analyzes and obtains an influence signal; performing fault processing on the link based on the influence signal; the method and the device judge the mutual influence degree among the links on the link by analyzing the abnormal values of the link duration, so that the link is convenient to quickly detect and maintain when the link is abnormal, and the normal operation of the link is ensured.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a system block diagram of embodiment 1 of the present invention;
fig. 2 is a system block diagram of embodiment 2 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, the invention discloses an enterprise credit information link analysis algorithm based on a knowledge graph, which comprises the following steps:
step 1: dividing information links for constructing a knowledge graph by using information (including financial data) related to different enterprises into different links, and obtaining energy efficiency data of the different links;
wherein, different links comprise information acquisition, transmission, storage, processing, exchange and the like; the energy efficiency data includes energy efficiency differences;
in some embodiments, obtaining information related to different enterprises, and constructing information links of a knowledge graph, dividing the information links into different links according to the information links, and marking the different links as Li, wherein the steps comprise information acquisition, transmission, storage and processing;
obtaining the processing speeds of different chain links Li, respectively carrying out difference calculation on the processing speeds of the different chain links Li and the corresponding standard processing speeds of the chain links (preset by a person skilled in the art according to data) to obtain a chain link processing speed difference value, and dividing the chain link processing speed difference value by the corresponding standard processing speed of the chain links to obtain an energy efficiency difference value of the chain links;
step 2: analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link;
in some embodiments, energy efficiency differences for different links are obtained, and the energy efficiency differences for different links are compared to energy efficiency difference thresholds;
if the energy efficiency difference value is greater than or equal to the energy efficiency difference threshold value, generating a link abnormal signal, and marking the link abnormal signal as a normal link;
if the energy efficiency difference value is smaller than the energy efficiency difference threshold value, generating a link normal signal and marking the link normal signal as an abnormal link;
carrying out association analysis on each abnormal chain link to obtain an input end abnormal length value and an output end abnormal length value, and marking the input end abnormal length value and the output end abnormal length value as ZDr and ZDC respectively;
by the formulaCalculating to obtain an abnormal chain link representation value ZBl; wherein, a1 and a2 are weight proportion coefficients, a1+a2=1, the value of a1 is 0.59, and the value of a2 is 0.41; the values of a1 and a2 are respectively expressed as the influence degree proportion of the input end length value ZDr and the output end length value ZDc to the abnormal chain link representation value ZBl;
specifically, the acquisition process of the input end abnormal length value ZDr is as follows:
the method comprises the steps of obtaining a transmission route in abnormal links, dividing the transmission route into an input route and an output route, obtaining the number of continuous abnormal links in each branch in the input route, recording the number as an abnormal length value of each branch, adding and summing the abnormal length values of all branches in the input route, and obtaining an input end abnormal length value ZDr;
similarly, the obtaining process of the abnormal length value ZDc at the output end is as follows:
the method comprises the steps of obtaining a transmission route in abnormal links, dividing the transmission route into an input route and an output route, obtaining the number of continuous abnormal links in each branch in the output route, recording the number as an abnormal length value of each branch, adding and summing the abnormal length values of all branches in the output route, and obtaining an output end abnormal length value ZDc;
it should be explained that: the successive abnormal links are expressed as: the analyzed abnormal chain links are taken as starting points, the abnormal chain links which are uninterrupted along the branch direction are arranged, and the normal chain links are not contained in the middle of the abnormal chain links;
for example, if there are 3 links on one leg, two links near the abnormal link analyzed are abnormal links, while the third is a normal link, the abnormal length value of the leg is 2;
step 3: based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal;
wherein the link quality signal comprises a link quality high signal and a link quality low signal;
in some embodiments, obtaining an abnormal representation value of each link, and summing the abnormal representation values of all links on the information link to obtain a link abnormal representation total value;
comparing the obtained link abnormal expression total value with a link abnormal expression total threshold value;
if the total abnormal link performance value is greater than or equal to the total abnormal link performance threshold, generating a low link quality signal;
if the total abnormal link performance value is smaller than the total abnormal link performance threshold, generating a high link quality signal;
the technical scheme of the embodiment of the invention comprises the following steps: dividing information links of knowledge maps constructed by information related to different enterprises into different links, and obtaining energy efficiency data of the different links; analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link; based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal; the embodiment of the invention analyzes the information link in a mode of calculating each link individually, can judge the abnormal state and abnormal representation value of each link to reflect the quality of each link, can also analyze the information link in a mode of calculating each link comprehensively, and can judge the overall quality level of the information link.
Example 2
Referring to fig. 2, the invention relates to an enterprise credit information link analysis algorithm based on a knowledge graph, which further comprises the following steps:
step 4: acquiring a time length abnormal value of a link based on the low-quality signal of the link and the abnormal link representation value, and analyzing to obtain an influence signal;
wherein the influence signals include a link-associated influence small signal and a link-associated influence large signal;
in some embodiments, when a link quality low signal is obtained, a two-dimensional coordinate system is constructed by taking the link running time as an X axis and the abnormal link expression value as a Y axis, the obtained real-time abnormal link expression value is substituted into the two-dimensional coordinate system, and a plurality of abnormal link expression coordinate points are obtained;
taking a link abnormal representation standard value (preset by a person skilled in the art according to data) as a horizontal dividing line, taking the midpoint of the link running time as a vertical dividing line, substituting the horizontal dividing line and the vertical dividing line into a coordinate system, obtaining the number of coordinate points positioned on the upper side and the lower side of the horizontal dividing line, respectively marking the number of coordinate points as the number of coordinate points on the horizontal side and the number of coordinate points on the horizontal side, and dividing the number of coordinate points on the horizontal side by the number of coordinate points on the horizontal side to obtain a horizontal distribution ratio;
the method comprises the steps of obtaining the number of coordinates located on the left side and the right side of a vertical dividing line, marking the number of coordinates as the number of vertical left coordinate points and the number of vertical right coordinate points respectively, dividing the number of vertical right coordinate points by the number of vertical left coordinate points, and obtaining a vertical distribution ratio;
multiplying the obtained horizontal distribution ratio by the vertical distribution ratio to obtain a link duration abnormal value; the larger the abnormal value of the chain link length is, the better the chain link is in terms of treatment efficiency, and the smaller the abnormal value of the chain link length is, the worse the chain link is in terms of treatment efficiency;
performing variance calculation on all abnormal values of the chain link duration to obtain abnormal values of the link duration;
comparing the abnormal time length value of the link with the abnormal time length threshold value of the link;
if the abnormal value of the duration of the link is greater than or equal to the abnormal threshold of the duration of the link, generating a link association influence small signal;
if the abnormal value of the duration of the link is smaller than the abnormal threshold of the duration of the link, generating a link association influence large signal;
it should be noted that, the link association influence small signal indicates that the independent operation capability of a plurality of links on the link is strong, and the interference degree between the links is low; the link association influence small signal indicates that the independent operation capability of a plurality of links on a link is low and the interference degree between the links is high;
step 5: based on the influence signal, checking the chain links in the link;
in some embodiments, when a link association impact small signal is obtained, all abnormal link representation values ZBl are obtained, the obtained abnormal link representation values ZBl are ordered from large to small, and sequential troubleshooting is performed according to the order in which the abnormal link representation values ZBl are sized;
when a large link association influence signal is obtained, extracting a link with the maximum abnormal value of the link duration, and performing fault investigation on the link;
the fault detection items comprise hardware faults and software faults corresponding to the links;
the technical scheme of the embodiment of the invention comprises the following steps: acquiring a time length abnormal value of a link based on the low-quality signal of the link and the abnormal link representation value, and analyzing to obtain an influence signal; performing fault processing on the link based on the influence signal; the method and the device judge the mutual influence degree among the links on the link by analyzing the abnormal values of the link duration, so that the link is convenient to quickly detect and maintain when the link is abnormal, and the normal operation of the link is ensured.
The working principle of the invention is as follows: dividing information links of knowledge maps constructed by information related to different enterprises into different links, and obtaining energy efficiency data of the different links; analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link; based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal;
acquiring a time length abnormal value of a link based on the low-quality signal of the link and the abnormal link representation value, and analyzing to obtain an influence signal; based on the impact signal, the link is processed for failure.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The foregoing describes one embodiment of the present invention in detail, but the description is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention. All equivalent changes and modifications within the scope of the present invention are intended to be covered by the present invention.
Claims (10)
1. The enterprise credit information link analysis algorithm based on the knowledge graph is characterized by comprising the following steps of:
step 1: dividing the information links of the knowledge graph into different links, and obtaining energy efficiency data of the different links; the energy efficiency data comprises energy efficiency differences of the links;
step 2: analyzing each chain link based on the energy efficiency data of different chain links to obtain an abnormal state and an abnormal representation value of each chain link;
if the energy efficiency difference value is smaller than the energy efficiency difference threshold value, generating a link normal signal and marking the link normal signal as an abnormal link;
carrying out association analysis on each abnormal chain link to obtain an input end abnormal length value and an output end abnormal length value, and marking the input end abnormal length value and the output end abnormal length value as ZDr and ZDC respectively;
by the formulaCalculating to obtain an abnormal chain link representation value ZBl, wherein a1 and a2 are weight proportion coefficients;
step 3: based on the abnormal representation value of each chain link, analyzing the quality of the information link to obtain a link quality signal;
wherein the link quality signal comprises a link quality high signal and a link quality low signal;
step 4: substituting the obtained real-time abnormal chain link representation values into a two-dimensional coordinate system when the link quality low signal is obtained, and obtaining a plurality of abnormal chain link representation coordinate points; substituting the horizontal dividing line and the vertical dividing line into a coordinate system by taking a standard value of abnormal link performance as a horizontal dividing line and a midpoint of link operation time as a vertical dividing line, obtaining the number of coordinate points positioned on the upper side and the lower side of the horizontal dividing line, respectively marking the number of coordinate points as the number of coordinate points on the horizontal side and the number of coordinate points on the horizontal side, and dividing the number of coordinate points on the horizontal side by the number of coordinate points on the horizontal side to obtain a horizontal distribution ratio;
the method comprises the steps of obtaining the number of coordinates located on the left side and the right side of a vertical dividing line, marking the number of coordinates as the number of vertical left coordinate points and the number of vertical right coordinate points respectively, dividing the number of vertical right coordinate points by the number of vertical left coordinate points, and obtaining a vertical distribution ratio;
and multiplying the obtained horizontal distribution ratio by the obtained vertical distribution ratio to obtain the abnormal value of the chain link duration.
2. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 1, wherein in step 1, the information links are divided into different links, and the different links are marked as Li;
and obtaining the processing speeds of different chain links Li, respectively carrying out difference calculation on the processing speeds of the different chain links Li and the corresponding standard processing speeds of the chain links to obtain a chain link processing speed difference value, and dividing the chain link processing speed difference value by the corresponding standard processing speed of the chain links to obtain the energy efficiency difference value of the chain links.
3. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 1, wherein in step 2, if the energy efficiency difference value is greater than or equal to the energy efficiency difference threshold, a link abnormality signal is generated and marked as a normal link.
4. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 3, wherein in step 2, the process of obtaining the input end abnormal length value ZDr is:
the method comprises the steps of obtaining a transmission route in abnormal links, dividing the transmission route into an input route and an output route, obtaining the number of continuous abnormal links in each branch in the input route, recording the number as an abnormal length value of each branch, adding and summing the abnormal length values of all branches in the input route, and obtaining an input end abnormal length value ZDr.
5. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 4, wherein in step 2, the process of obtaining the output end abnormal length value ZDc is as follows:
the number of continuous abnormal chain links in each branch in the output route is obtained and recorded as the abnormal length value of each branch, and the abnormal length values of all the branches in the output route are added and summed to obtain the abnormal length value ZDC of the output end.
6. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 1, wherein in step 3, an abnormal representation value of each link is obtained, and the abnormal representation values of all links on the information link are summed together to obtain a link abnormal representation total value;
comparing the obtained link abnormal expression total value with a link abnormal expression total threshold value;
if the total abnormal link performance value is greater than or equal to the total abnormal link performance threshold, generating a low link quality signal;
and if the total link abnormal performance value is smaller than the total link abnormal performance threshold, generating a link quality high signal.
7. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 1, wherein in step 4, a two-dimensional coordinate system is constructed with the link run time as the X axis and abnormal link representation values as the Y axis.
8. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 7, wherein if the abnormal value of the link duration is greater than or equal to the abnormal threshold of the link duration, a link association influence small signal is generated;
and if the abnormal value of the time length of the link is smaller than the abnormal threshold value of the time length of the link, generating a link association influence large signal.
9. The knowledge-graph-based enterprise credit information link analysis algorithm of claim 1, further comprising the steps of:
step 5: when the link association influence small signal is obtained, all abnormal link expression values ZBl are obtained, the obtained abnormal link expression values ZBl are ordered from large to small, and the sequential fault checking is performed according to the order of the abnormal link expression values ZBl.
10. The knowledge-graph-based enterprise credit information link analysis algorithm according to claim 9, wherein when a link association influence large signal is obtained, a link with the maximum link duration anomaly value is extracted, and the link is subjected to fault investigation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410180336.4A CN117726436B (en) | 2024-02-18 | 2024-02-18 | Knowledge graph-based enterprise credit information link analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410180336.4A CN117726436B (en) | 2024-02-18 | 2024-02-18 | Knowledge graph-based enterprise credit information link analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117726436A true CN117726436A (en) | 2024-03-19 |
| CN117726436B CN117726436B (en) | 2024-06-07 |
Family
ID=90205682
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410180336.4A Active CN117726436B (en) | 2024-02-18 | 2024-02-18 | Knowledge graph-based enterprise credit information link analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117726436B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118245551A (en) * | 2024-05-28 | 2024-06-25 | 杭州跃翔科技有限公司 | Low-code platform data transmission method and system based on BI platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220166787A1 (en) * | 2020-11-24 | 2022-05-26 | Panasonic Automotive Systems Company Of America, Division Of Panasonic Corporation Of North America | Link anomaly detector |
| CN114978939A (en) * | 2022-06-10 | 2022-08-30 | 中煤科工重庆设计研究院(集团)有限公司 | Method for detecting network link quality |
| CN115766514A (en) * | 2022-11-02 | 2023-03-07 | 中国第一汽车股份有限公司 | Full link quality monitoring method and device of Internet of vehicles, storage medium and vehicle |
| CN116150493A (en) * | 2023-02-27 | 2023-05-23 | 江苏智云天工科技有限公司 | Industrial Internet high-quality supply chain recommendation method and system based on knowledge graph |
| CN117472625A (en) * | 2023-10-20 | 2024-01-30 | 联想(北京)有限公司 | Method and device for determining abnormal information of PCIE link |
-
2024
- 2024-02-18 CN CN202410180336.4A patent/CN117726436B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220166787A1 (en) * | 2020-11-24 | 2022-05-26 | Panasonic Automotive Systems Company Of America, Division Of Panasonic Corporation Of North America | Link anomaly detector |
| CN114978939A (en) * | 2022-06-10 | 2022-08-30 | 中煤科工重庆设计研究院(集团)有限公司 | Method for detecting network link quality |
| CN115766514A (en) * | 2022-11-02 | 2023-03-07 | 中国第一汽车股份有限公司 | Full link quality monitoring method and device of Internet of vehicles, storage medium and vehicle |
| CN116150493A (en) * | 2023-02-27 | 2023-05-23 | 江苏智云天工科技有限公司 | Industrial Internet high-quality supply chain recommendation method and system based on knowledge graph |
| CN117472625A (en) * | 2023-10-20 | 2024-01-30 | 联想(北京)有限公司 | Method and device for determining abnormal information of PCIE link |
Non-Patent Citations (1)
| Title |
|---|
| 许刚 等: "基于链路状态数据库的数据中心网络异常检测算法", 计算机研究与发展, no. 04, 15 April 2018 (2018-04-15) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118245551A (en) * | 2024-05-28 | 2024-06-25 | 杭州跃翔科技有限公司 | Low-code platform data transmission method and system based on BI platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117726436B (en) | 2024-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117726436B (en) | Knowledge graph-based enterprise credit information link analysis method | |
| CN107370732A (en) | System is found based on neutral net and the industrial control system abnormal behaviour of optimal recommendation | |
| CN110798314B (en) | Quantum key distribution parameter optimization method based on random forest algorithm | |
| CN111191767A (en) | Vectorization-based malicious traffic attack type judgment method | |
| CN113687972B (en) | Processing method, device, equipment and storage medium for abnormal data of business system | |
| CN110334105B (en) | Stream data abnormity detection method based on Storm | |
| CN110020637B (en) | Analog circuit intermittent fault diagnosis method based on multi-granularity cascade forest | |
| JP2018169959A (en) | Extraction system, extraction method, and extraction program of high contribution degree item for improving performance of multilayer neural network (deep learning) | |
| CN115237717A (en) | Micro-service abnormity detection method and system | |
| CN110188837A (en) | A kind of MVB network fault diagnosis method based on fuzzy neural | |
| CN112417766A (en) | Fault diagnosis method mainly based on fault-free data | |
| CN115293383A (en) | Game theory fused transformer risk cause analysis method | |
| CN113593605B (en) | Industrial audio fault monitoring system and method based on deep neural network | |
| CN115878992A (en) | Monitoring method and monitoring system for comprehensive pipe rack power supply system | |
| CN117614741A (en) | Network security vulnerability position detection method and system | |
| CN117389779A (en) | Micro-service fault root cause positioning method | |
| CN111415326A (en) | Method and system for detecting abnormal state of railway contact net bolt | |
| CN114978878B (en) | Positioning method, positioning device, electronic equipment and computer readable storage medium | |
| CN114500229B (en) | Network alarm positioning and analyzing method based on space-time information | |
| CN116955059A (en) | Root cause positioning method, root cause positioning device, computing equipment and computer storage medium | |
| CN112418255A (en) | Unsupervised anomaly detection scheme based on one-dimensional convolution generation type countermeasure network | |
| CN112132819A (en) | Communication network management monitoring method based on artificial intelligence | |
| Li et al. | A fault diagnosis method based on decision tree for wireless mesh network | |
| CN117974166B (en) | Power consumption information acquisition full-link anomaly monitoring and causal tracing method and system | |
| CN114217591B (en) | Network behavior self-learning system for industrial control system |
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 | ||
| GR01 | Patent grant |