CN109885700B - Unstructured data analysis method based on industrial knowledge graph - Google Patents
Unstructured data analysis method based on industrial knowledge graph Download PDFInfo
- Publication number
- CN109885700B CN109885700B CN201910139921.9A CN201910139921A CN109885700B CN 109885700 B CN109885700 B CN 109885700B CN 201910139921 A CN201910139921 A CN 201910139921A CN 109885700 B CN109885700 B CN 109885700B
- Authority
- CN
- China
- Prior art keywords
- node
- sequence
- nodes
- industrial
- degree
- 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.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides an unstructured data analysis method based on an industrial knowledge graph, the method comprises the steps of obtaining the industrial knowledge graph, wherein the industrial knowledge graph comprises nodes and edges, each node represents an industrial entity existing in the real world, and each edge represents the relation between the industrial entity and the industrial entity; determining the degree and associated nodes of each node in the industrial knowledge graph; determining the association probability of the associated nodes according to the degree of each node; and determining fusion information of each node according to the association probability. According to the method, the degree and the associated node of each node in the industrial knowledge graph are determined, the associated probability of the associated node is determined according to the degree of each node, the fusion information of each node is further determined according to the associated probability, and unstructured data analysis is carried out on the industrial knowledge graph.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an unstructured data analysis method based on an industrial knowledge graph.
Background
The internet has gradually transitioned from the Document world wide Web (Web of documents) to the data world wide Web (Web of data). A Knowledge Graph (Knowledge Graph) can be viewed as a large Graph, where nodes represent entities or concepts and edges in the Graph form relationships. In the knowledge graph, each entity and concept is identified using a globally unique deterministic ID, which corresponds to the identifier (identifier) of the target.
The knowledge graph is a semantic network essentially, is a data structure based on a graph, and is a semantic network formed by connecting knowledge points. Any network is composed of nodes (points) and edges (edges), the knowledge graph technology is a relational network obtained by connecting all different kinds of information together, the relational network is stored in a graph database, each node represents an entity existing in the real world, each Edge is a relation between the entities, and a knowledge base is formed after a large number of knowledge graphs are integrated and classified and organized according to a knowledge system.
In recent years, some knowledge bases are built by the industry in a mode of automatically extracting internet information, a knowledge graph is a basic means for realizing machine intelligence, and compared with the traditional knowledge base, the knowledge graph has the advantage that more accurate and intelligent retrieval can be performed.
How to perform unstructured data analysis based on an industrial knowledge graph becomes a key concern in the industry.
Disclosure of Invention
In order to solve the above problem, an unstructured data analysis method based on an industrial knowledge graph is provided in an embodiment of the present application.
In order to achieve the purpose, the invention adopts the main technical scheme that:
a method of unstructured data analysis based on an industrial knowledge graph, the method comprising:
s101, acquiring an industrial knowledge graph, wherein the industrial knowledge graph comprises nodes and edges, each node represents an industrial entity existing in the real world, and each edge represents a relation between the industrial entity and the industrial entity;
s102, determining the degree and the associated node of each node in the industrial knowledge graph;
s103, determining the association probability of the associated nodes according to the degree of each node;
and S104, determining fusion information of each node according to the association probability.
Optionally, the step of determining the associated node of each node in the industrial knowledge graph in S102 includes:
for any one of the nodes i, the node i,
s201, determining a node j communicated with any node i;
s202, determining the number of nodes between each connected node j and any node i;
s203, calculating the association degree of each connected node j;
and S204, determining the associated nodes according to the association degree of each connected node j.
Optionally, the industrial entity has a hierarchical attribute;
the S203 includes:
for any connected node j, the degree of association
Wherein m isjIs the grade, D, of the industrial entity represented by any of the connected nodes jiDegree, n, of said any node iijThe number of nodes between any connected node j and any node i.
Optionally, the S204 includes:
s204-1, sequencing according to the relevance of each connected node from small to large to obtain a first sequence;
s204-2, if the total number of the connected nodes is NiIs odd, it is determined that the first sequence is in the first sequence
A1, determining a1 as a first value b 1; wherein the content of the first and second substances,
is an upper rounding function;
s204-3, if the total number of the connected nodes is NiIs even, it is determined that the first sequence is in the first sequence
A2, located in the first sequence
A3, calculating
S204-4, calculating the deviation between the association degree of each connected node and b 1;
and S204-5, determining the associated node according to the deviation.
Optionally, the S204-4 includes:
for any one of the connected nodes j,
wherein, DeltajIs the deviation of the degree of association of the connected node j from b 1.
Optionally, the S204-5 includes:
s204-5-1, by DeltajSelecting from large to small
A plurality of connected nodes forming a second sequence;
s204-5-2, selecting 1 connected node in the second sequence optionally, putting the selected connected node into a third sequence, and deleting the selected connected node in the second sequence; the third sequence is initially a null sequence;
s204-5-3, sequentially selecting 1 connected node in the second sequence, and deleting the currently selected connected node in the second sequence if a node in the third sequence exists between the currently selected connected node and any node i; if the nodes between the currently selected connected node and the any node i do not exist in the third sequence, but the currently selected connected node is a node between any node in the third sequence and the any node i, deleting the currently selected connected node in the second sequence; if the nodes between the currently selected connected node and the any node i do not exist in the third sequence, but the currently selected connected node is not the node between each node in the third sequence and the any node i, putting the currently selected connected node into the third sequence, and deleting the currently selected connected node in the second sequence;
and S204-5-4, all the connected nodes in the third sequence are associated nodes.
Optionally, the S103 includes:
association probability of any associated node k of any node i
Wherein alpha is a correlation coefficient, SkThe association degree of any associated node k.
Alternatively,
wherein n isikIs the number of nodes, m, between any associated node k and any node ikIs the level of the industrial entity represented by any one of the associated nodes k.
Alternatively,
wherein n isikIs the number of nodes, m, between any associated node k and any node ikIs the grade, beta, of the industrial entity represented by any one of the associated nodes kiThe total number of associated nodes of any node i.
Optionally, the S104 includes:
for any one of the nodes i, the node i,
s104-1, determining the associated node max with the maximum associated probability;
s104-2, determining the node information of the maximum associated node max as the fusion information of any node i; or determining the node information of the maximum associated node and the node information of the intermediate node as the fusion information of any node i;
and the intermediate nodes are all nodes between the maximum associated node max and the any node i.
The invention has the beneficial effects that: the degree and the associated node of each node in the industrial knowledge graph are determined, the associated probability of the associated node is determined according to the degree of each node, the fusion information of each node is further determined according to the associated probability, and unstructured data analysis is carried out on the industrial knowledge graph.
Drawings
Specific embodiments of the present application will be described below with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart of an unstructured data analysis method based on an industrial knowledge graph according to an embodiment of the present application;
FIG. 2 illustrates an industrial knowledge graph structure provided by an embodiment of the present application.
Detailed Description
In recent years, some knowledge bases are built by the industry in a mode of automatically extracting internet information, a knowledge graph is a basic means for realizing machine intelligence, and compared with the traditional knowledge base, the knowledge graph has the advantage that more accurate and intelligent retrieval can be performed. However, how to perform unstructured data analysis based on the industrial knowledge graph becomes a key concern in the industry.
Based on the method, the degree and the associated node of each node in the industrial knowledge graph are determined, the associated probability of the associated node is determined according to the degree of each node, the fusion information of each node is further determined according to the associated probability, and unstructured data analysis is carried out on the industrial knowledge graph.
Referring to fig. 1, the implementation process of the unstructured data analysis method based on the industrial knowledge graph provided by this embodiment is as follows:
and S101, acquiring an industrial knowledge graph.
The industrial knowledge graph comprises nodes and edges, each node represents one industrial entity existing in the real world, and each edge represents the industrial entity and the relation between the industrial entities.
In addition, industrial entities have hierarchical attributes.
Taking the ship industrial knowledge graph shown in fig. 2 as an example, each circle in fig. 2 is a node, and each node represents an industrial entity existing in the real world, such as a node corresponding to a ship and a node corresponding to a power device. Each edge in fig. 2 represents an industrial entity and a relationship between industrial entities, such as an edge between a ship and a conventional submarine represents a relationship between a ship and a conventional submarine.
Different industrial entities have membership and thus have hierarchical attributes.
Still taking fig. 2 as an example, if the ship is level 1, then the destroyer, the cruiser, the conventional submarine, the guided missile boat, the nuclear submarine, and the guard ship are the first-level subdivision of the ship, so the destroyer, the cruiser, the conventional submarine, the guided missile boat, the nuclear submarine, and the guard ship are all level 2, and so on.
A level describes the degree of abstraction of an industrial entity, with higher levels being more abstract and covering a wider range.
And S102, determining the degree and the associated node of each node in the industrial knowledge graph.
Wherein the content of the first and second substances,
1) the manner of determining the degree of each node in the industrial knowledge graph is as follows:
the degree of any node i is the number of edges associated with any node i. I.e. the number of edges connecting the any node i with other nodes.
2) The mode of determining the associated node of each node in the industrial knowledge graph is as follows:
for any one of the nodes i, the node i,
s201, determining a node j communicated with any node i.
Here, a connected node is a node that can reach any node i by an edge.
Can be reached without direct arrival.
Taking fig. 2 as an example, if any node i is a node corresponding to a ship, the node corresponding to the guardship can directly reach the node corresponding to the ship as a connected node. Although the node corresponding to the control system cannot directly reach the node corresponding to the ship, the node corresponding to the conventional submarine can indirectly reach the node corresponding to the ship, so that the node corresponding to the control system is also a connected node.
S202, determining the number of nodes between each connected node j and any node i.
Taking fig. 2 as an example, if any node i is a node corresponding to a ship, there is no node between the node corresponding to the guard ship and the node corresponding to the ship, so the number of nodes between the node corresponding to the guard ship and the node corresponding to the ship is 0. Nodes corresponding to the conventional submarine are directly arranged between the nodes corresponding to the operating system and the nodes corresponding to the ship, so that the number of the nodes between the nodes corresponding to the operating system and the nodes corresponding to the ship is 1.
S203, calculating the association degree of each connected node j.
Wherein, for any connected node j, the association degree thereof
Wherein m isjLevel of industrial entity represented by any connected node j, DiDegree, n, of any node iijIs the number of nodes between any connected node j and any node i。
As shown in fig. 2, if the number of nodes between the node corresponding to the guard ship and the node corresponding to the ship is 0, the rank of the guard ship is 2, and the degree of the node corresponding to the ship is 6, the association degree of the node corresponding to the guard ship is (0+1)/(2 × 6) ═ 1/12.
During calculation of the association degree, the grade of the industrial entity represented by any connected node j, the degree of any node i and the number of nodes between any connected node j and any node i are considered. The level of the industrial entity represented by any connected node j describes the range size of any connected node j in the industry, and the larger the range is, the more things are covered, and the more important the connected node is. The number of nodes between any one connected node j and any one node i describes the distance between any one connected node j and any one node i, and the smaller the number of nodes between any one connected node j and any one node i is, the closer two nodes are, and the closer two nodes are. The degree of any node i reflects the degree of association of any node i, and the greater the degree of any node i, the more content any node i relates to, the more extensive the association, and further the lower the dependency of any node i on any connected node j.
The degree of any node i and the number of the nodes between any connected node j and any node i can be determined from the importance degree of any connected node j, the dependence degree of any node i on any connected node j and the tightness degree between any node i and any connected node j, so that the association degree can accurately reflect the association degree between any node i and any connected node j.
And S204, determining the associated nodes according to the association degree of each connected node j.
The implementation mode of the step is as follows:
s204-1, sorting the connected nodes from small to large according to the relevance of the connected nodes to obtain a first sequence.
S204-2, if the total number of the connected nodes is NiIs odd, it is determined that the first sequence is in the first sequence
A1, a1 is determined as the first value b 1.
Wherein the content of the first and second substances,
is a ceiling function.
With NiAs an example of 3, the number of the channels,
if the first sequence is 0.1,0.15,0.7, the number at position 2 in the first sequence is 0.15, and b 1-a 1-0.15.
S204-3, if the total number of the connected nodes is NiIs even, it is determined that the first sequence is in the first sequence
A2, located in the first sequence
A3, calculating
With NiAs an example of the case of 4,
if the first sequence is 0.1,0.15,0.7,0.71, the number at the 2 nd position in the first sequence is 0.15 ═ a2, the number at the 3 rd position in the first sequence is 0.7 ═ a3,
and S204-4, calculating the deviation between the association degree of each connected node and b 1.
For any one of the connected nodes j,
wherein, DeltajIs the deviation of the degree of association of the connected node j from b 1.
And S204-5, determining the associated node according to the deviation.
The implementation mode of S204-5 is as follows:
s204-5-1, by DeltajSelecting from large to small
And the connected nodes form a second sequence.
S204-5-2, optionally selecting 1 connected node in the second sequence, putting the selected connected node into the third sequence, and deleting the selected connected node in the second sequence.
Wherein the third sequence is initially a null sequence.
S204-5-3, sequentially selecting 1 connected node in the second sequence, and deleting the currently selected connected node in the second sequence if a node in the third sequence exists between the currently selected connected node and any node i; if the nodes between the currently selected connected node and any node i do not exist in the third sequence, but the currently selected connected node is a node between any node in the third sequence and any node i, deleting the currently selected connected node in the second sequence; and if the nodes between the currently selected connected node and any node i do not exist in the third sequence, but the currently selected connected node is not the node between each node in the third sequence and any node i, putting the currently selected connected node into the third sequence, and deleting the currently selected connected node in the second sequence.
For any node a, there is a node B between node a and any node i, and if node B is in the third sequence, node a is deleted in the second sequence. If node B is not in the third sequence, but node a is a node between node C and any node i in the third sequence, node a is deleted in the second sequence. If node B is not in the third sequence and no node A is between any node i and which node in the third sequence, node A is deleted in the second sequence.
After the step S204-5-3 is executed, the second sequence is a null sequence, the connected nodes in the original second sequence are positioned on the same chain, and only the connected node farthest from any node i (i.e. the most nodes exist between the connected nodes) enters the third sequence. And the connected nodes in the third sequence and any node i are ensured to be positioned on different chains and are the connected nodes farthest from any node i on the chain.
And S204-5-4, all the connected nodes in the third sequence are associated nodes.
The associated node is also a connected node, which is a special connected node, i.e. a connected node in the third sequence. And the connected nodes in the third sequence and any node i are positioned on different chains and are all the connected nodes which are farthest away from any node i on the chain. Therefore, the associated node is located on a different chain from any node i, and is a connected node farthest from any node i on the chain. The comprehensiveness of the pipe connection point is guaranteed, the remotest of the associated node is also guaranteed, and therefore the comprehensiveness of the final fusion information is guaranteed.
S103, determining the association probability of the associated nodes according to the degree of each node.
Association probability of any associated node k of any node i
Wherein alpha is a correlation coefficient, SkIs the degree of association of any associated node k.
nikis the number of nodes, m, between any associated node k and any node ikIs the rank, β, of the industrial entity represented by any associated node kiThe total number of associated nodes for any node i.
And S104, determining fusion information of each node according to the association probability.
After the association probabilities of the respective associated nodes are obtained, for any node i,
s104-1, determining the associated node max with the maximum associated probability.
S104-2, determining the node information of the maximum associated node max as the fusion information of any node i. Or, the node information of the maximum associated node max and the node information of the intermediate node are both determined as the fusion information of any node i.
The intermediate nodes are all nodes between the maximum associated node max and any node i.
The final fusion information is the information related to the associated node with the maximum associated probability, and the associated probability is obtained through the association degree of any associated node k, the number of nodes between any associated node k and any node i, the level of the industrial entity represented by any associated node k and the total number of associated nodes of any node i. The association degree of any associated node k reflects the association degree of any associated node k, the number of nodes between any associated node k and any node i reflects the dependency degree between any associated node k and any node i, the level of the industrial entity represented by any associated node k reflects the importance degree of any associated node k, and the total number of associated nodes of any node i reflects the association degree of any node i. The association probability can be comprehensively reflected through the data, and the accuracy of the final fusion information is ensured.
The corresponding node potential relation can be known through the fusion information, and subsequent data analysis and application are facilitated.
According to the method provided by the invention, the degree and the associated node of each node in the industrial knowledge graph are determined, the associated probability of the associated node is determined according to the degree of each node, and then the fusion information of each node is determined according to the associated probability, so that the unstructured data analysis of the industrial knowledge graph is realized.
It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
Finally, it should be noted that: the above-mentioned embodiments are only used for illustrating the technical solution of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (1)
1. A method for unstructured data analysis based on an industrial knowledge graph, the method comprising:
s101, acquiring an industrial knowledge graph, wherein the industrial knowledge graph comprises nodes and edges, each node represents an industrial entity existing in the real world, each edge represents a relation between the industrial entity and the industrial entity, and the industrial entities have level attributes;
s102, determining the degree and the associated node of each node in the industrial knowledge graph;
s103, determining the association probability of the associated nodes according to the degree of each node;
s104, determining fusion information of each node according to the association probability;
in S102, the method for determining the degree of each node in the industrial knowledge graph is as follows: the degree of any node i is the number of edges associated with any node i;
the step of determining the associated node of each node in the industrial knowledge graph in S102 includes:
for any one of the nodes i, the node i,
s201, determining a node j communicated with any node i;
s202, determining the number of nodes between each connected node j and any node i;
s203, calculating the association degree of each connected node j, and for any connected node j, calculating the association degree
Wherein m isjIs the grade, D, of the industrial entity represented by any of the connected nodes jiDegree, n, of said any node iijThe number of nodes between any one connected node j and any one node i is defined;
s204, determining a correlation node according to the correlation degree of each connected node j; the S204 comprises the following steps:
s204-1, sequencing according to the relevance of each connected node from small to large to obtain a first sequence;
s204-2, if the total number of the connected nodes is NiIs odd, it is determined that the first sequence is in the first sequence
A1, determining a1 as a first value b 1; wherein the content of the first and second substances,
is an upper rounding function;
s204-3, if the total number of the connected nodes is NiIs even, it is determined that the first sequence is in the first sequence
A2, located in the first sequence
A3, calculating
S204-4, calculating the deviation of the relevance degree of each connected node from b1, and for any connected node j,
wherein, ΔjIs the deviation of the degree of association of the connected node j from b 1;
s204-5, determining a correlation node according to the deviation;
the S204-5 comprises:
s204-5-1, by ΔjSelecting from large to small
A plurality of connected nodes forming a second sequence;
s204-5-2, selecting 1 connected node in the second sequence optionally, putting the selected connected node into a third sequence, and deleting the selected connected node in the second sequence; the third sequence is initially a null sequence;
s204-5-3, sequentially selecting 1 connected node in the second sequence, and deleting the currently selected connected node in the second sequence if a node in the third sequence exists between the currently selected connected node and any node i; if the nodes between the currently selected connected node and the any node i do not exist in the third sequence, but the currently selected connected node is a node between any node in the third sequence and the any node i, deleting the currently selected connected node in the second sequence; if the nodes between the currently selected connected node and the any node i do not exist in the third sequence, but the currently selected connected node is not the node between each node in the third sequence and the any node i, putting the currently selected connected node into the third sequence, and deleting the currently selected connected node in the second sequence;
s204-5-4, all the connected nodes in the third sequence are associated nodes;
the S103 includes:
association probability of any associated node k of any node i
Wherein alpha is a correlation coefficient, SkThe degree of association of any associated node k
Wherein n isikIs the number of nodes, m, between any associated node k and any node ikThe grade of the industrial entity represented by any associated node k; or
Wherein, betaiThe total number of associated nodes of any node i;
the S104 includes:
for any one of the nodes i, the node i,
s104-1, determining the associated node max with the maximum associated probability;
s104-2, determining the node information of the maximum associated node max as the fusion information of any node i; or determining the node information of the maximum associated node max and the node information of the intermediate node as the fusion information of any node i;
and the intermediate nodes are all nodes between the maximum associated node max and the any node i.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910139921.9A CN109885700B (en) | 2019-02-26 | 2019-02-26 | Unstructured data analysis method based on industrial knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910139921.9A CN109885700B (en) | 2019-02-26 | 2019-02-26 | Unstructured data analysis method based on industrial knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109885700A CN109885700A (en) | 2019-06-14 |
| CN109885700B true CN109885700B (en) | 2020-10-27 |
Family
ID=66929359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910139921.9A Active CN109885700B (en) | 2019-02-26 | 2019-02-26 | Unstructured data analysis method based on industrial knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109885700B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469359B (en) * | 2021-06-10 | 2023-04-18 | 西安交通大学 | Bullet matching method, system and equipment based on knowledge graph and readable storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9235653B2 (en) * | 2013-06-26 | 2016-01-12 | Google Inc. | Discovering entity actions for an entity graph |
| CN106445988A (en) * | 2016-06-01 | 2017-02-22 | 上海坤士合生信息科技有限公司 | Intelligent big data processing method and system |
| CN107341215B (en) * | 2017-06-07 | 2020-05-12 | 北京航空航天大学 | Multi-source vertical knowledge graph classification integration query system based on distributed computing platform |
| CN107491555B (en) * | 2017-09-01 | 2020-11-20 | 北京纽伦智能科技有限公司 | Knowledge graph construction method and system |
| CN108920527A (en) * | 2018-06-07 | 2018-11-30 | 桂林电子科技大学 | A kind of personalized recommendation method of knowledge based map |
| CN109189940A (en) * | 2018-09-05 | 2019-01-11 | 南京大学 | A kind of knowledge sharing method of servicing based on crowdsourcing and graphical spectrum technology |
| CN109376269B (en) * | 2018-12-05 | 2021-01-19 | 西安交通大学 | Cross-course video subgraph recommendation method based on map association |
-
2019
- 2019-02-26 CN CN201910139921.9A patent/CN109885700B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109885700A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108874878B (en) | Knowledge graph construction system and method | |
| US10540965B2 (en) | Semantic re-ranking of NLU results in conversational dialogue applications | |
| CN108345690B (en) | Intelligent question and answer method and system | |
| JP2019020893A (en) | Non-factoid type question answering machine | |
| US20190163835A1 (en) | Structuring incoherent nodes by superimposing on a base knowledge graph | |
| CN110597804B (en) | Facilitating spatial indexing on a distributed key value store | |
| CN112989055B (en) | Text recognition method and device, computer equipment and storage medium | |
| CN109960722B (en) | Information processing method and device | |
| CN105183770A (en) | Chinese integrated entity linking method based on graph model | |
| CN104573130A (en) | Entity resolution method based on group calculation and entity resolution device based on group calculation | |
| US20210326525A1 (en) | Device and method for correcting context sensitive spelling error using masked language model | |
| CN116303971A (en) | Few-sample form question-answering method oriented to bridge management and maintenance field | |
| CN104699767A (en) | Large-scale ontology mapping method for Chinese languages | |
| CN110727769B (en) | Corpus generation method and device and man-machine interaction processing method and device | |
| CN113094593A (en) | Social network event recommendation method, system, device and storage medium | |
| US10936638B2 (en) | Random index pattern matching based email relations finder system | |
| CN116244333A (en) | Database query performance prediction method and system based on cost factor calibration | |
| CN110532393B (en) | Text processing method and device and intelligent electronic equipment thereof | |
| CN104572632B (en) | A kind of method in the translation direction for determining the vocabulary with proper name translation | |
| CN109885700B (en) | Unstructured data analysis method based on industrial knowledge graph | |
| CN109582771B (en) | Intelligent customer interaction method based on mobile application and oriented to electric power field | |
| CN113342982B (en) | Enterprise industry classification method integrating Roberta and external knowledge base | |
| Nath et al. | Resolving scalability issue to ontology instance matching in semantic web | |
| CN115455302A (en) | Knowledge graph recommendation method based on optimized graph attention network | |
| CN110309258A (en) | A kind of input checking method, server and computer readable storage medium |
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 | ||
| GR01 | Patent grant |