CN112100400A - Node recommendation method and device based on knowledge graph - Google Patents
Node recommendation method and device based on knowledge graph Download PDFInfo
- Publication number
- CN112100400A CN112100400A CN202010959476.3A CN202010959476A CN112100400A CN 112100400 A CN112100400 A CN 112100400A CN 202010959476 A CN202010959476 A CN 202010959476A CN 112100400 A CN112100400 A CN 112100400A
- Authority
- CN
- China
- Prior art keywords
- nodes
- node
- centrality
- graph
- knowledge
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000004364 calculation method Methods 0.000 claims abstract description 24
- 238000004590 computer program Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 239000000969 carrier Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 241001441571 Hiodontidae Species 0.000 description 1
- 229910000792 Monel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010225 co-occurrence analysis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a node recommendation method and device based on a knowledge graph, relates to the technical field of knowledge graphs, and mainly aims to calculate the centrality of relative nodes in the knowledge graph by combining query contents of a user and improve the accuracy of the centrality calculation of recommended nodes. The main technical scheme of the invention is as follows: matching related nodes in the knowledge graph according to the query content of the user; respectively calculating the centrality of each node according to the in-degree relation of the nodes in the knowledge graph; and determining the sequence of recommending the nodes to the user according to the centrality of the nodes. Thereby improving the accuracy of the result feedback to the user. The method is used for determining the node centrality aiming at different user requirements for the nodes in the knowledge graph.
Description
Technical Field
The invention relates to the technical field of knowledge graphs, in particular to a node recommendation method and device based on a knowledge graph.
Background
Knowledge map (Knowledge Graph) is a series of different graphs displaying Knowledge development process and structure relationship in the book intelligence field, describing Knowledge resources and carriers thereof by using visualization technology, mining, analyzing, constructing, drawing and displaying Knowledge and mutual relation between Knowledge resources and Knowledge carriers. The knowledge graph is a modern theory which achieves the aim of multi-discipline fusion by combining theories and methods of applying subjects such as mathematics, graphics, information visualization technology, information science and the like with methods such as metrology introduction analysis, co-occurrence analysis and the like and utilizing a visualized graph to vividly display the core structure, development history, frontier field and overall knowledge framework of the subjects. Therefore, the knowledge graph has more and more extensive applications in different industries and different fields, such as question answering, searching, personalized recommendation and the like.
The node centrality calculation is one of core contents for analyzing the knowledge graph, and a centrality algorithm (centrality algorithm) is used for understanding roles of specific nodes in the graph and influences of the roles on the network, so that understanding of group dynamics such as credibility, accessibility and object propagation speed is facilitated. Currently, the common centrality algorithm is mainly divided into two ways: one method is to calculate the degree of entry and exit of a node, such as a degree centrality calculation method, which judges the centrality of the node by comparing the sum of the degrees of entry and exit of the node, and because the degree of association between the nodes is not considered, the calculation result has large redundancy and the calculation accuracy is not high. The other is to calculate by the shortest distance between nodes, such as a near-centrality algorithm, that is, if the sum of the distances from a certain node to all nodes is shortest, the centrality of the node is highest, but the algorithm needs to consider all nodes in the whole graph, the calculation complexity is high, and the applicability to a specific scene is not good, that is, the accuracy difference of the calculation result for different scenes is obvious.
Disclosure of Invention
In view of the above problems, the present invention provides a node recommendation method and apparatus based on a knowledge graph, and mainly aims to calculate the centrality of relevant nodes in the knowledge graph in combination with query content of a user, and improve the accuracy of the centrality calculation of recommended nodes.
In order to achieve the purpose, the invention mainly provides the following technical scheme:
in a first aspect, the present invention provides a node recommendation method based on a knowledge graph, including:
matching related nodes in the knowledge graph according to the query content of the user;
respectively calculating the centrality of each node according to the in-degree relation of the nodes in the knowledge graph;
and determining the sequence of recommending the nodes to the user according to the centrality of the nodes.
Preferably, the matching of relevant nodes in the knowledge graph according to the query content of the user includes:
determining query types according to query contents of users, wherein the query types are types contained in nodes of the knowledge graph;
matching nodes in the knowledge-graph having the query class.
Preferably, the calculating the centrality of each node according to the in-degree relationship of the node in the knowledge graph includes:
acquiring the relation weight of corresponding edges in the knowledge graph based on the in-degree relation of the nodes;
and calculating the centrality of each node by using the relation weight and the entrance degree relation.
Preferably, the calculating the centrality of each node by using the relationship weight and the introductivity relationship includes:
carrying out weighted summation on the degree-of-entry relationship of the nodes according to the relationship weight to obtain the weight of the degree-of-entry relationship of the nodes;
and carrying out normalization processing on the entry relation weight of the nodes according to the number of the nodes and the nodes with the entry relation to obtain the centrality of the nodes.
Preferably, the calculating the centrality of each node according to the in-degree relationship of the node in the knowledge graph further includes:
judging whether the nodes are positioned in the same sub-graph in the knowledge graph;
if the nodes are located in the same subgraph, normalizing the entry relation weight of each node according to each node and the number of the nodes with the entry relation to obtain the centrality of each node;
and if the nodes are positioned in different subgraphs, taking the specific weight of the subgraph to which the nodes belong in the knowledge graph as the weight, and carrying out weighted calculation on the centrality of each node to determine the centrality of each node relative to the knowledge graph.
Preferably, the method further comprises:
calculating the centrality of each node through a formula, wherein the formula is as follows:
wherein weighted _ index represents the weight of the edge corresponding to the in-degree relation; total _ weighted _ index represents the total weight of the edge corresponding to the degree relation between each node and the same subgraph; m represents the number of nodes having an in-degree relationship with the calculated nodes; n represents the number of centrality nodes needing to be calculated in the same subgraph; v represents the number of nodes in the subgraph; v represents the number of nodes in the knowledge-graph.
In a second aspect, the present invention provides a node recommendation apparatus based on a knowledge-graph, the apparatus comprising:
the matching unit is used for matching related nodes in the knowledge graph according to the query content of the user;
the calculating unit is used for respectively calculating the centrality of each node obtained by the matching unit according to the in-degree relation of the node in the knowledge graph;
and the recommending unit is used for determining the order of recommending the nodes to the user according to the centrality of the nodes obtained by the calculating unit.
Preferably, the matching unit includes:
the determining module is used for determining query types according to query contents of users, wherein the query types are types contained in nodes of the knowledge graph;
a matching module for matching nodes in the knowledge-graph having the query class determined by the determining module.
Preferably, the calculation unit includes:
the acquisition module is used for acquiring the relation weight of the corresponding edge in the knowledge graph based on the in-degree relation of the nodes;
and the calculating module is used for calculating the centrality of each node by using the relation weight and the degree-of-entry relation obtained by the obtaining module.
Preferably, the calculation module is specifically configured to:
carrying out weighted summation on the degree-of-entry relationship of the nodes according to the relationship weight to obtain the weight of the degree-of-entry relationship of the nodes;
and carrying out normalization processing on the entry relation weight of the nodes according to the number of the nodes and the nodes with the entry relation to obtain the centrality of the nodes.
Preferably, the computing unit further includes:
the judging module is used for judging whether the nodes are positioned in the same sub-graph in the knowledge graph;
the first calculation module is used for carrying out normalization processing on the entry relation weight of each node according to each node and the number of the nodes with the entry relation if the judgment module is determined to be positioned in the same sub-graph, so as to obtain the centrality of each node;
and the second calculation module is used for performing weighted calculation on the centrality of each node by taking the specific gravity of the sub-graph to which the node belongs in the knowledge graph as the weight if the judgment module determines that the node is positioned in different sub-graphs, and determining the centrality of each node relative to the knowledge graph.
In another aspect, the present invention further provides a processor, where the processor is configured to execute a program, where the program executes the method for node recommendation based on a knowledge graph according to the first aspect.
In another aspect, the present invention further provides a storage medium, where the storage medium is used to store a computer program, where the computer program is run to control an apparatus in which the storage medium is located to execute the method for node recommendation based on a knowledge graph according to the first aspect.
By means of the technical scheme, when nodes obtained based on the knowledge graph are recommended to a user, the nodes needing to calculate the centrality in the knowledge graph are filtered in the knowledge graph according to the query content of the user, calculation of all the nodes in the knowledge graph is avoided, then the centrality of the nodes is calculated according to the relation of the nodes in the knowledge graph, the degree-out relation irrelevant to the query content of the user is eliminated, the evaluation standard of the centrality is more inclined to the query content of the user, the node recommendation sequence determined by the centrality is more in line with the query requirement of the user, the query accuracy is improved, and meanwhile the query experience of the user is improved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart illustrating a method for node recommendation based on a knowledge-graph according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating another knowledge-graph based node recommendation method proposed by an embodiment of the present invention;
FIG. 3 illustrates a knowledge-graph for computing node centrality in an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a node recommendation apparatus based on a knowledge-graph according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of another node recommendation apparatus based on a knowledge-graph according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The embodiment of the invention provides a node recommendation method based on a knowledge graph, which is mainly used for an inquiry application scene constructed based on the knowledge graph, a user inputs inquiry content, a system matches required content nodes for the user by using the corresponding knowledge graph, in the process, the system calculates the centrality of the nodes in the knowledge graph according to the inquiry content input by the user so as to determine the sequence of recommending the nodes to the user, and the specific execution steps are shown in figure 1 and comprise the following steps:
101. and matching related nodes in the knowledge graph according to the query content of the user.
The method comprises the steps of analyzing query contents input by a user based on a natural language processing technology to match the contents of nodes in a knowledge graph, determining which nodes are contents which the user wants to view, and determining the nodes as nodes needing to calculate centrality.
The method aims to filter nodes in the knowledge graph spectrum, select nodes needing to calculate the centrality, and avoid overlarge calculation load caused by calculating the centrality for all the nodes.
102. And respectively calculating the centrality of each node according to the in-degree relation of the nodes in the knowledge graph.
A knowledge graph may be generally represented as a directed graph, where a relationship is referred to as an in-degree when the relationship points to a node, and vice versa. The degree of entry relationship in this step may be understood as an edge of the knowledge graph spectrum pointing to the central node currently being calculated, and in this embodiment, it is considered that the more the degree of entry relationship is for a node, the greater the importance of the node with respect to the content queried by the user is. Thus, for the centrality of a node, it can be quantified as the number of in-degree relationships.
In order to make the centrality corresponding to different nodes comparable, the quantized in-degree relationship needs to be normalized, so as to obtain the quantized value of the centrality of the node required in this embodiment.
103. And determining the sequence of recommending the nodes to the user according to the centrality of the nodes.
As can be seen from the above steps, the obtained centrality of the node is a quantized numerical value subjected to normalization processing, and therefore, the importance degree of the node can be determined according to the magnitude of the numerical value, and generally, the node with the greater centrality is more important with respect to the content of the user query. Therefore, in this step, the nodes are sorted according to the centrality value, so as to obtain the order of recommending the nodes to the user, so that when the result is fed back to the user, the corresponding node content is displayed according to the order.
Based on the implementation manner of fig. 1, it can be seen that the node recommendation method provided in the embodiment of the present invention is a method for performing ranking recommendation on nodes by calculating the centrality of the nodes in an application scene based on a knowledge graph. The method mainly comprises the steps of screening nodes, avoiding huge calculation load caused by calculation of all nodes, considering content relations among the nodes, determining the centrality of the nodes based on the in-degree relations, and further determining a recommendation sequence. Compared with the existing centrality algorithm for the nodes, the nodes needing to calculate the centrality in the embodiment of the invention have more pertinence, save a large amount of calculation resources, and further limit the relation among the nodes, so that the centrality has stronger tendency, that is, when the same node aims at different problems of users, the corresponding centrality is different, thus the difference of user feedback can be improved, and the users can obtain more accurate query results.
Further, on the basis of fig. 1, in the embodiment of the present invention, when the centrality is calculated, different weights may be given to the introductivity relationship of the node according to the relevance of the content or the service requirement, so as to change the corresponding centrality of the node when querying the user. The specific implementation steps are shown in fig. 2 and include:
201. and determining the query category according to the query content of the user.
The query type refers to a type contained in a node of the knowledge graph, and generally, the node in the knowledge graph has a type attribute, so that the node can be classified and managed.
The query type in this step may be obtained by parsing based on a natural language processing technique, and a specific manner of the query type is not limited in this embodiment of the present invention.
202. Nodes having query categories are matched in the knowledge graph.
In the step, screening is carried out based on the query type determined by the query content of the user, and a conforming node is selected to calculate the centrality of the node.
203. And acquiring the relation weight of the corresponding edge in the knowledge graph based on the in-degree relation of the nodes.
In the embodiment, the relation weight of the edge is mainly determined manually, in a specific application, after a user inputs query content, the system acquires graph data from a graph database, the graph database records specific data of a knowledge graph, wherein the relation weight corresponding to the edge in the graph is included, the relation weight is generally determined by manual assignment, and a service person or a maintenance person can adjust the value of the relation weight in real time according to actual service requirements.
204. And calculating the centrality of each node by using the relation weight and the entrance relation.
In the embodiment, when the node centrality is calculated, in addition to considering from the number of the in-degree relations, the quality of each in-degree relation is also evaluated, that is, the quality of one in-degree relation is determined by the relation weight, and the influence of the in-degree relation with a larger weight on the node centrality is larger.
The specific process for calculating the centrality of the node comprises the following steps:
firstly, weighting and summing the degree relation of the node according to the relation weight to obtain the weight of the degree relation of the node. I.e. the weighted sum of the in-degree relations corresponding to the node.
And secondly, normalizing the entry relation weight of the node according to the number of the nodes and the nodes with the entry relation to obtain the centrality of the node. The centrality of the node after the normalization process can be compared with the centrality of other nodes.
205. And determining the sequence of recommending the nodes to the user according to the centrality of the nodes.
This step is the same as step 103 shown in fig. 1, and is not described herein again.
Further, in another preferred embodiment of the present invention, since the directed graph in the knowledge graph may be divided into a plurality of subgraphs, and there is no association between the subgraphs, but the computation manner in the above embodiment cannot effectively perform sorting on nodes between different subgraphs.
And when the nodes are positioned in the same subgraph, normalizing the weight of the degree relation of each node according to each node and the number of the nodes with the degree relation to obtain the centrality of each node. I.e. calculated according to the steps described above in the embodiment shown in fig. 2.
And when the nodes are positioned in different subgraphs, the proportion of the subgraph to which the nodes belong in the knowledge graph is taken as the weight, the centrality of each node is calculated in a weighting mode, and the centrality of each node relative to the knowledge graph is determined. That is, based on the centrality calculated in fig. 2, the centrality is weighted, the added weight is the proportion of the sub-graph to which the node belongs in the knowledge graph, and the proportion can be calculated based on the number of nodes in the sub-graph and the number of all nodes in the knowledge graph. Thus, there is also comparability to nodes in different subgraphs.
Based on the content of the above embodiments, the calculation of the node centrality in the present invention can be calculated by the following formula, which is expressed as:
wherein weighted _ index represents the weight of the edge corresponding to the in-degree relation; total _ weighted _ index represents the total weight of the edge corresponding to the degree relation between each node and the same subgraph; m represents the number of nodes having an in-degree relationship with the calculated nodes; n represents the number of centrality nodes needing to be calculated in the same subgraph; v represents the number of nodes in the subgraph; v represents the number of nodes in the knowledge-graph.
When the nodes needing to calculate the centrality are all located in the same sub-graph, the formula can be simplified as follows:
the process of determining the order of the recommended nodes will be illustrated below with respect to the above formula in conjunction with the knowledge graph shown in fig. 3. As shown in fig. 3, what is recorded in the knowledge-graph is the painting and its author, and the related information of the author, so it can be known that the nodes of the knowledge-graph can be divided into three categories: works, authors, author related information.
When the user queries based on the knowledge graph, if the query content is: which are known by modern artists. According to the query content, the nodes in the knowledge graph can be screened, and the nodes with the types as authors are selected, wherein the method comprises the following steps: DaVinci, Monel, Zhang Daqian, Xupeshao hong.
After the nodes are determined, the centrality of each node is calculated one by using the formula based on the relation weight of the edges in the knowledge graph acquired from the graph database.
Since fig. 3 includes two sub-graphs, when calculating the centrality, the above complete formula is used for calculation, that is, the formula:
taking node (da vinci) as an example, the centrality calculation process is:
wherein, the total number of 5 edges having an in-degree relation with the node is provided, wherein the relation weight of the 'Luo-Floating House' and the 'dinner after left' is 1.5, the relation weight of the 'Mona Lisa' is 2, and when the in-degree relation is 1, the sigma is calculatedmweighted _ index is 1+1+1.5+1.5+2 ═ 7; the subgraph of the node of the 'DaVinci' is provided with two nodes needing to be calculated, namely the 'DaVinci' node and the 'Mooney', therefore, the sigmantotal _ weighted _ index is (1+1+1.5+1.5+2) + (0.5+0.5+1+1+2) ═ 12; according to the knowledge graph, V is the number (17) of nodes of a sub-graph where the 'DaVinci' nodes are located, and V is the number (28) of all nodes in the knowledge graph. The centrality value of the resulting "da vinci" node is 0.346.
By analogy, the centrality of a "Monel" node is
The centrality of the Zhangda Qian node is
The centrality of the "sorrow" node is
Based on the comparison of the centrality, the order of recommending the nodes to the user is obtained as follows: DaVinci, Xupeshahong, Monel, Zhang Daqian.
In combination with the above example, the node recommendation method based on the knowledge graph provided by the embodiment of the present invention calculates the centrality of the node in a multi-dimensional manner based on the relation weight of the in-degree relation and the in-degree relation, and also considers the relation between sub-graphs when the knowledge graph has a plurality of sub-graphs, so as to further highlight the importance of the node relative to the user query content.
Further, as an implementation of the method embodiments shown in fig. 1 and 2, an embodiment of the present invention provides a node recommendation device based on a knowledge graph, where the device may calculate the centrality of a relevant node in the knowledge graph in combination with query content of a user, so as to improve the accuracy of calculating the centrality of a recommended node. The embodiment of the apparatus corresponds to the foregoing method embodiment, and details in the foregoing method embodiment are not repeated in this embodiment for convenience of reading, but it should be clear that the apparatus in this embodiment can correspondingly implement all the contents in the foregoing method embodiment. As shown in fig. 4 in detail, the apparatus includes:
a matching unit 31, configured to match relevant nodes in the knowledge graph according to the query content of the user;
the calculating unit 32 is configured to calculate the centrality of each node obtained by the matching unit 31 according to the degree-of-entry relationship of the node in the knowledge graph;
and the recommending unit 33 is configured to determine an order of recommending the nodes to the user according to the centrality of the nodes obtained by the calculating unit 32.
Further, as shown in fig. 5, the matching unit 31 includes:
a determining module 311, configured to determine a query category according to query content of a user, where the query category is a category included in a node of the knowledge graph;
a matching module 312, configured to match nodes in the knowledge-graph with the query category determined by the determining module 311.
Further, as shown in fig. 5, the calculating unit 32 includes:
an obtaining module 321, configured to obtain a relationship weight of a corresponding edge in the knowledge graph based on an in-degree relationship of a node;
a calculating module 322, configured to calculate the centrality of each node by using the relationship weight and the entrance relation obtained by the obtaining module 321.
Further, the calculating module 322 is specifically configured to:
carrying out weighted summation on the degree-of-entry relationship of the nodes according to the relationship weight to obtain the weight of the degree-of-entry relationship of the nodes;
and carrying out normalization processing on the entry relation weight of the nodes according to the number of the nodes and the nodes with the entry relation to obtain the centrality of the nodes.
Further, as shown in fig. 5, the calculating unit 32 further includes:
a judging module 323, configured to judge whether each node is located in the same sub-graph in the knowledge graph;
a first calculating module 324, configured to, if the determining module 323 determines that the nodes are located in the same sub-graph, perform normalization processing on the entry relation weight of each node according to each node and the number of nodes having an entry relation, so as to obtain the centrality of each node;
a second calculating module 325, configured to, if the determining module 323 determines that the node is located in a different sub-graph, perform weighted calculation on the centrality of each node by using the specific gravity of the sub-graph to which the node belongs in the knowledge graph as a weight, and determine the centrality of each node relative to the knowledge graph.
Further, an embodiment of the present invention further provides a processor, where the processor is configured to execute a program, where the program executes the method for node recommendation based on a knowledge graph as described in fig. 1 to 2.
Further, an embodiment of the present invention further provides a storage medium, where the storage medium is used to store a computer program, where the computer program is run to control a device in which the storage medium is located to perform the method for node recommendation based on a knowledge graph as described in fig. 1-2 above.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
It will be appreciated that the relevant features of the method and apparatus described above are referred to one another. In addition, "first", "second", and the like in the above embodiments are for distinguishing the embodiments, and do not represent merits of the embodiments.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The algorithms and displays presented herein are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. Moreover, the present invention is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.
In addition, the memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A knowledge graph-based node recommendation method, the method comprising:
matching related nodes in the knowledge graph according to the query content of the user;
respectively calculating the centrality of each node according to the in-degree relation of the nodes in the knowledge graph;
and determining the sequence of recommending the nodes to the user according to the centrality of the nodes.
2. The method of claim 1, wherein matching relevant nodes in the knowledge-graph based on the query content of the user comprises:
determining query types according to query contents of users, wherein the query types are types contained in nodes of the knowledge graph;
matching nodes in the knowledge-graph having the query class.
3. The method according to claim 1 or 2, wherein the step of calculating the centrality of each node according to the in-degree relation of the node in the knowledge graph comprises the following steps:
acquiring the relation weight of corresponding edges in the knowledge graph based on the in-degree relation of the nodes;
and calculating the centrality of each node by using the relation weight and the entrance degree relation.
4. The method of claim 3, wherein calculating the centrality of each node using the relationship weights and the in-degree relationship comprises:
carrying out weighted summation on the degree-of-entry relationship of the nodes according to the relationship weight to obtain the weight of the degree-of-entry relationship of the nodes;
and carrying out normalization processing on the entry relation weight of the nodes according to the number of the nodes and the nodes with the entry relation to obtain the centrality of the nodes.
5. The method of claim 4, wherein the centrality of each node is calculated according to the in-degree relationship of the node in the knowledge-graph, further comprising:
judging whether the nodes are positioned in the same sub-graph in the knowledge graph;
if the nodes are located in the same subgraph, normalizing the entry relation weight of each node according to each node and the number of the nodes with the entry relation to obtain the centrality of each node;
and if the nodes are positioned in different subgraphs, taking the specific weight of the subgraph to which the nodes belong in the knowledge graph as the weight, and carrying out weighted calculation on the centrality of each node to determine the centrality of each node relative to the knowledge graph.
6. The method of claim 5, further comprising:
calculating the centrality of each node through a formula, wherein the formula is as follows:
wherein weighted _ index represents the weight of the edge corresponding to the in-degree relation; total _ weighted _ index represents the total weight of the edge corresponding to the degree relation between each node and the same subgraph; m represents the number of nodes having an in-degree relationship with the calculated nodes; n represents the number of centrality nodes needing to be calculated in the same subgraph; v represents the number of nodes in the subgraph; v represents the number of nodes in the knowledge-graph.
7. An apparatus for node recommendation based on a knowledge-graph, the apparatus comprising:
the matching unit is used for matching related nodes in the knowledge graph according to the query content of the user;
the calculating unit is used for respectively calculating the centrality of each node obtained by the matching unit according to the in-degree relation of the node in the knowledge graph;
and the recommending unit is used for determining the order of recommending the nodes to the user according to the centrality of the nodes obtained by the calculating unit.
8. The apparatus of claim 7, wherein the matching unit comprises:
the determining module is used for determining query types according to query contents of users, wherein the query types are types contained in nodes of the knowledge graph;
a matching module for matching nodes in the knowledge-graph having the query class determined by the determining module.
9. A processor, configured to execute a program, wherein the program when executed performs the method for node recommendation based on a knowledge-graph of any one of claims 1-6.
10. A storage medium storing a computer program, wherein the computer program is operable to control an apparatus in which the storage medium is located to perform the method for node recommendation based on knowledge-graph according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010959476.3A CN112100400A (en) | 2020-09-14 | 2020-09-14 | Node recommendation method and device based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010959476.3A CN112100400A (en) | 2020-09-14 | 2020-09-14 | Node recommendation method and device based on knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112100400A true CN112100400A (en) | 2020-12-18 |
Family
ID=73752140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010959476.3A Pending CN112100400A (en) | 2020-09-14 | 2020-09-14 | Node recommendation method and device based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112100400A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112861034A (en) * | 2021-02-04 | 2021-05-28 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for detecting information |
| WO2023226311A1 (en) * | 2022-05-23 | 2023-11-30 | 华为云计算技术有限公司 | Data asset graph management method and related device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150045281A (en) * | 2013-10-18 | 2015-04-28 | 경희대학교 산학협력단 | Recommender system and method for trust-aware recommender system |
| US20150178775A1 (en) * | 2013-12-23 | 2015-06-25 | Yahoo! Inc. | Recommending search bid phrases for monetization of short text documents |
| US20180253653A1 (en) * | 2017-03-03 | 2018-09-06 | International Business Machines Corporation | Rich entities for knowledge bases |
| US20180330248A1 (en) * | 2017-05-12 | 2018-11-15 | Adobe Systems Incorporated | Context-aware recommendation system for analysts |
| CN109710621A (en) * | 2019-01-16 | 2019-05-03 | 福州大学 | In conjunction with the keyword search KSANEW algorithm of semantic category node and side right weight |
| CN110457573A (en) * | 2019-07-04 | 2019-11-15 | 平安科技（深圳）有限公司 | Products Show method, apparatus, computer equipment and storage medium |
-
2020
- 2020-09-14 CN CN202010959476.3A patent/CN112100400A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150045281A (en) * | 2013-10-18 | 2015-04-28 | 경희대학교 산학협력단 | Recommender system and method for trust-aware recommender system |
| US20150178775A1 (en) * | 2013-12-23 | 2015-06-25 | Yahoo! Inc. | Recommending search bid phrases for monetization of short text documents |
| US20180253653A1 (en) * | 2017-03-03 | 2018-09-06 | International Business Machines Corporation | Rich entities for knowledge bases |
| US20180330248A1 (en) * | 2017-05-12 | 2018-11-15 | Adobe Systems Incorporated | Context-aware recommendation system for analysts |
| CN109710621A (en) * | 2019-01-16 | 2019-05-03 | 福州大学 | In conjunction with the keyword search KSANEW algorithm of semantic category node and side right weight |
| CN110457573A (en) * | 2019-07-04 | 2019-11-15 | 平安科技（深圳）有限公司 | Products Show method, apparatus, computer equipment and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112861034A (en) * | 2021-02-04 | 2021-05-28 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for detecting information |
| CN112861034B (en) * | 2021-02-04 | 2023-08-15 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for detecting information |
| WO2023226311A1 (en) * | 2022-05-23 | 2023-11-30 | 华为云计算技术有限公司 | Data asset graph management method and related device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11354587B2 (en) | Systems and methods for organizing and finding data | |
| US9542477B2 (en) | Method of automated discovery of topics relatedness | |
| US10410138B2 (en) | System and method for automatic generation of features from datasets for use in an automated machine learning process | |
| Wu et al. | A multiple attribute group decision making framework for the evaluation of lean practices at logistics distribution centers | |
| US11182364B2 (en) | Data analysis support apparatus and data analysis support method | |
| KR102249466B1 (en) | Data catalog providing method and system for providing recommendation information using artificial intelligence recommendation model | |
| CN111143685B (en) | Commodity recommendation method and device | |
| CN112100400A (en) | Node recommendation method and device based on knowledge graph | |
| Thomas et al. | Applications of decision trees | |
| Jalali | Graph-based process mining | |
| CN110472742B (en) | Model variable determination method, device and equipment | |
| CN111797995B (en) | Method and device for generating interpretation report of model prediction sample | |
| Bernard et al. | Discovering customer journeys from evidence: a genetic approach inspired by process mining | |
| EP2541409B1 (en) | Parallelization of large scale data clustering analytics | |
| CN112465012A (en) | Machine learning modeling method and device, electronic equipment and readable storage medium | |
| US20130124484A1 (en) | Persistent flow apparatus to transform metrics packages received from wireless devices into a data store suitable for mobile communication network analysis by visualization | |
| CN111708945A (en) | Product recommendation method and device, electronic equipment and computer storage medium | |
| CN115422346A (en) | Knowledge graph-based multi-turn topic recommendation method | |
| CN110781211A (en) | Data analysis method and device | |
| CN110019771B (en) | Text processing method and device | |
| JP2020166443A (en) | Data processing method recommendation system, data processing method recommendation method, and data processing method recommendation program | |
| Atagün et al. | A comparison of data mining tools and classification algorithms: Content producers on the video sharing platform | |
| CN116561134B (en) | Business rule processing method, device, equipment and storage medium | |
| Yankov et al. | Digital advantages for the construction industry | |
| Silva et al. | Exploring Dataset Patterns for New Demand Response Participants Classification |
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 |