CN111460168B - Knowledge graph verification and updating method based on block chain distributed double consensus - Google Patents
Knowledge graph verification and updating method based on block chain distributed double consensus Download PDFInfo
- Publication number
- CN111460168B CN111460168B CN202010228791.9A CN202010228791A CN111460168B CN 111460168 B CN111460168 B CN 111460168B CN 202010228791 A CN202010228791 A CN 202010228791A CN 111460168 B CN111460168 B CN 111460168B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- machine
- machine node
- node
- nodes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
- H04L9/50—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols using hash chains, e.g. blockchains or hash trees
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computing Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a knowledge graph verification and updating method based on block chain distributed double consensus, which comprises the following steps: all machine nodes i carry out local newly-added knowledge mining, and are matched with the latest knowledge graph spectrum of the original block chain to find out different knowledge to form a knowledge list; all machine nodes share the knowledge list to other machine nodes to form a total task table Si(ii) a The host machine node connects the local node SpThe entity relationship pairs k in the machine node p are sequentially broadcasted to other machine nodes and used for verifying whether the entity relationship pairs k corresponding to each number n of the main machine node p are provided by the machine node i or not; after the main machine node p receives f +1 Confirm messages, the agreed SpThe entity relation in the system initiates local verification on k; and when the host machine node p receives the messages of more than f +1 local-complete, updating the latest knowledge graph, performing virtual currency reward on the providing node i of the entity relation pair k, and writing the newly added knowledge graph and the transaction into a block chain.
Description
Technical Field
The invention relates to the technical field of combination of a block chain and a knowledge graph, in particular to a knowledge graph verification and updating method based on block chain distributed double consensus.
Background
The essence of the knowledge graph is a knowledge base based on semantic network, which is mainly used to describe concepts, entities, events and their relationships, and is usually stored in the form of triples. With the development of science and technology, the application of knowledge maps is more and more extensive, for example, decision assistance is provided for enterprises, potential customers are positioned for markets, domain knowledge is updated in real time for new people in the field to learn, and the like. Therefore, various industries put high demands on the reliability and accuracy of knowledge maps. When new knowledge is introduced into the knowledge map, it is determined whether the newly added knowledge is correct? Is there a conflict with the original knowledge? The reliability, authority and continuous expansibility of the knowledge graph become main bottlenecks limiting the rapid development and landing of the knowledge graph.
In the automatic construction process of the knowledge graph at the present stage, a large amount of knowledge noise and knowledge conflict are often introduced, so that the reliability of the knowledge graph is poor. Meanwhile, due to the fact that the quality levels of the data sources are different, a high-quality knowledge graph is difficult to dig out. Secondly, constructing a knowledge graph in the vertical field requires high cost and expense due to factors such as limited data sources, high industry barriers, real-time change of data content and business content, and the like, so as to construct a high-quality industry knowledge graph. Therefore, how to arouse that each industry actively provides a high-quality data source, excavate high-quality knowledge, and construct an authoritative knowledge map is a problem which needs to be solved urgently.
The characteristics of decentralization, traceability, non-tamper property, incentive mechanism and the like of the block chain provide new possibility and direction for constructing a reliable knowledge map and bring new thought and economic value for knowledge sharing. Blockchains are distributed ledgers that are primarily used to solve trust and consistency issues between multiple principals. The distributed consensus mechanism of the blockchain is mainly used for solving the problem of distributed consistency, namely, data is confirmed, verified and updated. Therefore, the block structure with the time stamp, the consensus mechanism of the distributed nodes and the reward mechanism based on the consensus algorithm provide a new idea for verification and updating of the knowledge graph.
Disclosure of Invention
Aiming at the problems of poor reliability, uneven knowledge quality level, high cost, high barrier and the like of the current knowledge graph for constructing the knowledge graph in the vertical field, the invention provides a method for verifying and updating the knowledge graph based on block chain distributed double consensus.
In order to achieve the purpose, the invention adopts the following technical scheme:
a knowledge graph verification and updating method based on block chain distributed double consensus comprises the following specific steps:
step 1: all distributed machine nodes i participating in verification conduct newly added knowledge mining on logs, articles and webpages in the professional field of the machine nodes to form a local knowledge base; the knowledge source diversity is ensured, and meanwhile, the accuracy of knowledge verification in the step 4 is ensured due to the irrelevance of the knowledge mining method; the information stored by each block within the chain of blocks includes: the latest knowledge graph stored in the form of RDF, information of reward transaction and time stamp, wherein RDF is a resource description framework;
step 2, after all the machine nodes complete the newly added knowledge mining, each machine node downloads the knowledge graph in the latest block on the block chain, namely the latest knowledge graph built currently is subjected to entity matching, namely the local knowledge base formed by the machine node i in the step 1 is compared with the knowledge downloaded on the block chain, and different knowledge, namely newly added knowledge and conflict knowledge, is formed into a knowledge list Li,i＝A,B,C,D...；LiThe method comprises the following steps: the RDF form entity relationship pair and the generation time corresponding to each entity relationship pair; each entity relationship pair is corresponding knowledge;
step 3, forming a knowledge list L by all machine nodesiThen, a first recognition mechanism, namely a recognition phase of the knowledge list, is entered, and the recognition mechanism involves 5 phases:
s1: a list sharing stage; each machine node stores the knowledge list L of the machine nodeiWith { Share, LiThe message form of the node is broadcasted to other machine nodes, so that each machine node is provided with a knowledge list L provided by all the machine nodesiThe knowledge lists L areiSummarizing to form a total task list LTAlso included are self-formed lists;
s2, preparation phase; after each machine node receives R-1 list sharing messages, entering a preparation stage, wherein R is the number of the machine nodes participating in knowledge graph verification; in the preparation stage, each machine node collects the machine nodes to form a total task list LTScreening, matching, duplicate removal and sequencing one by one; finding out the same entity relationship pair, reserving a group of earliest entity relationship pairs according to the time sequence of the time stamp, and deleting other repeated relationships to ensure that the machine node i is the machine node providing the earliest entity relationship pair; finally, each machine node obtains a group of task tables S which have no repeated relation, are sorted in time sequence and contain the knowledge provided by other nodesi(ii) a Task list SiThe method comprises the following steps: the method comprises the following steps that entity relationship pairs, machine providing nodes corresponding to each entity relationship pair and generation time of each entity relationship pair are obtained;
s3, starting; when all machine nodes form a task table SiThen, entering a starting stage; selecting a main machine node p, wherein the selection mode of the main machine node p is a polling mechanism, and when a knowledge graph is constructed in the v-th round, p is vmod | R |; host machine node p connects local machine node SpNumbering n by each knowledge in the database from early to late according to the timestamp, wherein n is a natural number which is continuously increased, and broadcasting messages in the form of { Start, number n, entity relationship pair k, timestamp t and knowledge-providing machine node i } to other machine nodes;
s4, list verification phase; the other auxiliary machine nodes verify the received broadcast message Start; the entity relation pair k in the broadcast message Start message, the timestamp t, the machine node i providing knowledge and a task table S formed by the local machine node in the preparation stageiCarrying out comparison and verification; if the broadcast message Start (k, t, i) is associated with the task table S in the local machine nodeiIf the (k, t and i) in the information is consistent, the auxiliary machine node sends a message { Verified, the number n, the entity relation pair k, the timestamp t and the machine node i providing knowledge } to all other machine nodes; this is achieved byThe stage is mainly used for verifying whether a provider of an entity relationship pair k corresponding to each number n of a main machine node p is a machine node i;
s5, list confirmation; after the auxiliary machine node receives 2f +1 Verified messages, entering a list confirmation stage, wherein the stage lays a foundation for the step 4, wherein f is the number of the problem machine nodes, and the number R of the main machine nodes required to participate in verification is more than 3f + 1; at this stage, all the machine nodes receiving 2f +1 Verified messages send messages { Confirm, number n, entity relationship pair k, timestamp t, machine node i providing knowledge } to the main machine node p; after the host machine node p receives f +1 Confirm signals, the entity relationship pair k enters a knowledge verification stage, namely a second recognition mechanism stage;
step 4, all auxiliary machine nodes immediately carry out consensus verification on the entity relationship in the knowledge list which achieves the authentication in the step 3, namely, the local knowledge base formed in the step 1 carries out comparison verification on the entity relationship in the knowledge list of the main machine node p; this step is a second consensus mechanism, which is divided into 5 steps:
s1, locally verifying the entity relationship pair k with the number n by all the auxiliary machine node pairs; the method comprises the following steps that a main machine node p broadcasts a message { local-verified, an entity relation pair k, a timestamp t and a knowledge providing machine node i } to all auxiliary machine nodes, and the auxiliary machine nodes locally verify the entity relation pair k;
s2, if the entity relation pair k is consistent with the entity relation pair stored in the local knowledge base, the entity relation pair k passes the verification of the auxiliary machine node, and the auxiliary machine node sends a message { local-confirmed, the entity relation pair k, a timestamp t and a machine node i providing knowledge to all machine nodes including the main machine node p;
s3, all auxiliary machine nodes count the local-confirmed messages, and if 2f +1 local-confirmed messages are received, the messages { local-complete, entity relation pair k, timestamp t, machine node i providing knowledge } are sent to the main machine node p;
s4, the main machine node p counts local-complete messages for the entity relation pair k, when more than f +1 messages are received, the knowledge graph downloaded in the step 2 is updated, the entity relation pair k is updated and written into the knowledge graph, and virtual currency reward is carried out on the machine node i providing knowledge;
s5, the host machine node p writes the updated knowledge graph and the reward transaction information into a new block chain;
the host machine node p completes the knowledge list S of the host machine node ppAfter all entity relationship pairs k are verified, the knowledge graph verification and updating of the current round is completed, then the step 1 is skipped to, the next round of the knowledge graph verification and updating process is started, and the main machine node of the next round is replaced immediately.
Based on the flow, a reliable and credible authoritative knowledge map can be obtained through a double verification mechanism of the block chain, and the knowledge map realizes the functions of automatic verification, updating, completion, conflict removal and the like. Because knowledge comes from different data sources and different mining algorithms are equivalent to the realization of a completely irrelevant extraction mode for the same nouns, the extraction accuracy is greatly improved; meanwhile, the verification method of the common knowledge graph at the present stage comprises the following steps: compared with the verification based on confidence coefficient and the manual verification based on crowdsourcing, the verification method based on the block chain combines the advantages of the two, realizes the participation of multiple people, and has the characteristics of low cost, automatic verification, high confidence coefficient and the like.
The invention not only solves the problems of credibility of the knowledge map and time-efficiency traceability of the knowledge map, but also protects the value of knowledge contributors to the maximum extent and stimulates the contribution of more ginseng and high-quality knowledge because a reward mechanism of virtual currency is added.
Drawings
Fig. 1 is a knowledge graph of the latest blocks of a downloaded block chain for each machine node.
FIG. 2 is a process of knowledge list consensus.
Fig. 3 shows a new knowledge graph and a transaction information write block chain.
Detailed Description
The invention is described in more detail below with reference to the accompanying drawings and the construction and verification of a knowledge graph of power defects in the power field as an example.
Step 1: mining newly added knowledge of all distributed machine nodes i participating in verification aiming at the power defect logs, power defect field documents and power field webpages of the machine nodes to form a local knowledge base; the knowledge source diversity is ensured, and meanwhile, the accuracy of knowledge verification in the step 4 is ensured due to the irrelevance of the knowledge mining method; each block within the block chain stores information including: the latest knowledge graph stored in the form of RDF, information of reward transaction and time stamp, wherein RDF is a resource description framework;
step 2, as shown in the flow of fig. 1, after all the machine nodes complete the newly added knowledge mining, each machine node downloads the knowledge graph in the latest block on the block chain, namely the latest knowledge graph built currently is subjected to entity matching, namely the local knowledge base formed in the step 1 by the machine node i is compared with the knowledge downloaded on the block chain, and different knowledge, namely newly added knowledge and conflict knowledge, is formed into a knowledge list LiI ═ a, B, C, d.; as shown in Table 1, LiThe method comprises the following steps: the RDF form entity relationship pair and the generation time corresponding to each entity relationship pair; each entity relationship pair is corresponding knowledge;
TABLE 1
Step 3, forming a knowledge list L by all machine nodesiThen, enter the first recognition mechanism, i.e. the recognition phase of the knowledge list, which involves 5 phases, and the flow is shown in fig. 2:
s1: a list sharing stage; each machine node stores the knowledge list L of the machine nodeiWith { Share, LiThe message form of the node is broadcasted to other machine nodes, so that each machine node is provided with a knowledge list L provided by all the machine nodesiThe knowledge lists L areiSummarizing to form a total task list LTAlso included are self-formed lists;
s2 preparationA stage; after each machine node receives R-1 list sharing messages, entering a preparation stage, wherein R is the number of the machine nodes participating in knowledge graph verification; in the preparation stage, each machine node collects the machine nodes to form a total task list LTScreening, matching, duplicate removal and sequencing one by one; finding out the same entity relationship pair, reserving a group of earliest entity relationship pairs according to the time sequence of the time stamp, and deleting other repeated relationships to ensure that the machine node i is the machine node providing the earliest entity relationship pair; finally, each machine node obtains a group of task tables S which have no repeated relation, are sorted in time sequence and contain the knowledge provided by other nodesi(ii) a As shown in Table 2, task Table SiThe method comprises the following steps: the method comprises the following steps that entity relationship pairs, machine providing nodes corresponding to each entity relationship pair and generation time of each entity relationship pair are obtained;
TABLE 2
S3, starting; when all machine nodes form a task table SiThen, entering a starting stage; selecting a main machine node p, wherein the selection mode of the main machine node p is a polling mechanism, and when a knowledge graph is constructed in the v-th round, p is vmod | R |; host machine node p connects local machine node SpNumbering n by each knowledge in the database from early to late according to the timestamp, wherein n is a natural number which is continuously increased, and broadcasting messages in the form of { Start, number n, entity relationship pair k, timestamp t and knowledge-providing machine node i } to other machine nodes;
s4, list verification phase; the other auxiliary machine nodes verify the received broadcast message Start; the entity relation pair k in the broadcast message Start message, the timestamp t, the machine node i providing knowledge and a task table S formed by the local machine node in the preparation stageiCarrying out comparison and verification; if the broadcast message Start (k, t, i) is associated with the task table S in the local machine nodeiIf the (k, t, i) in the message is consistent, the slave node sends a message { Verified, number n, entity relationship pairk, timestamp t, machine node i providing knowledge to all other machine nodes; the stage is mainly used for verifying whether the provider of the entity relationship pair k corresponding to each number n of the main machine node p is a machine node i;
s5, list confirmation; after the auxiliary machine node receives 2f +1 Verified messages, entering a list confirmation stage, wherein the stage lays a foundation for the step 4, wherein f is the number of the problem machine nodes, and the number R of the main machine nodes required to participate in verification is more than 3f + 1; at this stage, all the slave nodes receiving 2f +1 Verified messages send messages { Confirm, number n, entity relationship pair k, timestamp t, knowledge-providing machine node i } to the master node p; after the host machine node p receives f +1 Confirm signals, the entity relationship pair k enters a knowledge verification stage, namely a second recognition mechanism stage;
step 4, all auxiliary machine nodes immediately carry out consensus verification on the entity relationship in the knowledge list which achieves the authentication in the step 3, namely, the local knowledge base formed in the step 1 carries out comparison verification on the entity relationship in the knowledge list of the main machine node p; this step is a second consensus mechanism, which is divided into 5 steps:
s1, locally verifying the entity relationship pair k with the number n by all the auxiliary machine node pairs; the method comprises the following steps that a main machine node p broadcasts a message { local-verified, an entity relation pair k, a timestamp t and a knowledge providing machine node i } to all auxiliary machine nodes, and the auxiliary machine nodes locally verify the entity relation pair k;
s2, if the entity relation pair k is consistent with the entity relation pair stored in the local knowledge base, the entity relation pair k passes the verification of the auxiliary machine node, and the auxiliary machine node sends a message { local-confirmed, the entity relation pair k, a timestamp t and a machine node i providing knowledge to all machine nodes including the main machine node p;
s3, all auxiliary machine nodes count the local-confirmed messages, and if 2f +1 local-confirmed messages are received, the messages { local-complete, entity relation pair k, timestamp t, machine node i providing knowledge } are sent to the main machine node p;
s4, the main machine node p counts local-complete messages for the entity relation pair k, when more than f +1 messages are received, the knowledge graph downloaded in the step 2 is updated, the entity relation pair k is updated and written into the knowledge graph, and virtual currency reward is carried out on the machine node i providing knowledge;
s5, the host machine node p writes the updated knowledge graph and the reward transaction information into a new block chain;
the host machine node p completes the knowledge list S of the host machine node ppAfter all entity relationship pairs k are verified, the knowledge graph verification and updating of the current round is completed, then the step 1 is skipped to, the next round of the knowledge graph verification and updating process is started, and the main machine node of the next round is replaced immediately.
Claims (1)
1. A knowledge graph verification and updating method based on block chain distributed dual consensus is characterized in that: the method comprises the following specific steps:
step 1: all distributed machine nodes i participating in verification conduct newly added knowledge mining on logs, articles and webpages in the professional field of the machine nodes to form a local knowledge base; the knowledge source diversity is ensured, and meanwhile, the accuracy of knowledge verification in the step 4 is ensured due to the irrelevance of the knowledge mining method; the information stored by each block within the chain of blocks includes: the latest knowledge graph stored in the form of RDF, information of reward transaction and time stamp, wherein RDF is a resource description framework;
step 2: after all machine nodes complete newly added knowledge mining, each machine node downloads a knowledge graph in the latest block on the block chain, namely the latest knowledge graph constructed currently is subjected to entity matching, namely the local knowledge base formed by the machine node i in the step 1 is compared with the knowledge downloaded on the block chain, different knowledge, namely newly added knowledge and conflict knowledge, is formed into a knowledge list Li，i＝A，B，C，D...；LiThe method comprises the following steps: the RDF form entity relationship pair and the generation time corresponding to each entity relationship pair; each entity relationship pair is corresponding knowledge;
and step 3: when all the machine nodes form the knowledge listSingle LiThen, a first recognition mechanism, namely a recognition phase of the knowledge list, is entered, and the recognition mechanism involves 5 phases:
s1: a list sharing stage; each machine node stores the knowledge list L of the machine nodeiWith { Share, LiThe message form of the node is broadcasted to other machine nodes, so that each machine node is provided with a knowledge list L provided by all the machine nodesiThe knowledge lists L areiSummarizing to form a total task list LTAlso included are self-formed lists;
s2: a preparation stage; after each machine node receives R-1 list sharing messages, entering a preparation stage, wherein R is the number of the machine nodes participating in knowledge graph verification; in the preparation stage, each machine node collects the machine nodes to form a total task list LTScreening, matching, duplicate removal and sequencing one by one; finding out the same entity relationship pair, reserving a group of earliest entity relationship pairs according to the time sequence of the time stamp, and deleting other repeated relationships to ensure that the machine node i is the machine node providing the earliest entity relationship pair; finally, each machine node obtains a group of task tables S which have no repeated relation, are sorted in time sequence and contain the knowledge provided by other nodesi(ii) a Task list SiThe method comprises the following steps: the method comprises the following steps that entity relationship pairs, machine providing nodes corresponding to each entity relationship pair and generation time of each entity relationship pair are obtained;
s3: a start phase; when all machine nodes form a task table SiThen, entering a starting stage; selecting a main machine node p, wherein the selection mode of the main machine node p is a polling mechanism, and when a v-th wheel constructs a knowledge graph, p is v mod | R |; the main machine node p lists the tasks S of the local machine nodepNumbering n by each knowledge in the database from early to late according to the timestamp, wherein n is a natural number which is continuously increased, and broadcasting messages in the form of { Start, number n, entity relationship pair k, timestamp t and knowledge-providing machine node i } to other machine nodes;
s4: a list verification stage; the other auxiliary machine nodes verify the received broadcast message Start message; general will be widePlaying entity relation pair k in message Start message, time stamp t, knowledge providing machine node i and task table S formed by local machine node in preparation stageiCarrying out comparison and verification; if the broadcast message Start (k, t, i) is associated with the task table S in the local machine nodeiIf the (k, t and i) in the information is consistent, the auxiliary machine node sends a message { Verified, the number n, the entity relation pair k, the timestamp t and the machine node i providing knowledge } to all other machine nodes; the stage is mainly used for verifying whether the provider of the entity relationship pair k corresponding to each number n of the main machine node p is a machine node i;
s5: a list confirmation stage; after the auxiliary machine node receives 2f +1 Verified messages, entering a list confirmation stage, wherein the stage lays a foundation for the step 4, wherein f is the number of the problem machine nodes, and the number R of the main machine nodes required to participate in verification is larger than 3f + 1; at this stage, all the machine nodes receiving 2f +1 Verified messages send messages { Confirm, number n, entity relationship pair k, timestamp t, machine node i providing knowledge } to the main machine node p; after the host machine node p receives f +1 Confirm signals, the entity relationship pair k enters a knowledge verification stage, namely a second recognition mechanism stage;
and 4, step 4: all the auxiliary machine nodes immediately carry out consensus verification on the k according to the entity relationship in the knowledge list which achieves the authentication in the step 3, namely, the k is compared and verified according to the entity relationship in the knowledge list of the main machine node p in the local knowledge base formed in the step 1; this step is a second consensus mechanism, which is divided into 5 steps:
s1: all the auxiliary machine node pairs carry out local verification on the entity relationship with the number n; the method comprises the following steps that a main machine node p broadcasts a message { local-verified, an entity relation pair k, a timestamp t and a knowledge providing machine node i } to all auxiliary machine nodes, and the auxiliary machine nodes locally verify the entity relation pair k;
s2: if the entity relationship pair k is consistent with the entity relationship pair stored in the local knowledge base, the entity relationship pair k passes the verification of the auxiliary machine node, and the auxiliary machine node sends a message { local-confirmed, the entity relationship pair k, a timestamp t and a machine node i providing knowledge } to all machine nodes including the main machine node p;
s3: all the auxiliary machine nodes count the local-confirmed messages, and if 2f +1 local-confirmed messages are received, the messages { local-complete, entity relation pair k, timestamp t and knowledge providing machine nodes i } are sent to the main machine node p;
s4: the main machine node p counts local-complete messages aiming at the entity relation pair k, when more than f +1 messages are received, the knowledge graph downloaded in the step 2 is updated, the entity relation pair k is updated and written into the knowledge graph, and virtual currency reward is carried out on a machine node i providing knowledge;
s5: writing the updated knowledge graph and the reward transaction information into a new block chain by the main machine node p;
the host machine node p completes its task table SpAfter all entity relations in the node k are verified, the knowledge graph verification and updating of the current round is completed, then the step 1 is skipped to, the next round of the knowledge graph verification and updating process is started, and the host machine node of the next round is replaced immediately;
based on the steps, a reliable and credible authoritative knowledge map is obtained through a double verification mechanism of the block chain, and the knowledge map realizes the functions of automatic verification, updating, completion and conflict removal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010228791.9A CN111460168B (en) | 2020-03-27 | 2020-03-27 | Knowledge graph verification and updating method based on block chain distributed double consensus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010228791.9A CN111460168B (en) | 2020-03-27 | 2020-03-27 | Knowledge graph verification and updating method based on block chain distributed double consensus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111460168A CN111460168A (en) | 2020-07-28 |
| CN111460168B true CN111460168B (en) | 2021-06-22 |
Family
ID=71680218
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010228791.9A Active CN111460168B (en) | 2020-03-27 | 2020-03-27 | Knowledge graph verification and updating method based on block chain distributed double consensus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111460168B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783653A (en) * | 2019-02-16 | 2019-05-21 | 南京樯图数据科技有限公司 | A kind of inquiry system of management and the retrospect of the knowledge mapping based on block chain technology |
| CN110069637A (en) * | 2019-03-01 | 2019-07-30 | 安徽省优质采科技发展有限责任公司 | Online way of invitation for bid and system based on block chain |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10721069B2 (en) * | 2018-08-18 | 2020-07-21 | Eygs Llp | Methods and systems for enhancing privacy and efficiency on distributed ledger-based networks |
| CN110750698A (en) * | 2019-09-09 | 2020-02-04 | 深圳壹账通智能科技有限公司 | Knowledge graph construction method and device, computer equipment and storage medium |
| CN110597992B (en) * | 2019-09-10 | 2023-08-29 | 腾讯科技（深圳）有限公司 | Semantic reasoning method and device based on knowledge graph and electronic equipment |
-
2020
- 2020-03-27 CN CN202010228791.9A patent/CN111460168B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109783653A (en) * | 2019-02-16 | 2019-05-21 | 南京樯图数据科技有限公司 | A kind of inquiry system of management and the retrospect of the knowledge mapping based on block chain technology |
| CN110069637A (en) * | 2019-03-01 | 2019-07-30 | 安徽省优质采科技发展有限责任公司 | Online way of invitation for bid and system based on block chain |
Non-Patent Citations (1)
| Title |
|---|
| 国外区块链研究主题及展望;程晨等;《电子政务》;20181231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111460168A (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Catalini | Microgeography and the direction of inventive activity | |
| CN110570202B (en) | Hybrid consensus method based on fragmentation technology | |
| US8719222B2 (en) | Synchronization and collaboration within peer-to-peer and client/server environments | |
| CN111082943B (en) | Efficient block chain consensus method | |
| CN112395640A (en) | Industry Internet of things data lightweight credible sharing technology based on block chain | |
| US11610173B2 (en) | Intelligent collaborative project management | |
| JP2018511851A (en) | Information transmission method and apparatus | |
| CN107659467B (en) | Community detection method based on alliance formation game in dynamic social network | |
| CN111008521B (en) | Method, device and computer storage medium for generating wide table | |
| Stejskal et al. | Modelling collaboration and innovation in creative industries using fuzzy set qualitative comparative analysis | |
| CN112100282B (en) | Data synchronization method and device of alliance chain, readable medium and electronic equipment | |
| CN106992966B (en) | Network information transmission implementation method for true and false messages | |
| CN101060428A (en) | A P2P network dispatching method | |
| CN111460168B (en) | Knowledge graph verification and updating method based on block chain distributed double consensus | |
| CN114491615A (en) | Asynchronous longitudinal federal learning fair incentive mechanism method based on block chain | |
| Inoue et al. | Analysis of cooperative research and development networks on Japanese patents | |
| CN115633035B (en) | Improved PBFT (physical layer transmission) based block chain consensus algorithm for Internet of things | |
| CN109410078A (en) | A kind of information propagation prediction method for the mobile social networking shared suitable for object oriented file | |
| CN106776756A (en) | The method and apparatus of server end, client process weather resource data | |
| CN114372533A (en) | Hybrid table arrangement data processing method and device for matching activities | |
| Hoekman et al. | Trade and value chain participation: Domestic firms and FDI spillovers in Africa | |
| Zhou et al. | Coevolution of synchronization and cooperation in real networks | |
| CN115002111B (en) | Block chain consensus method based on group tree structure | |
| CN117114886B (en) | Block chain carbon transaction method and system based on double-layer consensus mechanism | |
| AHMAD | Nigerian Agricultural Posts on Facebook and Instagram within the West African Agricultural Messaging Framework |
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 |