CN111581534B - Rumor propagation tree structure optimization method based on consistency of vertical place - Google Patents
Rumor propagation tree structure optimization method based on consistency of vertical place Download PDFInfo
- Publication number
- CN111581534B CN111581534B CN202010438369.6A CN202010438369A CN111581534B CN 111581534 B CN111581534 B CN 111581534B CN 202010438369 A CN202010438369 A CN 202010438369A CN 111581534 B CN111581534 B CN 111581534B
- Authority
- CN
- China
- Prior art keywords
- node
- rumor
- child
- propagation tree
- child node
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 238000010586 diagram Methods 0.000 claims description 16
- 230000004044 response Effects 0.000 claims description 5
- 230000008451 emotion Effects 0.000 claims description 3
- 230000007480 spreading Effects 0.000 claims description 2
- 238000003062 neural network model Methods 0.000 abstract description 4
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005295 random walk Methods 0.000 description 1
- 230000000306 recurrent 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of authenticity of social media information release, and particularly relates to a rumor propagation tree structure optimization method based on the consistency of the legislation. Aiming at the scene of rumor detection of a top-down recursion neural network model based on a tree structure in social media, the invention enables the nodes with the same vertical field in the rumor propagation tree structure to be combined into a super node, strengthens the vertical field representation of the node to the father node of the node, optimizes the propagation tree structure and improves the rumor detection performance. The invention can greatly reduce the number of branches and nodes in the propagation tree, thereby reducing the computation time complexity of rumor detection and realizing optimization on the performance of rumor detection.
Description
Technical Field
The invention belongs to the technical field of authenticity of social media information release, and particularly relates to a rumor propagation tree structure optimization method based on the consistency of the legislation.
Background
Rumors are generally defined as information that appears and propagates between people whose genuine value has not been confirmed or intentionally faked. In recent years, the popularity of social media over Twitter, facebook, etc. has further facilitated rumor dissemination by enabling unreliable sources to disseminate large amounts of unverified information between people. Therefore, real-time tracking and uncovering rumors becomes especially important. However, the conventional feature-based method extracts features from the statistics of the spurious message, the author of the message, and their responses, and forms a flat feature vector for rumor detection. However, the method ignores the propagation structure of the message, so that the existing method has high complexity of computation time and low efficiency. Therefore, research into rumor transmission structural characteristics is receiving increasing attention from researchers.
Wu et al propose a hybrid SVM classifier that deeply combines radial basis function RBF with a random walk based graph kernel to detect rumors on the green microblog by capture plane and propagation mode. Ma et al captured the similarity of propagation trees by computing their similar substructures using the tree kernel to identify different types of rumors on Twitter. Ma et al also propose a recurrent neural network model based on bottom-up and top-down tree structures for representation learning and classification of rumors. However, current rumor detection studies based on rumor propagation structural features have not considered the problem of optimizing on the propagation structure of messages. In the rumor propagation mode tree diagram, the propagation tree structure optimization method based on the consistent ground uses the different ground of the parent node expression of each node to classify, the nodes with the same ground can implement the propagation tree optimization strategy to form a super node, thereby simplifying the propagation tree structure.
Disclosure of Invention
The invention aims to provide a rumor propagation tree structure optimization method based on the consistency of the stand, which is applied to a social media rumor detection scene.
The purpose of the invention is realized by the following technical scheme: the method comprises the following steps:
step 1: inputting a rumor propagation tree diagram G;
and 2, step: base ofTraversing each node in the rumor propagation tree graph G in a breadth-first manner in a consistent manner to obtain an optimized rumor propagation tree graph G 1 ；
Step 2.1: selecting a node in the rumor propagation tree graph G, and judging whether the node has a child node or not; if the node has a child node, executing step 2.2; if the node has no child node, ending the operation on the node, and executing the step 2.4;
step 2.2: traversing each child node of the node, and acquiring the position of each child node;
step 2.3: judging whether the child nodes have the same position or not; if the child nodes have the same position, merging the child nodes with the same position into a supernode; if the child nodes do not have the same position, ending the operation on the node, and executing the step 2.4;
step 2.4: judging whether the traversal of all nodes in the rumor propagation tree graph G is finished or not; if the traversal of all nodes in the rumor propagation tree graph G is completed, outputting the optimized propagation tree graph G 1 (ii) a Otherwise, returning to the step 2.1;
and step 3: depth-first traversal propagation tree graph G based on consistency of ground 1 Obtaining the optimized propagation tree graph G by each node in the tree 2 ；
Step 3.1: selecting optimized rumor propagation tree graph G 1 The node in (4) determines whether the node has a child node. If the node has a child node, executing the step 3.2; if the node has no child node, ending the operation on the node, and executing the step 3.5;
step 3.2: selecting a child node of the node, and judging whether the child node has a next-level child node; if the child node has a next-level child node, executing step 3.3; if the child node does not have a next level child node, executing step 3.4;
step 3.3: checking whether the position of the next-level child node of the child node is supported; if the position of the child node at the next level of the child node is the support, combining the child node and the child node at the next level into a super node; otherwise, executing step 3.4;
step 3.4: judging whether the traversal of all child nodes of the node is completed or not; if the traversal of all child nodes of the node is finished, executing the step 3.5; otherwise, returning to the step 3.2;
step 3.5: judging whether the optimized rumor spreading tree diagram G is finished or not 1 Traversing all nodes in the tree; if the optimized rumor propagation tree diagram G is completed 1 Outputting the final optimized propagation tree graph G after traversing all the nodes 2 (ii) a Otherwise, the step 3.1 is returned.
The invention has the beneficial effects that:
the invention provides a rumor propagation tree structure optimization method based on the consistency of the vertical field aiming at the scene of rumor detection of a top-down recursion neural network model based on a tree structure in social media. The invention can greatly reduce the number of branches and nodes in the propagation tree, thereby reducing the complexity of calculation time of rumor detection and realizing optimization on the performance of rumor detection.
Drawings
Fig. 1 is a diagram of a rumor propagation tree in social media.
Fig. 2 is a schematic diagram of rumor propagation in social media optimized based on a place-consistent breadth-first propagation tree structure.
Fig. 3 is a schematic diagram of rumor propagation in social media optimized based on a position-consistent depth-first propagation tree structure.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
By taking fig. 1 as an example, the propagation tree optimization problem based on breadth and depth priority with consistent ground is introduced in sequence. Fig. 1 is a tree diagram of rumor propagation structure in a part of original social media, and fig. 2 and 3 are tree diagrams of rumor propagation structure of social media after being applied to propagation tree optimization methods based on breadth-first and depth-first, respectively. The invention aims to provide a propagation tree optimization algorithm based on consistency of a place under a social media rumor detection scene, which comprises a propagation tree optimization method based on breadth-first and a propagation tree optimization method based on depth-first, and can greatly reduce the number of branches and nodes in a propagation tree, thereby reducing the computation time complexity of rumor detection and realizing optimization on the performance of the rumor detection.
The purpose of the invention is realized as follows:
1. inputting a rumor propagation tree diagram G;
2. traversing each node in the propagation tree graph G in a priority mode based on the breadth with consistent ground to obtain the optimized propagation tree graph G 1 ；
3. Depth-first traversal propagation tree graph G based on consistency of ground 1 Obtaining the optimized propagation tree graph G at each node 2 ；
4. Returning to the finally optimized propagation tree graph G 2 I.e. graph G'.
The invention provides a rumor propagation tree structure optimization method based on the consistency of the vertical sites aiming at a rumor detection scene of a top-down recursive neural network model based on a tree structure in social media. The invention can greatly reduce the number of branches and nodes in the propagation tree, thereby reducing the complexity of calculation time of rumor detection and realizing optimization on the performance of rumor detection.
1. The invention relates to a directed graph with emotion labels: top-down rumor propagation tree plots. It naturally follows the direction of propagation of rumor information and defines the rumor detection dataset as a set of statements C = { C = } 1 ,C 2 ,…,C |c| A, wherein each declaration C i Corresponding to a source tweet, the tweet ideally consists of all the relevant response tweets arranged in time sequence. Thus, the propagation tree is denoted T (r) =<V i ,E i >In which V is i ＝C i Composed of all related posts as nodes, and E i Represents a set of directed links where u → v represents information that flows from u to v, which sees it and provides a response to u, i.e., an emotion label on a directed edge.
2. Breadth-first traversal of each node V in the graph i : and (1) judging whether the node has a child node or not. If not, then based on
Ending the breadth-first propagation tree optimization algorithm with consistent ground; (2) Traversing each child node of the node and obtaining the position of each child node, such as support, denial or question; and (3) judging whether the child nodes have the same position. If not, ending the propagation tree optimization algorithm based on the breadth-first with consistent position; (4) And executing a breadth-first propagation tree optimization method based on consistent position. As shown in fig. 1, when traversing to node 2, it is found that the positions of nodes 6 and 7 in its child nodes are the same, and they are both considered, so we can use our proposed optimization strategy based on the breadth-first propagation tree with consistent positions to merge the child nodes with consistent positions into a super node, and simplify the structure of the propagation tree. Figure 2 is a simplified rumor propagation tree diagram.
3. Each node V in the depth-first traversal graph i : and (1) judging whether the node has a child node or not. If not, finishing the depth-first propagation tree optimization algorithm based on the consistency of the position; (2) And traversing each child node of the node, and judging whether the child node has the child node. If not, finishing the propagation tree optimization algorithm based on depth priority; (3) see if their position is supported. If not, finishing the propagation tree optimization algorithm based on depth priority; if yes, combining the child node and the next-level child node into a super node (4), traversing the child node of the node, judging whether the child node has the child node, and simultaneously checking whether the ground of the child node is supported; (5) Step (4) is executed in a circulating way until a certain node has no child nodes or all the child nodes of the certain node are not supported from the standpoint; (6) And executing a depth-first propagation tree optimization strategy based on position consistency. Such asAs shown in fig. 2, when traversing to a super node, it is found that the node 9 supports the position of its parent node, and the node 11 supports the position of its parent node 9, so that we can find that the positions of the super node, the node 9 and the node 11 are identical, that is, the positions of the super node, the node 9 and the node 11 are identical, and whether they are all considered, so we can implement a depth-first propagation tree optimization strategy based on the position agreement. Fig. 3 is a diagram of an optimized rumor propagation tree structure.
4. Thus, an optimized structural tree diagram of the rumor propagation tree is generated.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (1)
1. A rumor propagation tree structure optimization method based on position consistency is characterized by comprising the following steps:
step 1: inputting a rumor propagation tree diagram G;
the rumor propagation tree map G is a top-down rumor propagation tree map that naturally follows the propagation direction of the rumor information, and defines the rumor detection dataset as a set of statements C = { C = { (C) } 1 ,C 2 ,…,C |c| A, wherein each declaration C i Corresponding to one source tweet, which ideally consists of all related response tweets arranged in time sequence, the propagation tree is therefore denoted T (r) = T<V i ,E i >In which V is i ＝C i Composed of all related posts as nodes, and E i Representing a set of directed links, where u → v represents information that flows from u to v, v sees it and provides a response to u, i.e., an emotion label on a directed edge;
and 2, step: traversing each node in the rumor propagation tree graph G based on breadth first of consistency to obtain an optimized rumor propagation tree graph G 1 ；
Step 2.1: selecting a node in the rumor propagation tree graph G, and judging whether the node has a child node or not; if the node has a child node, executing the step 2.2; if the node has no child node, ending the operation on the node, and executing the step 2.4;
step 2.2: traversing each child node of the node, and acquiring the position of each child node;
step 2.3: judging whether the child nodes have the same position or not; if the child nodes have the same position, combining the child nodes with the same position into a super node; if the child nodes do not have the same position, ending the operation on the node, and executing the step 2.4;
step 2.4: judging whether the traversal of all nodes in the rumor propagation tree graph G is finished or not; if the traversal of all nodes in the rumor propagation tree graph G is completed, outputting the optimized propagation tree graph G 1 (ii) a Otherwise, returning to the step 2.1;
and step 3: depth-first traversal propagation tree graph G based on consistency of ground 1 Obtaining the optimized propagation tree graph G by each node in the tree 2 ；
Step 3.1: selecting optimized rumor propagation tree graph G 1 Judging whether the node has a child node or not; if the node has a child node, executing the step 3.2; if the node has no child node, ending the operation on the node, and executing the step 3.5;
step 3.2: selecting a child node of the node, and judging whether the child node has a next-level child node; if the child node has a next-level child node, executing the step 3.3; if the child node does not have a next level child node, executing step 3.4;
step 3.3: checking whether the position of the next-level child node of the child node is supported; if the position of the child node at the next level of the child node is the support, combining the child node and the child node at the next level into a super node; otherwise, executing step 3.4;
step 3.4: judging whether the traversal of all child nodes of the node is finished or not; if the traversal of all the child nodes of the node is finished, executing the step 3.5; otherwise, returning to the step 3.2;
step 3.5: judging whether the optimized rumor spreading tree diagram G is finished or not 1 Traversing all nodes in the tree; if the optimized rumor propagation tree diagram G is completed 1 The final optimized propagation tree graph G is output after the traversal of all the nodes in the tree graph 2 (ii) a Otherwise, returning to the step 3.1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010438369.6A CN111581534B (en) | 2020-05-22 | 2020-05-22 | Rumor propagation tree structure optimization method based on consistency of vertical place |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010438369.6A CN111581534B (en) | 2020-05-22 | 2020-05-22 | Rumor propagation tree structure optimization method based on consistency of vertical place |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111581534A CN111581534A (en) | 2020-08-25 |
| CN111581534B true CN111581534B (en) | 2022-12-13 |
Family
ID=72113892
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010438369.6A Active CN111581534B (en) | 2020-05-22 | 2020-05-22 | Rumor propagation tree structure optimization method based on consistency of vertical place |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111581534B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113434684B (en) * | 2021-07-01 | 2022-03-08 | 北京中科研究院 | Rumor detection method, system, equipment and storage medium for self-supervision learning |
| CN115048514B (en) * | 2022-06-09 | 2023-06-16 | 安徽大学 | Personality-based social network rumor detection and standing classification method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467535A (en) * | 2010-11-12 | 2012-05-23 | 金蝶软件（中国）有限公司 | Method, device and system for acquiring tree data output coordinate |
| CN105608086A (en) * | 2014-11-17 | 2016-05-25 | 中兴通讯股份有限公司 | Transaction processing method and device of distributed database system |
| CN106354845A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Microblog rumor recognizing method and system based on propagation structures |
| CN109597944A (en) * | 2018-10-30 | 2019-04-09 | 四川大学 | A kind of single classification microblogging rumour detection model based on deepness belief network |
| CN110046224A (en) * | 2019-04-15 | 2019-07-23 | 哈尔滨工程大学 | A kind of social networks rumour suppressing method based on region |
| CN110069718A (en) * | 2019-04-15 | 2019-07-30 | 哈尔滨工程大学 | A kind of social networks rumour suppressing method based on theme |
| CN110362818A (en) * | 2019-06-06 | 2019-10-22 | 中国科学院信息工程研究所 | Microblogging rumour detection method and system based on customer relationship structure feature |
| CN110443355A (en) * | 2019-08-06 | 2019-11-12 | 苏州思必驰信息科技有限公司 | Dialogue method and system applied to compound conversation tasks |
| CN111008337A (en) * | 2019-11-06 | 2020-04-14 | 武汉大学 | Deep attention rumor identification method and device based on ternary characteristics |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7054867B2 (en) * | 2001-09-18 | 2006-05-30 | Skyris Networks, Inc. | Systems, methods and programming for routing and indexing globally addressable objects and associated business models |
| US8908565B2 (en) * | 2012-07-16 | 2014-12-09 | Cisco Technology, Inc. | Methods and apparatus for efficient decentralized information dissemination in a network |
-
2020
- 2020-05-22 CN CN202010438369.6A patent/CN111581534B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467535A (en) * | 2010-11-12 | 2012-05-23 | 金蝶软件（中国）有限公司 | Method, device and system for acquiring tree data output coordinate |
| CN105608086A (en) * | 2014-11-17 | 2016-05-25 | 中兴通讯股份有限公司 | Transaction processing method and device of distributed database system |
| CN106354845A (en) * | 2016-08-31 | 2017-01-25 | 上海交通大学 | Microblog rumor recognizing method and system based on propagation structures |
| CN109597944A (en) * | 2018-10-30 | 2019-04-09 | 四川大学 | A kind of single classification microblogging rumour detection model based on deepness belief network |
| CN110046224A (en) * | 2019-04-15 | 2019-07-23 | 哈尔滨工程大学 | A kind of social networks rumour suppressing method based on region |
| CN110069718A (en) * | 2019-04-15 | 2019-07-30 | 哈尔滨工程大学 | A kind of social networks rumour suppressing method based on theme |
| CN110362818A (en) * | 2019-06-06 | 2019-10-22 | 中国科学院信息工程研究所 | Microblogging rumour detection method and system based on customer relationship structure feature |
| CN110443355A (en) * | 2019-08-06 | 2019-11-12 | 苏州思必驰信息科技有限公司 | Dialogue method and system applied to compound conversation tasks |
| CN111008337A (en) * | 2019-11-06 | 2020-04-14 | 武汉大学 | Deep attention rumor identification method and device based on ternary characteristics |
Non-Patent Citations (4)
| Title |
|---|
| False rumors detection on Sina Weibo by propagation structures;Ke Wu 等;《2015 IEEE 31st International Conference on Data Engineering》;20150601;651-662 * |
| 基于k-核过滤的社交网络影响最大化算法;李阅志 等;《计算机应用》;20181231;第38卷(第2期);464-470 * |
| 基于标记信息级联传播树特征的谣言检测新方法;蔡国永 等;《计算机工程与科学》;20181231;第40卷(第8期);1488-1495 * |
| 社会媒体大数据分析研究综述;杜治娟 等;《计算机科学与探索》;20171231;第11卷(第1期);1-23 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111581534A (en) | 2020-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111581534B (en) | Rumor propagation tree structure optimization method based on consistency of vertical place | |
| CN105956184B (en) | Collaborative and organized junk information issue the recognition methods of group in a kind of microblogging community network | |
| Huang et al. | DeepDiff: Learning deep difference features on human body parts for person re-identification | |
| CN112329444B (en) | Early rumor detection method fusing text and propagation structure | |
| CN109218304A (en) | A kind of network risks blocking-up method based on attack graph and coevolution | |
| CN115809327B (en) | Real-time social network rumor detection method based on multimode fusion and topics | |
| CN109788067A (en) | A kind of information processing method and information processing system based on block chain | |
| CN109561100A (en) | Method and system based on the distributed duplexing energized network attacking and defending with artificial intelligence | |
| Cheng et al. | MultiJAF: Multi-modal joint entity alignment framework for multi-modal knowledge graph | |
| CN116595467A (en) | Abnormal user detection method based on dynamic weighted graph convolution and storage medium | |
| CN113537272B (en) | Deep learning-based semi-supervised social network abnormal account detection method | |
| Manna et al. | Exploring topic models on short texts: a case study with crisis data | |
| CN115238773A (en) | Malicious account detection method and device for heterogeneous primitive path automatic evaluation | |
| CN114219089A (en) | Construction method and equipment of new-generation information technology industry knowledge graph | |
| Sun et al. | Reinforced Contrastive Graph Neural Networks (RCGNN) for Anomaly Detection | |
| Dutta et al. | A genetic algorithm based tweet clustering technique | |
| Zhang et al. | SARC: Split-and-recombine networks for knowledge-based recommendation | |
| CN117131503B (en) | Threat chain identification method for user behavior | |
| Kamakshidasan et al. | Understanding merge trees with force-directed landscapes | |
| CN117556830B (en) | Rumor detection method based on potential hot topics and propagation process | |
| CN114493905B (en) | Social relationship identification method based on multilevel feature fusion | |
| CN113254494B (en) | New energy research and development classification recording method | |
| Zhao et al. | Airborne LiDAR point cloud classification based on transfer learning | |
| CN104765847A (en) | Frequent closed item set mining method based on order-preserving characteristic and preamble tree | |
| Barros et al. | ORCHNet: A robust global feature aggregation approach for 3D LiDAR-based place recognition in orchards |
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 |