CN115797871A - Analysis method and system for infant companion social network - Google Patents
Analysis method and system for infant companion social network Download PDFInfo
- Publication number
- CN115797871A CN115797871A CN202211654400.5A CN202211654400A CN115797871A CN 115797871 A CN115797871 A CN 115797871A CN 202211654400 A CN202211654400 A CN 202211654400A CN 115797871 A CN115797871 A CN 115797871A
- Authority
- CN
- China
- Prior art keywords
- infant
- social
- social network
- child
- children
- 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
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of social networks, and provides a method and a system for analyzing a social network of children companions, wherein the method comprises the following steps: extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten; carrying out target identification on the multiple monitoring images, identifying individual children and calculating social distances among the children; constructing a child-partner social network in which each individual child identified is taken as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edges between is taken as e ij Weight w on edge ij η is the social distance threshold; analyzing and calculating the social network of the child companions according to the characteristic vector centrality analysis methodAnd the importance index of the nodes in the network is used for representing the social interaction capacity of the children. By the technical scheme, the problem of poor accuracy of social network analysis of the child companion relationship in the related technology is solved.
Description
Technical Field
The invention relates to the technical field of social networks, in particular to an analysis method and system for a social network of child companions.
Background
Social networks are an important branch of sociology. The research starting point of the social network is that actors and various links exist among the actors, the social network analysis is described by adopting a graph theory in a mathematical method, and a deep network mode hidden under the surface of a complex social system is researched. A social network is composed of a plurality of points and connecting lines among the points, the points are all social actors, the edges are all social relations among the actors, social network analysis is to establish a model of the relations, try to describe the structure of group relations, and study the influence of the structure on group functions or individuals in the group.
With the development of society and the progress of education, the attention of researchers to the social adaptability, interpersonal communication ability and cooperative communication ability of students in all ages is gradually improved. Meanwhile, as the number of patients with depression and autism is gradually increased and even the patients are attacked in the infancy, the social interaction capacity of students, particularly the social interaction of the infants, becomes the key point of the education and teaching research.
The social interaction of the infant is the need of growth and personality development, is the process of mutual interaction information and emotion between the infant and surrounding people, and is the process of completing individual socialization. The infant stage is a key stage of social development of a person, the social interaction ability of the infant is strong and weak, whether the infant can establish a good harmonious relationship with other people and a group in the interaction process is related, and more importantly, the infant enters the society after becoming mature, the self ability is fully exerted, and therefore a good foundation is laid for promoting the harmonious development of the society. The development of social interaction ability of the infants is concerned, the social development level of the infants is improved, the development of self-consciousness, social cognition and emotional feeling of the infants is facilitated, the requirement of the era development is met, the urgent requirement of talent quality is improved, and the method has great practical significance in quality education of the infants.
In the existing research, the infant companion social network is constructed by performing data acquisition through a companion relationship nomination method, an interview method and an observation method. The research aims to guide daily infant education work, a teacher carries out targeted intervention work according to a quantized network analysis result, and necessarily needs to feed back, explain and cooperate with parents, but the accuracy of the result is easily questioned by the captain due to too many subjective factors in the research process, and the interpretability is poor.
Disclosure of Invention
The invention provides an analysis method and system for a social network of child companions, and solves the problem of poor accuracy of analysis of the social network of child companions in the related technology.
The technical scheme of the invention is as follows:
in a first aspect, a method for analyzing a peer-to-peer social network of a child comprises the following steps:
extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
performing target identification on the plurality of monitoring images, identifying individual children and calculating social distances among the children;
constructing a child-partner social network in which each child individual identified is regarded as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edges between is taken as e ij Weight w on edge ij η is the social distance threshold;
and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance index of the nodes in the network, wherein the importance index of the nodes in the network is used for representing the social interaction capacity of the infant.
In a second aspect, include
An image sampling module: extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
an image processing module: performing target identification on the plurality of monitoring images, identifying individual children and calculating social distances among the children;
network architectureBuilding a module: constructing a child-partner social network in which each individual child identified is taken as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edges between is taken as e ij Weight w on edge ij (ii) a η is the social distance threshold;
the network analysis calculation module: and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance indexes of the nodes in the network, wherein the importance indexes of the nodes in the network are used for representing the social interaction capacity of the infant.
The working principle and the beneficial effects of the invention are as follows:
according to the method, each infant individual and the social distance between the infants are identified through a monitoring video, the infant individual is used as a point, for any two infants, when the social distance between the two infants is smaller than a social distance threshold value, an edge is added between the two infants, the proportion of the edge between the two infants in all monitoring images is used as the weight of the edge, an infant peer social network is constructed according to the point, the edge and the weight, the infant peer social network is analyzed through a feature vector centrality analysis method, the importance index of nodes in the network is calculated, the social interaction capacity of the infant individual is judged through the importance index of the nodes in the network, and the method is used for later infant peer intervention guidance teaching.
The social interaction ability of the individual is reflected through the live-action data of the activities of the infants, and compared with a companion naming method, an interview method and an observation method, the problems do not need to be answered by the infants, parents and teachers, and the problem of unreliable data caused by subjective factors is avoided.
Drawings
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
FIG. 1 is a flow chart of a method for analyzing a social network of a child partner according to the present invention;
FIG. 2 is a flowchart illustrating an exemplary method for analyzing a social network of a child partner according to the present invention;
fig. 3 is a schematic structural diagram of an analysis system of a peer-to-peer social network of a baby according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any inventive step, are intended to be within the scope of the present invention.
Example 1
As shown in fig. 1-2, the present embodiment provides a method for analyzing a social network of child associates, including the following steps:
s100: extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
in this embodiment, select certain angle camera data as main surveillance video in the indoor outer multi-angle camera of follow kindergarten, other angle surveillance videos are as supplementing. According to teaching requirements, a main monitoring video of indoor or outdoor independent activities of the infant without intervention of a teacher is collected once, the main monitoring video is edited, and the duration is a free activity class of the infant. And (3) carrying out frame image extraction on the video data of indoor and outdoor free activities of the infant in the free activity class of the infant to obtain m frame images as m monitoring images.
S200: performing target identification on the plurality of monitoring images, identifying individual children and calculating social distances among the children;
in this embodiment, a deep learning algorithm in the YoloV5 model is applied to perform infant target detection and face recognition on each monitored image, identify an infant individual v, and calculate a social distance l between infants.
S300: constructing a child-partner social network in which each child individual identified is regarded as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij Multiple monitoring imagesMiddle v i And v j The ratio of the edges between is taken as e ij Weight w on edge ij η is the social distance threshold;
forming a point v in the social network by the child as an actor; in the infant v i And v j Social distance between l ij When l is ij When eta is less than or equal to eta, increase the edge e ij Else v i And v j There is no edge between them, all monitoring image points v i And point v j The ratio of the edges between is taken as e ij Weight on edge w ij And constructing a child companion undirected social network with right according to the points, edges and weights.
S400: and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance indexes of the nodes in the network, wherein the importance indexes of the nodes in the network are used for representing the social interaction capacity of the infant.
And judging the social interaction ability of the individual children through the importance indexes of the nodes in the network, and using the social interaction ability for later-stage infant companion intervention guidance teaching.
Further, still include:
the language communication capability is used as a part in the feature vector centrality calculation, each node in the child partner social network is endowed with a centrality initial value beta, and the node v in the child partner social network is corrected i The importance index is obtained after correction;
determining the social interaction ability of the infant according to the corrected importance index, and using the social interaction ability for later-stage infant companion intervention guidance teaching;
wherein any child node v is determined i Initial value of centrality of (beta) i The calculating method comprises the following steps:
carrying out target identification on the multiple monitoring images, calculating the height h and width w of the mouth of the infant, setting a threshold value mu for closing the mouth, identifying as opening the mouth when w/h is less than or equal to mu, and otherwise, identifying as closing the mouth;
in the embodiment, when frame image extraction is performed on the main monitoring video, the frame rate is set by comprehensively considering the speaking speed of the infant, so that the extracted frame images are continuous frames with open mouth and closed mouth, and a plurality of continuous frames are used as a plurality of monitoring images; and carrying out target recognition on the monitored images, and calculating the mouth height h and the mouth width w of the infant in each monitored image.
Determining the language communication time length d of the infant according to the sum of the time lengths of opening and closing the mouth of the infant;
according to the language communication time d of the infant, calculating a centrality initial value beta i The specific calculation formula is as follows:
wherein d is i Node infant v for infant peer social network i Duration of language exchange of d max For all children the most language exchanges the language exchange duration corresponding to the child, d min And the language communication time length corresponding to the child with the minimum language communication among all the children is obtained.
Therefore, the embodiment adopts an EVC-CPC (evolution-computational-feature-vector centrality algorithm) for infant peer social network integrating language communication, the language communication capability is used as a part of the centrality calculation of feature vectors, each node in the infant peer social network is endowed with a centrality initial value beta, the information is integrated into an infant peer social network node importance index calculation formula, and the node v in the infant peer social network is corrected i Importance index of, node v in the social network of child companions i The corrected importance index is calculated according to the formula:
wherein x is i For corrected infant v i Of importance, x j For the young children v j Of importance, A ij =w ij ，w ij For children in the social network v i And v j Weight of the middle edge, beta i For children node v i Initial value of centrality ofAnd alpha is a parameter.
Expressed as:
Wherein, X = (X) 1 ，x 2 ，x 3 ，…x i ，…，x n )，x i For the young children v i N is the number of infants, I is the identity matrix, A is the adjacency matrix, A is the significance index of ij Is an element in the adjacency matrix, and beta is a centrality initial value beta i Vector of compositions, β = (β) 1 ，β 2 ，β 3 ，…β i ，…，β n ) And alpha is a parameter.
Calculating the feature vector centrality value requires a free parameter α, which is responsible for adjusting the balance between the importance index and β. In general, α cannot be an arbitrarily large number, and if α → 0, the formula is equal to β, so the social interaction ability importance index of all children is equal to the social interaction ability of children represented by the language communication. As α increases from zero, the significance index also increases and eventually reaches a point where the significance index diverges. Therefore, an appropriate α value should be selected to ensure convergence of the importance index and balance the centrality of the infant social network feature vector and the infant language communication ability β. By current practice, α typically takes a value less than the maximum eigenvalue of the adjacency matrix a.
The language communication fused infant peer social network feature vector centrality algorithm (EVC-CPC) not only comprises infant peer social network feature vector centrality analysis, but also integrates node importance influence brought by infant language communication capacity, and evaluation results are comprehensive and efficient. The analysis result can also be used for digging key nodes (children with better social ability) and isolated points (children with an autism tendency) in the social network of the child partner, and theoretical basis and support are provided for the intervention of the child partner in teaching activities.
Further, the target recognition is carried out on the multiple monitoring images, the mouth height h and the mouth width w of the infant are calculated, and the method specifically comprises the following steps: target detection is applied to each monitored image, face recognition is performed by applying a dat model library of 68 key point detections of the face, and the mouth height h between 52 and 58 points and the mouth width w between 49 and 55 points are calculated.
Because the activity scene of the infant is complex, the video sound is noisy, and the video sound and the infant node v cannot be directly passed through i Therefore, in the embodiment, the target detection in the Yolov5 model is combined with the dat model library for detecting 68 key points of the human face to perform the human face recognition, and the language communication duration d of the infant is determined.
Example 2
As shown in fig. 3, based on the same concept as that of embodiment 1, the present embodiment further provides an analysis system for social networks of children's peers, which includes
An image sampling module: extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
an image processing module: carrying out target identification on the multiple monitoring images, identifying individual children and calculating social distances among the children;
a network construction module: constructing a child-partner social network in which each individual child identified is taken as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edges between is taken as e ij Weight w on edge ij (ii) a η is the social distance threshold;
the network analysis calculation module: and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance indexes of the nodes in the network, wherein the importance indexes of the nodes in the network are used for representing the social interaction capacity of the infant.
Further, the network analysis computation module further comprises:
a correction calculation module for using the language communication capability as a feature directionIn part of the amount centrality calculation, each node in the infant companion social network is endowed with a centrality initial value beta, and a node v in the infant companion social network is corrected i Obtaining the corrected importance index;
the correction analysis module is used for determining the social interaction ability of the infant according to the corrected importance index;
wherein any child node v is determined i Initial value of centrality of (beta) i The method comprises the following specific steps:
carrying out target identification on the multiple monitoring images, calculating the height h and width w of the mouth of the infant, setting a threshold value mu for closing the mouth, identifying as opening the mouth when w/h is less than or equal to mu, and otherwise, identifying as closing the mouth; determining the language communication time length d of the infant according to the sum of the time lengths of opening and closing the mouth of the infant; according to the language communication time d of the infant, calculating a centrality initial value beta i 。
Further, the image sampling module is specifically configured to:
editing the monitoring video according to the teaching requirement, and collecting the monitoring video of the independent activities of the children indoors or outdoors without intervention of a teacher for t minutes;
and (4) carrying out frame image extraction on the monitoring video of the independent activity of the infant within t minutes to generate continuous frame images, wherein the continuous frame images are used as a plurality of monitoring images.
Further, the image processing module is specifically configured to:
and carrying out target detection and face recognition on the infants in each monitoring image through a deep learning algorithm, identifying individual infants, and calculating the social distance between the infants.
Various changes and specific examples of the method for analyzing a social network of a child partner in the foregoing embodiment 1 are also applicable to the system for analyzing a social network of a child partner in the foregoing embodiment, and a person skilled in the art can clearly know the method for implementing the system for analyzing a social network of a child partner in the present embodiment through the detailed description of the method for analyzing a social network of a child partner, so for the brevity of the description, detailed descriptions are omitted here.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions, improvements, etc. within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An analysis method of a child partner social network is characterized by comprising the following steps:
extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
carrying out target identification on the multiple monitoring images, identifying individual children and calculating social distances among the children;
constructing a child-partner social network in which each individual child identified is taken as a node v in the child-partner social network i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edges between is taken as e ij Weight w on edge ij η is the social distance threshold;
and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance index of the nodes in the network, wherein the importance index of the nodes in the network is used for representing the social interaction capacity of the infant.
2. The method for analyzing a social network of child associates according to claim 1, further comprising:
the language communication capacity is used as a part in the feature vector centrality calculation, a centrality initial value beta is given to each node in the infant peer social network, the importance index of the node in the infant peer social network is corrected, and the corrected importance index is obtained;
determining the social interaction capacity of the infant according to the corrected importance index;
wherein any child node v i Initial value of centrality of (beta) i The calculating method comprises the following steps:
carrying out target identification on the multiple monitoring images, calculating the height h and width w of the mouth of the infant, setting a threshold value mu for closing the mouth, identifying as opening the mouth when w/h is less than or equal to mu, and otherwise, identifying as closing the mouth;
determining the language communication time length d of the infant according to the sum of the time lengths of opening and closing the mouth of the infant;
according to the language exchange time d of the child, calculating a centrality initial value beta i The specific calculation formula is as follows:
wherein d is i Node infant v for infant peer social network i Duration of language exchange of d max For all children the most language exchanges the language exchange duration corresponding to the child, d min And communicating the language communication time length corresponding to the least children for all the children.
3. The method for analyzing a social network of young children companions according to claim 1, wherein extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten specifically comprises:
editing the monitoring video according to the teaching requirement, and collecting the monitoring video of the independent activities of the children indoors or outdoors without intervention of a teacher for t minutes;
and (3) carrying out frame image extraction on the monitoring video of the independent activity of the infant within t minutes to generate continuous frame images, wherein the continuous frame images are used as a plurality of monitoring images.
4. The method for analyzing a social network of peer children according to claim 1, wherein the identifying of the target of the plurality of monitored images and the identification of individual children and social distances among children specifically comprises:
and carrying out target detection and face recognition on the infants in each monitoring image through a deep learning algorithm, identifying individual infants, and calculating the social distance between the infants.
5. The method for analyzing the social network of infant companions according to claim 2, wherein the method for correcting the importance index of the nodes in the social network of infant companions by assigning a centrality initial value β to each node in the social network of infant companions specifically comprises:
wherein, X = (X) 1 ，x 2 ，x 3 ，…x i ，…，x n )，x i For the young children v i N is the number of infants,Iis an identity matrix, A is w ij A composed adjacency matrix, w ij For v in the infant companion social network i And v j Weight of the edge between, β = (β) 1 ，β 2 ，β 3 ，…β i ，…，β n ) And alpha is a parameter.
6. The method for analyzing a social network of child associates according to claim 2, wherein the step of performing target recognition on the plurality of monitored images and calculating the height h and width w of the mouth of the child specifically comprises:
object detection is applied to each monitored image, face recognition is performed by applying a dat model library of face 68 key point detection, and the mouth height h between 52 and 58 points and the mouth width w between 49 and 55 points are calculated.
7. An analysis system for social networks of children and companions is characterized by comprising
An image sampling module: extracting a plurality of monitoring images from indoor and outdoor monitoring videos of a kindergarten;
an image processing module: carrying out target identification on the multiple monitoring images, identifying individual children and calculating social distances among the children;
a network construction module: constructing a peer-to-peer social network of children, wherein each child to be identified is identifiedThe child individual serves as a node v in the child companion social network, where the child v i And v j Social distance between l ij When eta is less than or equal to eta, increase the edge e ij V in a plurality of monitored images i And v j The ratio of the edge to the edge is taken as e ij Weight on edge w ij (ii) a η is the social distance threshold;
the network analysis calculation module: and analyzing the infant peer social network according to the characteristic vector centrality analysis method, and calculating the importance index of the nodes in the network, wherein the importance index of the nodes in the network is used for representing the social interaction capacity of the infant.
8. The system for analyzing peer-to-peer social networks in children of claim 7, wherein the network analysis computing module further comprises:
a correction calculation module for using the language communication ability as a part of the feature vector centrality calculation, giving a centrality initial value beta to each node in the infant peer social network, and correcting the node v in the infant peer social network i The importance index is obtained after correction;
the correction analysis module is used for determining the social interaction ability of the infant according to the corrected importance index;
wherein any child node v is determined i Initial value of centrality of (beta) i The method comprises the following specific steps:
carrying out target identification on the multiple monitoring images, calculating the height h and width w of the mouth of the infant, setting a threshold value mu for closing the mouth, identifying as opening the mouth when w/h is less than or equal to mu, and otherwise, identifying as closing the mouth; determining language communication duration d of the infant according to the sum of the durations of opening and closing the mouth of the infant; according to the language communication time d of the infant, calculating a centrality initial value beta i 。
9. The system for analyzing a social network of child associates according to claim 7, wherein the image sampling module is specifically configured to:
editing the monitoring video according to the teaching requirement, and collecting the monitoring video of the independent activities of the children indoors or outdoors without intervention of a teacher for t minutes;
and (3) carrying out frame image extraction on the monitoring video of the independent activity of the infant within t minutes to generate continuous frame images, wherein the continuous frame images are used as a plurality of monitoring images.
10. The system for analyzing a social network of child associates according to claim 7, wherein the image processing module is specifically configured to:
and carrying out target detection and face recognition on the infants in each monitoring image through a deep learning algorithm, identifying individual infants, and calculating the social distance between the infants.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211654400.5A CN115797871A (en) | 2022-12-22 | 2022-12-22 | Analysis method and system for infant companion social network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211654400.5A CN115797871A (en) | 2022-12-22 | 2022-12-22 | Analysis method and system for infant companion social network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115797871A true CN115797871A (en) | 2023-03-14 |
Family
ID=85427635
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211654400.5A Pending CN115797871A (en) | 2022-12-22 | 2022-12-22 | Analysis method and system for infant companion social network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115797871A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130132158A1 (en) * | 2011-05-27 | 2013-05-23 | Groupon, Inc. | Computing early adopters and potential influencers using transactional data and network analysis |
| CN104616438A (en) * | 2015-03-02 | 2015-05-13 | 重庆市科学技术研究院 | Yawning action detection method for detecting fatigue driving |
| US20150170295A1 (en) * | 2013-12-17 | 2015-06-18 | Palo Alto Research Center Incorporated | System and method for identifying key targets in a social network by heuristically approximating influence |
| CN105117422A (en) * | 2015-07-30 | 2015-12-02 | 中国传媒大学 | Intelligent social network recommender system |
| CN111640048A (en) * | 2020-05-20 | 2020-09-08 | 合肥巴灵瑞教育科技有限公司 | Social ability evaluation system based on infant behavior track analysis |
| CN112200698A (en) * | 2020-09-30 | 2021-01-08 | 黄日光 | Campus social relationship big data analysis system based on artificial intelligence |
| US20210334311A1 (en) * | 2018-09-28 | 2021-10-28 | Suzhou Dajiaying Information Technology Co., Ltd. | Method, apparatus, device and storage medium for determining a central vertex in a social network |
| CN114513426A (en) * | 2022-03-02 | 2022-05-17 | 郑州轻工业大学 | CCN community division method based on node similarity and influence |
-
2022
- 2022-12-22 CN CN202211654400.5A patent/CN115797871A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130132158A1 (en) * | 2011-05-27 | 2013-05-23 | Groupon, Inc. | Computing early adopters and potential influencers using transactional data and network analysis |
| US20150170295A1 (en) * | 2013-12-17 | 2015-06-18 | Palo Alto Research Center Incorporated | System and method for identifying key targets in a social network by heuristically approximating influence |
| CN104616438A (en) * | 2015-03-02 | 2015-05-13 | 重庆市科学技术研究院 | Yawning action detection method for detecting fatigue driving |
| CN105117422A (en) * | 2015-07-30 | 2015-12-02 | 中国传媒大学 | Intelligent social network recommender system |
| US20210334311A1 (en) * | 2018-09-28 | 2021-10-28 | Suzhou Dajiaying Information Technology Co., Ltd. | Method, apparatus, device and storage medium for determining a central vertex in a social network |
| CN111640048A (en) * | 2020-05-20 | 2020-09-08 | 合肥巴灵瑞教育科技有限公司 | Social ability evaluation system based on infant behavior track analysis |
| CN112200698A (en) * | 2020-09-30 | 2021-01-08 | 黄日光 | Campus social relationship big data analysis system based on artificial intelligence |
| CN114513426A (en) * | 2022-03-02 | 2022-05-17 | 郑州轻工业大学 | CCN community division method based on node similarity and influence |
Non-Patent Citations (5)
| Title |
|---|
| CAROLYN PARKINSON等: "The Neuroscience of Social Networks", THE OXFORD HANDBOOK OF SOCIAL NETWORKS * |
| MOHAMMED SAQR等: "The Curious Case of Centrality Measures: A Large-Scale Empirical Investigation", JOURNAL OF LEARNING ANALYTICS * |
| 武澎等: "基于特征向量中心性的社交信息超网络中重要节点的评判", 情报理论与实践 * |
| 王顺晔等: "基于社会网络分析的网路舆情管理研究", 电脑知识与技术 * |
| 胥伟岚;: "基于人际网络的学术社交网络知识交流模式研究", 图书馆学研究 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113298191A (en) | User behavior identification method based on personalized semi-supervised online federal learning | |
| CN109408911A (en) | A kind of group's evolution method based on ACP theory at CPSS | |
| CN109743196A (en) | It is a kind of based on the network characterisation method across double-layer network random walk | |
| CN114582030A (en) | Behavior recognition method based on service robot | |
| CN115775085A (en) | Smart city management method and system based on digital twin | |
| CN115346149A (en) | Rope skipping counting method and system based on space-time diagram convolution network | |
| CN110889335B (en) | Human skeleton double interaction behavior identification method based on multichannel space-time fusion network | |
| CN113762716A (en) | Method and system for evaluating running state of transformer area based on deep learning and attention | |
| CN116680358A (en) | Rumor detection method and device based on bidirectional graph attention network | |
| Mansour et al. | Design of integrated artificial intelligence techniques for video surveillance on iot enabled wireless multimedia sensor networks | |
| CN116206327A (en) | Image classification method based on online knowledge distillation | |
| CN117576267B (en) | Digital person generation method based on LLM and ANN and application of digital person generation method in cloud video | |
| Pang et al. | Federated learning for crowd counting in smart surveillance systems | |
| Chen | A novel long short-term memory network model for multimodal music emotion analysis in affective computing | |
| CN115797871A (en) | Analysis method and system for infant companion social network | |
| CN108629362B (en) | A kind of learning behavior habit discovery quantization system and method towards mobile environment | |
| CN115797557A (en) | Self-supervision 3D scene flow estimation method based on graph attention network | |
| CN111178141B (en) | LSTM human body behavior identification method based on attention mechanism | |
| CN114154024A (en) | Link prediction method based on dynamic network attribute representation | |
| CN114004299A (en) | Parent-child relationship verification algorithm based on deep learning | |
| CN114969143A (en) | Inter-city atmospheric pollutant space-time correlation characteristic analysis method based on data mining | |
| CN114359816A (en) | Dynamic capacity-expansion video analysis desk based on edge calculation and intelligent identification method | |
| CN111274866A (en) | Crowd behavior identification method based on subgroup division and dynamic and static feature fusion | |
| KR101687763B1 (en) | System for automatic categorization based on co-movement using life-logging data and method thereof | |
| Paulavičius et al. | Client tuned federated learning for rssi-based indoor localisation |
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 |