A social network user identification method and device based on multi-dimensional social fingerprint

Abstract

The present invention discloses a social network user identification method and device based on multi-dimensional social fingerprints, belonging to the technical field of user identification. The method comprises collecting social network data and performing preprocessing, extracting multi-dimensional social fingerprint features of users and constructing multi-dimensional social fingerprints; calculating the social fingerprint distance between users; inputting the multi-dimensional social fingerprint distance as a data set into a model for training and testing to determine whether users belong to the same person, thereby identifying the identity of an unknown user. The present invention can improve the accuracy and effectiveness of user identification.

Description

Technical Field

[0001] The present invention relates to the technical field of user identification, and more particularly to a method and device for identifying social network users based on multi-dimensional social fingerprints. Background Art

[0002] With the rapid development of social networks, a large amount of user interaction data has been accumulated on social networking platforms. This data includes not only users' personal information, but also interactions between users, published content, comments, likes, etc. This rich data not only provides social networking platforms with deep user insights, but also brings new opportunities for the dissemination and sharing of information. However, this also brings a series of problems, such as user identity forgery, the proliferation of fake accounts, and the increase of malicious behavior. These problems not only affect the healthy development of social networks, but also pose a threat to users' privacy and security.

[0003] In order to deal with these problems, user identification has become a key technology. Through effective user identification methods, real users and fake accounts can be distinguished, and social robot users can also be identified, thereby improving the security and credibility of social networks. Existing social network user identification methods cover a variety of approaches, including graph-based algorithms, machine learning techniques, and crowdsourcing strategies. Although these methods have their own characteristics, they also have certain limitations. For example, graph theory algorithms are good at discovering the social relationships of users in social networks, but attackers may evade identification by forging users' social relationship graphs. In addition, the relationship graphs relied on by this method are static and cannot capture the dynamic evolution of user relationships. Machine learning technology can extract the behavior patterns of social network users from massive user behavior data, and then train the identification model to improve the recognition efficiency and accuracy. However, this method is affected by data quality, the complexity of user behavior, and the dynamic changes of social networks. Crowdsourcing strategies have advantages in data accuracy and reliability, but their implementation in large-scale networks may lead to high costs and privacy issues.

[0004] Among these diverse recognition algorithms, extracting effective features is the key to identifying social network users. Effective features need to be able to accurately capture the behavioral patterns of social network users. However, since the behaviors and strategies of social network users may continue to evolve, this increases the difficulty of feature engineering.

[0005] At present, graph-based methods have shown certain effectiveness in identifying social network users, but most methods still need to collect diverse feature data of users, which not only increases the complexity of data collection, but also most of the collected features are often difficult to accurately capture the dynamic evolution of user relationships in social networks. Therefore, when extracting features that can accurately identify social network users, it is also necessary to explore their applicability in different social networks to solve the current challenges.

[0006] For example, patent 202010814842.6 discloses a user identification method, device, equipment and storage medium in a social network, which relies on graph structure data for analysis and identification. The method uses propagation data in a social network to characterize the connection relationship between users, generates graph structure data based on the above propagation data to represent the connection between users, and then uses the graph structure data to determine the attribute data of each user, such as node degree, close centrality and betweenness centrality, and finally identifies key users and sets corresponding labels based on the user's attribute data. However, this method mainly relies on static graph structure data and cannot fully capture the dynamic changes of user relationships in social networks. The behavior and relationships of users are constantly changing in social networks, and static graph structures may not reflect these changes, thereby affecting the accuracy of identification. For another example, patent 202210429087.9 discloses a method for identifying associated users based on a social network topology graph, which collects data sets of two social networks, including friend relationships between users. According to the user and friend relationships, the topological graphs of social networks G-1 and G-2 are constructed respectively. Using the random walk method, a set of node sequences is generated, and the skip-gram model is used to map the node sequence set into a feature space, and the low-dimensional vector representation of the node is learned. The target feature mapping matrix is ​​obtained through training, and the two feature spaces are mapped to the same space, the similarity between nodes is calculated, and the associated users are identified. However, this method mainly relies on the static topological graph of the social network, and calculates the node features and user associations through the graph structure, which is also unable to fully capture the dynamic changes of user behavior and relationships. In addition, the process of mapping the node sequence into a feature space using the skip-gram model and training the target feature mapping matrix usually requires a lot of computing resources and training time. This computing resource requirement may limit the scalability of this method, especially when it is used in real-time in large-scale social networks.

[0007] Therefore, how to provide a social network user identification method and device based on multi-dimensional social fingerprints that can solve the above problems is a problem that technical personnel in this field urgently need to solve. Summary of the invention

[0008] In view of this, the present invention provides a social network user identification method and device based on multi-dimensional social fingerprints, which constructs more unique and tamper-proof features to cope with the dynamically changing social network environment, thereby improving the accuracy and effectiveness of user identification.

[0009] In order to achieve the above object, the present invention provides the following technical solutions:

[0010] A method for identifying social network users based on multi-dimensional social fingerprints comprises the following steps:

[0011] S100: Collect social network data;

[0012] S200: Constructing social interaction networks based on social network data;

[0013] S300: Extracting multi-dimensional social fingerprint features of users based on social interaction networks;

[0014] S400: constructing a multi-dimensional social fingerprint based on the multi-dimensional social fingerprint feature;

[0015] S500: Calculate the multi-dimensional social fingerprint distance between users, use the social fingerprint distance between each unknown user and the known user as a multi-dimensional social fingerprint distance feature, and classify and label the user pairs to which the multi-dimensional social fingerprint distance feature belongs, and generate label sample data of the same user and different users;

[0016] S600: Construct an XGBoost recognition model, and input the multi-dimensional social fingerprint distance feature into the XGBoost recognition model for training and verification;

[0017] S700: Through the trained XGBoost recognition model, the social fingerprint distance features between unknown users and known users are used to identify users.

[0018] Preferably, S200 includes:

[0019] Extracting from the social network data a one-hop individual interaction graph of a user directly connected to its one-hop neighbor, and an individual interaction graph of a user and its one-hop neighbor and two-hop neighbors indirectly connected via the one-hop neighbor;

[0020] Constructing social interaction networks based on individual interaction graphs of two-hop neighbors.

[0021] Preferably, S300 includes:

[0022] Based on the social interaction network, multi-dimensional social fingerprint features between users and their corresponding contacts are extracted.

[0023] Preferably, the multi-dimensional social fingerprint features include: number of interactions, interaction reciprocity, topological overlap, connection diversity, total number of motifs, contact age, relative freshness, proportion of continuous calls within a preset time, and user activity diversity features.

[0024] Preferably, S400 includes:

[0025] The extracted multi-dimensional social fingerprint features are arranged in order from large to small, and the proportion of the value of each contact under each multi-dimensional social fingerprint feature to the total feature value under the feature is calculated and returned to construct the multi-dimensional social fingerprint of the user within the preset time interval.

[0026] Preferably, S500 includes:

[0027] Dividing the social network into different periods evenly according to time, wherein users in the first period are classified as known users, and users in other periods are classified as unknown users;

[0028] The social fingerprint distance between each unknown user in other periods and all known users in the first period is calculated in sequence and used as the multi-dimensional social fingerprint distance feature;

[0029] Among them, the true label of the unknown user is known. If the true label of the known user and the unknown user belongs to the same user, the multi-dimensional social fingerprint distance feature between the two users is labeled as 1. If they are not the same user, the multi-dimensional social fingerprint distance feature is labeled as 0 to generate label sample data of the same user and different users.

[0030] Preferably, S600 includes:

[0031] Divide the labeled sample data into training set, validation set and test set;

[0032] Input the multi-dimensional social fingerprint distance features between users in the training set and the pre-set labels in the training set into the XGBoost recognition model for training;

[0033] The optimal parameters of the XGBoost recognition model are determined through the validation set, and the final XGBoost recognition model is generated.

[0034] An electronic device comprises: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute a social network user identification method based on multi-dimensional social fingerprints.

[0035] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method and device for identifying social network users based on multi-dimensional social fingerprints, which captures the dynamic multi-dimensional social fingerprint features of users by processing user data on social networks. Multi-dimensional social fingerprints are constructed in the network and the social fingerprint distances between users are calculated, and these distance feature data sets are used to classify them to identify the user identity. This method only needs to extract the interactive information in the user network to accurately identify social network users, and the constructed multi-dimensional social fingerprints can effectively reflect the long-term and unique interactive behaviors of users in social networks. This overcomes the problem of the difficulty in adapting to dynamic environments faced by existing social network user identification technologies based on static information, and significantly improves the temporal stability of the algorithm. Therefore, the application of this method in actual systems is more convenient and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0037] Figure 1 A schematic diagram of a flow chart of a social network user identification method based on multi-dimensional social fingerprints provided by the present invention;

[0038] Figure 2 An overall flow chart for constructing the multi-dimensional social fingerprint provided by the present invention;

[0039] Figure 3 A flowchart of multi-dimensional social fingerprint feature calculation provided by the present invention;

[0040] Figure 4 A flow chart of constructing a social fingerprint based on feature values ​​provided by the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The purpose of the present invention is to provide a method and device for identifying social network users based on multi-dimensional social fingerprints. The method includes: collecting social network data and performing preprocessing, extracting multi-dimensional social fingerprint features of users and constructing multi-dimensional social fingerprints, calculating the social fingerprint distance between users, and inputting the multi-dimensional social fingerprint distance features as a data set into a model for training and testing, determining whether users belong to the same person, thereby identifying the identity of unknown users, and providing a solution to the problem of poor accuracy and effectiveness of existing user identification.

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] See also Figure 1 On the one hand, an embodiment of the present invention discloses a method for identifying social network users based on multi-dimensional social fingerprints, comprising the following steps:

[0045] S100: Collect social network data: Manually collect interaction data of some accounts on social platforms.

[0046] S200: Building a social interaction network based on social network data; specifically including: user nodes With another user node interaction, you can create a arrive Directed edge In this network, each directed edge is assigned a weight whose value is determined by the node With Node The number of interactions between them.

[0047] S300: Extracting multi-dimensional social fingerprint features of users based on social interaction networks; specifically including: based on the social interaction network of users and their contacts, the number of interactions between users and their contacts, interaction reciprocity, topological overlap, connection diversity, total number of motifs, contact age, relative freshness, proportion of continuous calls within 5 minutes, and user activity diversity features can be extracted.

[0048] S400: constructing a multi-dimensional social fingerprint based on the multi-dimensional social fingerprint features.

[0049] S500: Calculate the distance of the multi-dimensional social fingerprint constructed by each feature between users, take the social fingerprint distance between each unknown user and the known user as the multi-dimensional social fingerprint distance feature, and classify and label the user pairs to which the multi-dimensional social fingerprint distance feature belongs, and generate label sample data of the same user and different users.

[0050] S600: Build an XGBoost recognition model and input the multi-dimensional social fingerprint distance features into the XGBoost recognition model for training and verification;

[0051] S700: Through the trained XGBoost recognition model, the social fingerprint distance features between unknown users and known users are used to identify users.

[0052] Specifically, in a specific embodiment, collecting social network data in step S100 includes:

[0053] An email dataset from a large research organization, covering all incoming and outgoing email communication records between members and their contacts. The dataset is simplified to members, contacts, and timestamps, covering a period of 17 months from October 2003 to March 2005.

[0054] In a specific embodiment, building a social interaction network in step S200 includes:

[0055] The constructed social interaction network is a directed weighted network in the form of ,in, V is the set of nodes in the network, E is the set of edges in the network, usually represented as an ordered pair (e.g., Represents a slave node u To Node v ), W is the weight set in the network, representing the weight of each edge, usually expressed as a function (a set of real numbers), that is, for each edge There is a corresponding weight .

[0056] Specifically, the social interaction network rules constructed in this step are: based on the social network interaction data collected in step S100, each user is abstracted as a node in the graph u , and convert the interaction between user nodes into edges in the network. Specifically, if the user node With another user node interaction, you can create a arrive Directed edge In this way, all users collected in step 1 are converted into nodes of the network, and the corresponding directed edges are formed according to the interactive relationship between them. When constructing a directed weighted network, the strength of the interactive relationship also needs to be considered. Specifically, if the user node With another user node If there are multiple interactions between them, a higher weight can be assigned to this directed edge. Indicates that from the user node With another user node Finally, a complete social interaction network is constructed, which accurately reflects the interactive relationship between users.

[0057] In a specific embodiment, extracting the user's multi-dimensional social fingerprint construction features in step S300 includes:

[0058] The interaction count feature is defined as:

[0059] ;

[0060] in, a Indicates user u For users v The number of interactions initiated, b Indicates user v For users u The number of interactions initiated.

[0061] The interactive reciprocity feature is defined as:

[0062] ;

[0063] The topological overlap feature is defined as:

[0064] ;

[0065] in, Indicates user u The neighbor set of Indicates user v The neighbor set of Indicates user u With users v The number of common neighbors of Indicates user u With users v The total number of neighbors.

[0066] The connection diversity feature is defined as:

[0067] ;

[0068] in, Indicates user u The number of neighbors of Indicates user v The number of neighbors.

[0069] Specifically, the total number of motifs feature is defined as . Indicates user u No. i The number of fourth-order motifs, .

[0070] The contact age feature is defined as:

[0071] ;

[0072] in, Indicates the last moment of the current period, Indicates user u With users v The moment of first interaction.

[0073] The relative freshness feature is defined as:

[0074] ;

[0075] in, Indicates user u With users v The time of the last interaction before the end of the current period. Represents the average inter-event time between interactions.

[0076] The continuous call ratio feature within 5 minutes is defined as:

[0077] ;

[0078] in, Indicates user u With users v In the current period n The time of interaction.

[0079] The user activity diversity feature is defined as:

[0080] ;

[0081] in, Indicates user u The number of connections created, Indicates user v The number of connections created.

[0082] According to the above formula, the multi-dimensional social fingerprint construction features are calculated, and the calculation process is as follows: Figure 2 As shown. Figure 2In the above, (a) is the structure of social network data; (b) extracts the one-hop individual interaction graph (1-IIG) directly connecting the user (ego) with its one-hop neighbor, and the individual interaction graph (2-IIG) between the user (ego) and its one-hop neighbor and the two-hop neighbor indirectly connected through the one-hop neighbor from the social network data; (c) accurately calculates the multi-dimensional social fingerprint characteristics of the intensity, time and structure between the user and his contacts based on the 1-IIG and 2-IIG, and the multi-dimensional social fingerprint characteristics include the number of interactions, interaction reciprocity, topological overlap, connection diversity, total number of motifs, contact age, relative freshness, continuous call ratio within 5 minutes and user activity diversity characteristics; (d) arranges the feature values ​​under each feature in order from large to small and calculates their proportions, thereby constructing the user's multi-dimensional social fingerprint. Among them, Figure 2 The specific calculation process of subgraph (c) in Figure 3 , the specific construction process of subgraph (d) can be found in Figure 4 .

[0083] In a specific embodiment, constructing a multi-dimensional social fingerprint in step S400 includes:

[0084] User collection in social interaction network U , which contains n users, i.e. At the same time, suppose there is a set of time intervals T , which contains m time interval, that is For each user and each time interval , a feature vector can be extracted from the data , indicating that the user In the time interval Behavioral characteristics of users. The contacts are classified according to the feature vector Arrange the values ​​in from large to small, and return the proportion of each contact's value to the total value, thereby constructing the user In the time interval Social fingerprint .

[0085] In a specific embodiment, calculating the social fingerprint distance between users in step S500 includes:

[0086] Jensen-Shannon divergence (JSD) is used as a tool to measure the distance of social fingerprints. The distance between users' social fingerprints at different times is calculated. And the reference distance of all users to other users at different times .

[0087] Among them, the social fingerprint distance is defined as:

[0088] ;

[0089] in, , are the social fingerprints of two users, is the Shannon entropy, k The number of digits of the last contact after sorting, which is equal to the total number of contacts. For ranking r The characteristic values ​​of the contacts, X is the sum of the feature values ​​of all contacts.

[0090] Specifically, first calculate the user The distance between social fingerprints in different time intervals . Second, calculate the user The reference distance between the social fingerprints of other users in a certain time interval ,in This embodiment is The label is 1, which represents the same user. The label is 0, which represents different users.

[0091] In a specific embodiment, in step S600, the multi-dimensional social fingerprint distance feature data is input into the XGBoost model for training and testing, including:

[0092] The multi-dimensional social fingerprint distance features are extracted from the sample data to form a feature data set, and the data set is divided into a training set, a validation set, and a test set, where the ratio of the training set to the validation set is 9:1. The feature data in the training set and the preset labels in the training set are input into the XGBoost model, and the optimal parameters of the model are trained through the validation set. The test set is used to judge the accuracy of the generated XGBoost recognition model. Finally, the user identity can be accurately identified based on the prediction results of the model.

[0093] Specifically, this embodiment uses the calculated social fingerprint distance of each detected user pair as a feature value. For an unknown user, the social fingerprint distance between the unknown user and each known user in the known user set is calculated, so as to accurately obtain the identity of the unknown user.

[0094] More specifically, the present invention extracts multi-dimensional social fingerprint features from the constructed social interaction network, calculates the social fingerprint distance between unknown users and known users, and inputs it as a feature into the XGBoost model for user identification. Multi-dimensional social fingerprints are formed based on a certain period of time and cover the strength, structure, and time characteristics of the network, providing clues for understanding the overall structural evolution of the network. Through network analysis using multi-dimensional social fingerprints, a finer granularity can be obtained, so that the structure and dynamics of the network can be grasped more accurately, which solves the problem that the existing social network user identification technology based on static information is difficult to cope with dynamic network environments. In addition, the selection and extraction process of multi-dimensional social fingerprint features is usually closely related to specific tasks and data sets, so it can be optimized and adjusted according to different application scenarios. This flexibility enables the algorithm to adapt well to different data and tasks, thereby improving its scalability.

[0095] On the other hand, an embodiment of the present invention further discloses an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors execute a social network user identification method based on multi-dimensional social fingerprints.

[0096] Specifically, through the above technical solution, the multi-dimensional social fingerprint features extracted from the social interaction network constructed based on Email data in the present invention can accurately capture the interactive information in the network. This information fully reflects the interactions and relationships between nodes, so that the unique interaction patterns of social network users can be more accurately identified, thereby improving the accuracy of social network user identification. And because of the flexibility and versatility of the multi-dimensional social fingerprint features, when this method is migrated to other social platform data, it can also excellently identify users. And because the accuracy of social network user identification is improved, it can be used to identify social robots in the network, so that malicious robots can be more effectively prevented from attacking social network platforms, spreading false information or committing fraud, thereby improving the security of social networks.

[0097] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0098] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A social network user identification method based on multi-dimensional social fingerprint, characterized in that: The following steps are involved: S100: Collect social network data; S200: Constructing social interaction networks based on social network data; S300: Extracting multi-dimensional social fingerprint features of users based on social interaction networks; S400: constructing a multi-dimensional social fingerprint based on the multi-dimensional social fingerprint feature, including: Arrange the extracted multi-dimensional social fingerprint features in descending order, calculate and return the ratio of the value of each contact under each multi-dimensional social fingerprint feature to the total feature value under the feature, and construct the multi-dimensional social fingerprint of the user within a preset time interval; The step S400 of constructing a multi-dimensional social fingerprint includes: The user set U in the social interaction network contains n users, that is, U = {u1, u2, ..., u n }; At the same time, suppose there is a time interval set T, which contains m time intervals, that is, T = {t1, t2, ..., t m }; For each user u i and each time interval t j , extract a feature vector x(u i , t j ), indicating user u i In the time interval t j Behavioral characteristics of user u i The contacts are classified according to the characteristic vector x(u i , t j ) are arranged from large to small, and the proportion of each contact's value to the total value is returned, thereby constructing the user u i In the time interval t j The social fingerprint P(u i , t j ); S500: Calculate the multi-dimensional social fingerprint distance between users, use the social fingerprint distance between each user to be identified and the known user as a multi-dimensional social fingerprint distance feature, and classify and label the user pairs to which the multi-dimensional social fingerprint distance feature belongs, and generate label sample data of the same user and different users, including: Dividing the social network into different periods evenly according to time, wherein users in the first period are classified as known users, and users in other periods are classified as unknown users; The social fingerprint distance between each unknown user in other periods and all known users in the first period is calculated in sequence and used as a multi-dimensional social fingerprint distance feature; The real label of the unknown user is known. If the real label of the known user and the unknown user is the same user, the multi-dimensional social fingerprint distance feature between the two users is labeled as 1. If they are not the same user, the multi-dimensional social fingerprint distance feature is labeled as 0 to generate label sample data of the same user and different users. The step S500 of calculating the social fingerprint distance between users includes: Jensen-Shannon divergence (JSD) is used as a tool to measure the distance of social fingerprints; the distance d between users’ social fingerprints at different times is calculated. self And the reference distance d of all users to other users at different times ref ; Among them, the social fingerprint distance is defined as: Where P1 and P2 are the social fingerprints of two users, H(P) is the Shannon entropy, k is the number of digits of the last contact after sorting, which is numerically equal to the total number of contacts; x(r) is the eigenvalue of the contact ranked r, and X is the sum of the eigenvalues ​​of all contacts; The specific process is as follows: First, calculate user u i The distance d between social fingerprints in different time intervals self (u i ); Secondly, calculate user u i The reference distance d between the social fingerprints of other users in a certain time interval ref (u i ,u j ), where u j ∈U-{u i }; S600: Construct an XGBoost recognition model, and input the multi-dimensional social fingerprint distance feature into the XGBoost recognition model for training and verification; S700: Through the trained XGBoost recognition model, the social fingerprint distance feature between the user to be identified and the known users is used to perform user identification.

2. The method for identifying social network users based on multi-dimensional social fingerprints according to claim 1, characterized in that: S200 includes: Extracting from the social network data a one-hop individual interaction graph of a user directly connected to its one-hop neighbor, and an individual interaction graph of a user and its one-hop neighbor and two-hop neighbors indirectly connected via the one-hop neighbor; Constructing social interaction networks based on individual interaction graphs of two-hop neighbors.

3. The method for identifying social network users based on multi-dimensional social fingerprints according to claim 1, characterized in that: S300 includes: Based on the social interaction network, multi-dimensional social fingerprint features between users and their corresponding contacts are extracted.

4. The method for identifying social network users based on multi-dimensional social fingerprints according to claim 1, characterized in that: The multi-dimensional social fingerprint features include: number of interactions, interaction reciprocity, topological overlap, connection diversity, total number of motifs, contact age, relative freshness, proportion of continuous calls within 5 minutes, and user activity diversity features.

5. The method for identifying social network users based on multi-dimensional social fingerprints according to claim 1, characterized in that: S600 includes: Divide the labeled sample data into training set, validation set and test set; Input the multi-dimensional social fingerprint distance features between users in the training set and the pre-set labels in the training set into the XGBoost recognition model for training; The optimal parameters of the XGBoost recognition model are determined through the validation set, and the final XGBoost recognition model is generated.

6. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the method according to any one of claims 1 to 5.

Images