A rumor detection method based on multi-feature fusion and sentiment difference analysis

Abstract

This invention discloses a rumor detection method based on multi-feature fusion and sentiment difference analysis, specifically relating to the field of rumor detection technology. Its key technical points are: inputting online text data and multiple online comment data into a BERT model for feature extraction to obtain the text semantic features and text sentiment features corresponding to the online text data, as well as the comment semantic features and comment sentiment features corresponding to each online comment data; fusing the text semantic features with the multiple comment semantic features to obtain fused semantic features, and calculating the differences between the text sentiment features and each comment sentiment feature to obtain a difference vector feature; fusing the fused semantic features and the difference vector feature to generate fused features, and using a fully connected network to identify the fused features to obtain the rumor identification result corresponding to the news event.

Description

Technical Field

[0001] This invention relates to the field of rumor detection technology, specifically to a rumor detection method based on multi-feature fusion and sentiment difference analysis. Background Technology

[0002] Rumors are generally considered to be unverified information without supporting evidence, authoritative information, or official statements. Unlike real events, rumors often use extreme or strong language in their emotional expression, or use exaggerated headlines and images that are not strongly related to the content to attract attention, attempting to manipulate readers' emotions and stances, eliciting strong emotional reactions, and thus making them easier to accept and spread.

[0003] In recent years, rumor detection technology has gained increasing attention; however, existing methods still have some significant shortcomings. Many traditional rumor detection studies rely heavily on semantic feature extraction, often neglecting the impact of sentiment features on rumor dissemination. Secondly, existing methods largely depend on manual annotation for sentiment labeling and typically employ a single sentiment classification system. This limits the in-depth understanding of complex emotions in comments, lacks effective mining of sentiment differences, and may lead to biased rumor judgments, affecting the overall performance of the model. Some studies have attempted to improve feature extraction using deep learning techniques, but shortcomings remain in feature fusion and sentiment analysis, failing to achieve an effective combination of semantic and sentiment features, thus impacting the accuracy and efficiency of rumor detection.

[0004] Therefore, the present invention aims to provide a rumor detection method based on multi-feature fusion and sentiment difference analysis to solve the aforementioned problems. Summary of the Invention

[0005] The technical problem this invention aims to solve is the failure of existing technologies to effectively combine semantic and sentiment features, thus affecting the accuracy and efficiency of rumor detection. The goal is to provide a rumor detection method based on multi-feature fusion and sentiment difference analysis. This method involves inputting the acquired news event's online text data and corresponding online comment data into a BERT model to obtain sentiment and semantic features for both the online text and comment data. The semantic features of the text are then fused with the semantic features of the comment data using fine-grained features to obtain fused semantic features. The differences between the sentiment features of the text and each comment are calculated to obtain a difference vector feature. Finally, the fused semantic features and the difference vector feature are fused to obtain the final fused feature. A fully connected network is used as a discriminator to perform rumor identification on this fused feature, thus achieving a rumor identification result. This method combines semantic and sentiment features to improve rumor identification performance and solves the problem of existing technologies failing to effectively combine semantic and sentiment features, thereby affecting the accuracy and efficiency of rumor detection.

[0006] This invention is achieved through the following technical solution:

[0007] A rumor detection method based on multi-feature fusion and sentiment difference analysis, the method includes:

[0008] Obtain the online text data of a news event and the corresponding online comment data;

[0009] The network text data and multiple network comment data are respectively input into the BERT model for feature extraction, to obtain the text semantic features and text sentiment features corresponding to the network text data, and to obtain the comment semantic features and comment sentiment features corresponding to each network comment data.

[0010] The semantic features of the main text are fused with the semantic features of multiple comments to obtain fused semantic features. The difference between the sentiment features of the main text and the sentiment features of each comment is calculated to obtain the difference vector feature.

[0011] The fused semantic features and differential vector features are fused to generate fused features. A pre-trained fully connected network is used to identify the fused features in order to obtain the rumor identification results corresponding to the news event.

[0012] Furthermore, the method also includes:

[0013] The error between the rumor identification result and the true result is calculated using the cross-entropy loss function. The parameters of the fully connected network are iteratively optimized based on the error value until the loss function converges to obtain the optimized fully connected network.

[0014] Furthermore, the online text data and multiple online comment data are respectively input into the BERT model for feature extraction, obtaining the semantic features and sentiment features of the online text data, and the semantic features and sentiment features of each online comment data, specifically:

[0015] The network text data and multiple network comment data are respectively input into the BERT model to obtain the text feature representation corresponding to the network text data and the comment feature representation corresponding to each network comment data.

[0016] Semantic features are extracted from the main text feature representation and multiple comment feature representations using a multilayer perceptron, resulting in main text semantic features and multiple comment semantic features, respectively.

[0017] Then, the sentiment features in the main text feature representation and multiple comment feature representation are extracted by multilayer perceptron to obtain the main text sentiment features and multiple comment sentiment features respectively.

[0018] Furthermore, the semantic features of the main text are fused with the semantic features of multiple comments to obtain fused semantic features. The differences between the sentiment features of the main text and the sentiment features of each comment are then calculated to obtain a difference vector feature, specifically:

[0019] A semantic feature matrix for comments is constructed using multiple semantic features of comments. After transposing the semantic feature matrix for comments, matrix multiplication is performed with the semantic features of the main text to obtain semantic scores. Then, the semantic scores are multiplied with the semantic feature matrix for comments to obtain fused semantic features.

[0020] By calculating the Euclidean distances between the sentiment features of the main text and the sentiment features of multiple comments, a dissimilarity vector feature is constructed using the multiple Euclidean distances.

[0021] Furthermore, the semantic features and differential vector features are fused to generate fused features. A fully connected network is then used to evaluate these fused features to obtain the rumor identification results corresponding to the network text data. Specifically:

[0022] The fused semantic features and the differential vector features are fused and concatenated to obtain the fused features;

[0023] The fused features and corresponding true / false labels are input into a fully connected network, and the softmax function in the fully connected network is used to identify rumors, thus obtaining the rumor identification results corresponding to the news event.

[0024] This invention also provides a rumor detection system based on multi-feature fusion and sentiment difference analysis, which is used in any of the rumor detection methods based on multi-feature fusion and sentiment difference analysis described above, the system comprising:

[0025] The data acquisition module is used to acquire the online text data of news events and the corresponding online comment data.

[0026] The feature extraction module is used to input the online text data and multiple online comment data into the BERT model for feature extraction, to obtain the semantic features and sentiment features of the online text data, and to obtain the semantic features and sentiment features of each online comment data.

[0027] The feature processing module is used to fuse the semantic features of the main text with the semantic features of multiple comments to obtain fused semantic features, and to calculate the difference between the sentiment features of the main text and the sentiment features of each comment to obtain the difference vector feature.

[0028] The rumor identification module is used to fuse semantic features and differential vector features to generate fused features. The fused features are then identified using a pre-trained fully connected network to obtain the rumor identification results corresponding to the news event.

[0029] Furthermore, the system also includes:

[0030] The network optimization module is used to calculate the error between the rumor identification result and the true result using the cross-entropy loss function. Based on the error value, the parameters of the fully connected network are iteratively optimized until the loss function converges to obtain the optimized fully connected network.

[0031] The present invention also provides a computer device, including a system memory and a processor, wherein the system memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described above.

[0032] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of any one of the methods described above.

[0033] The present invention also provides a computer program product containing instructions, characterized in that, when the instructions are executed by a cluster of computer devices, the cluster of computer devices causes the cluster of computer devices to perform the method described in any of the preceding claims.

[0034] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0035] In this invention, the acquired news event's online text data and corresponding online comment data are input into a BERT model to obtain sentiment features and semantic features of the online text data and multiple online comment data, respectively. Then, the text semantic features and multiple comment semantic features are fused at a fine-grained level to obtain fused semantic features. The difference between the text sentiment features and each comment sentiment feature is calculated to obtain a difference vector feature. Finally, the fused semantic features and the difference vector feature are fused to obtain the final fused feature. A fully connected network is used as a discriminator to perform rumor identification on this fused feature to obtain the rumor identification result. This achieves the combination of semantic features and sentiment features to improve the rumor identification effect, solving the problem in existing technologies where the effective combination of semantic features and sentiment features is not achieved, thus affecting the accuracy and efficiency of rumor detection. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0037] Figure 1 This is a flowchart of a rumor detection method based on multi-feature fusion and sentiment difference analysis in this embodiment;

[0038] Figure 2 This is a schematic diagram of the module connections of a rumor detection system based on multi-feature fusion and sentiment difference analysis in this embodiment;

[0039] Figure 3 This is a schematic diagram of the structure of a computer device. Detailed Implementation

[0040] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0041] In this disclosure, unless otherwise stated, the use of terms such as "first," "second," etc., to describe various elements is not intended to define the positional, temporal, or importance relationships of these elements; such terms are merely used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of that element, while in other cases, based on the context, they may refer to different instances.

[0042] The terminology used in the description of the various examples in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context explicitly indicates otherwise, an element may be one or more unless the number of elements is specifically limited. Furthermore, the term "and / or" as used in this disclosure covers any one of the listed items and all possible combinations thereof.

[0043] Example 1

[0044] See Figure 1 As shown in the figure, this embodiment provides a rumor detection method based on multi-feature fusion and sentiment difference analysis. The method includes:

[0045] S1: Obtain the online text data of the news event and the corresponding online comment data. Specifically:

[0046] After obtaining the online text data of the news event and the corresponding online comment data, the online text data and the multiple online comment data are first processed to remove redundant information not covered by the news event, and the number of online comment data is uniformly set to N; then the sentiment tags of the online text data and the multiple online comment data are obtained, and the format of the online text data and the multiple online comment data is converted.

[0047] S2: Input the online text data and multiple online comment data into the BERT model for feature extraction, respectively, to obtain the semantic features and sentiment features of the online text data, and the semantic features and sentiment features of each online comment data, specifically:

[0048] The main text data and multiple online comment data are input into the BERT model to obtain the text feature representation T corresponding to the main text data. t And the comment feature representation T corresponding to each network comment data. cN Specifically:

[0049]

[0050] Among them, X t For web text data; X c ={Xc1 ,X c2 ,…,X cN} represents online comment data; T cN For each piece of online comment data, a comment feature representation is provided, where N is the number of comments and is a positive integer; T c A comment feature representation matrix constructed for multiple comment feature representations;

[0051] The text feature representation T is obtained through a multilayer perceptron. t And multiple comment features represent T cN Semantic features are extracted from the text to obtain the semantic features F of the main text. t and multiple comment semantic features F cN Specifically:

[0052]

[0053] Among them, W t W is the weight matrix of the semantic features of the main text; c b is the weight matrix of the semantic features of the comment group; t b c Both are bias vectors; d fc F represents the feature dimension output by the fully connected layer. c A comment semantic feature matrix constructed from multiple comment semantic features;

[0054] Then, the text feature representation T is processed through a multilayer perceptron. t And multiple comment features represent T cN The sentiment features in the text are extracted to obtain the sentiment features E of the main text. t and multiple comment sentiment features E cN Specifically:

[0055]

[0056] Among them, W et W is the weight matrix for the sentiment features of the main text. ec b is the weight matrix for the sentiment features of the comments. et b ec Both are bias vectors; d fc E represents the feature dimension output by the fully connected layer. c A comment sentiment feature matrix constructed for multiple comment sentiment features.

[0057] S3: The semantic features of the main text are fused with the semantic features of multiple comments to obtain fused semantic features. The differences between the sentiment features of the main text and the sentiment features of each comment are then calculated to obtain a difference vector feature. Specifically:

[0058] Utilizing multiple comment semantic features FcN Construct the semantic feature matrix F of the comments c For the semantic feature matrix F of the comments c After transposition, they are compared with the semantic features F of the main text. t Perform matrix multiplication to obtain the semantic score R, and then multiply the semantic score R with the comment semantic feature matrix F. c Perform matrix multiplication to obtain the fused semantic features F fusion Specifically:

[0059]

[0060] F fusion =R·F c

[0061] in, This is the transposed semantic feature matrix of the comments;

[0062] By calculating the sentiment feature E of the main text t Each comment sentiment feature E cN The Euclidean distance D between them N Using multiple Euclidean distances D N The difference vector feature D is constructed as follows:

[0063]

[0064] S4: The semantic features and differential vector features are fused to generate fused features. A pre-trained fully connected network is then used to identify these fused features to obtain the rumor identification results for the news event. Specifically:

[0065] For the fusion semantic features F fusion The feature F is obtained by fusing and concatenating the feature D with the difference vector. final =[F fusion ;D];

[0066] F fusion feature F final The corresponding true / false labels y are input into a fully connected network, and rumor identification is performed through the softmax function in the fully connected network. The fused features F are then used to identify the rumor. final Mapping to the true and false classification target spaces, we obtain the classification results. This means obtaining the rumor identification results corresponding to the news event, specifically:

[0067]

[0068] Where W represents the weight matrix in the classifier; b represents the bias term.

[0069] In another embodiment, the method further includes: calculating the error value between the rumor identification result and the true result using the cross-entropy loss function, and iteratively optimizing the parameters of the fully connected network based on the error value until the loss function converges to obtain the optimized fully connected network, specifically:

[0070] Using cross-entropy loss function Calculate the error between the rumor identification result and the true result, and optimize the parameters of the rumor detection task. The true result of the event is y. i ∈{0,1} (0 represents real news, 1 represents rumor), the model identification result is The calculation results are as follows:

[0071]

[0072] Among them, y i This represents the true result of the i-th event; This represents the identification result of the i-th event;

[0073] Meanwhile, the true sentiment tag for each comment is y. c ∈{0,1,2} (representing negative, neutral, and positive sentiment respectively), the sentiment probability predicted by the model is The cross-entropy loss function is used to calculate the error in sentiment classification of comments:

[0074]

[0075] Where k represents the comment index; j represents the sentiment category index; This represents the true sentiment label of the k-th comment; This represents the true sentiment label of the k-th comment; C is the number of sentiment categories.

[0076] The final overall loss function is defined as the weighted sum of the classification loss for rumor detection and the sentiment classification loss.

[0077]

[0078] Where λ is the weighting coefficient.

[0079] Specifically, in this embodiment, the acquired news event's online text data and the corresponding multiple online comment data are input into the BERT model to obtain sentiment features and semantic features of the online text data and multiple online comment data, respectively. Then, the text semantic features and multiple comment semantic features are fused at a fine-grained level to obtain fused semantic features. The difference between the text sentiment features and each comment sentiment feature is calculated to obtain a difference vector feature. Finally, the fused semantic features and the difference vector feature are fused to obtain the final fused feature. A fully connected network is used as a discriminator to perform rumor identification on this fused feature to obtain the rumor identification result. This achieves the combination of semantic features and sentiment features to improve the rumor identification effect, solving the problem in the prior art that failed to effectively combine semantic features and sentiment features, thus affecting the accuracy and efficiency of rumor detection.

[0080] Example 2

[0081] See Figure 2 As shown, this embodiment also provides a rumor detection system based on multi-feature fusion and sentiment difference analysis. This system is used in any of the rumor detection methods described above based on multi-feature fusion and sentiment difference analysis. The system includes:

[0082] The data acquisition module 100 is used to acquire the online text data of a news event and the corresponding online comment data.

[0083] The feature extraction module 200 is used to input the online text data and multiple online comment data into the BERT model to extract semantic features, so as to obtain the text semantic features and text sentiment features corresponding to the online text data, and to obtain the comment semantic features and comment sentiment features corresponding to each online comment data.

[0084] The feature processing module 300 is used to fuse the semantic features of the main text with the semantic features of multiple comments to obtain fused semantic features, and to calculate the difference between the sentiment features of the main text and the sentiment features of each comment to obtain the difference vector feature.

[0085] The rumor identification module 400 is used to fuse semantic features and differential vector features to generate fused features, and then use a pre-trained fully connected network to identify the fused features in order to obtain the rumor identification results corresponding to the news event.

[0086] Furthermore, the system also includes: using the cross-entropy loss function to calculate the error value between the rumor identification result and the true result, and iteratively optimizing the parameters of the fully connected network based on the error value until the loss function converges to obtain the optimized fully connected network.

[0087] It should be noted that the modules in the system of Embodiment 2 correspond to the steps in the method of Embodiment 1. The steps in the method of Embodiment 1 have been described in detail in Embodiment 1, and the module content in the system will not be described in detail in this Embodiment 2.

[0088] Example 3

[0089] See Figure 3 As shown, this embodiment also provides a computer device, including a system memory 1005 and a processor 1001. The system memory 1005 stores a computer program, and the processor 1001 executes the computer program to implement the steps of any of the methods described above.

[0090] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, when the processor 1001 executes the computer program, it implements the functions of each module / unit in the above system / device embodiments.

[0091] Specifically, in this embodiment, the computer program can be divided into one or more modules / units. One or more modules / units are stored in the system memory 1005 and executed by the processor 1001 to complete this application. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0092] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will understand that this does not constitute a limitation on the terminal device, and it may include more or fewer components than shown in the figures, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0093] The processor 1001 can be a Central Processing Unit (CPU), or other general-purpose processors 1001, digital signal processors 1001 (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 1001 can be a microprocessor or any conventional processor, etc.

[0094] System memory 1005 can be an internal storage unit of the terminal device, such as a hard drive or RAM. System memory 1005 can also be a storage device 1004 of the terminal device, such as an external hard drive, SmartMedia Card (SMC), Secure Digital (SD) card, or FlashCard. Furthermore, system memory 1005 can include both internal storage units and storage device 1004. System memory 1005 is used to store computer programs and other programs and data required by the terminal device. System memory 1005 can also be used to temporarily store data that has been output or will be output.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0096] Example 4

[0097] This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0098] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof, or any other form of computer-readable storage medium in the art.

[0099] An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside within an application-specific integrated circuit (ASIC). In embodiments of the invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device.

[0100] Example 5

[0101] This embodiment also provides a computer program product containing instructions that, when executed by a cluster of computer devices, cause the cluster of computer devices to perform the method described in Embodiment 1.

[0102] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rumor detection method based on multi-feature fusion and sentiment difference analysis, characterized in that, The methods include: Obtain the online text data of a news event and the corresponding online comment data; The main text data and multiple online comment data are respectively input into the BERT model for feature extraction, resulting in the semantic and sentiment features of the main text data, and the semantic and sentiment features of each online comment data. Specifically: The network text data and multiple network comment data are respectively input into the BERT model to obtain the text feature representation corresponding to the network text data and the comment feature representation corresponding to each network comment data. Semantic features are extracted from the main text feature representation and multiple comment feature representations using a multilayer perceptron, resulting in main text semantic features and multiple comment semantic features, respectively. Then, the sentiment features in the main text feature representation and multiple comment feature representations are extracted by a multilayer perceptron to obtain the main text sentiment features and multiple comment sentiment features respectively; The semantic features of the main text are fused with the semantic features of multiple comments to obtain fused semantic features. The differences between the sentiment features of the main text and the sentiment features of each comment are then calculated to obtain a difference vector feature. Specifically: A semantic feature matrix for comments is constructed using multiple semantic features of comments. After transposing the semantic feature matrix for comments, matrix multiplication is performed with the semantic features of the main text to obtain semantic scores. Then, the semantic scores are multiplied with the semantic feature matrix for comments to obtain fused semantic features. By calculating the Euclidean distance between the sentiment features of the main text and the sentiment features of multiple comments, a differential vector feature is constructed using the multiple Euclidean distances; The fused semantic features and differential vector features are fused to generate fused features. A pre-trained fully connected network is used to identify the fused features in order to obtain the rumor identification results corresponding to the news event.

2. The rumor detection method based on multi-feature fusion and sentiment difference analysis according to claim 1, characterized in that, The method also includes: The error between the rumor identification result and the true result is calculated using the cross-entropy loss function. The parameters of the fully connected network are iteratively optimized based on the error value until the loss function converges to obtain the trained fully connected network.

3. The rumor detection method based on multi-feature fusion and sentiment difference analysis according to claim 1, characterized in that, The semantic features and differential vector features are fused to generate fused features. A fully connected network is then used to evaluate these fused features to obtain the rumor identification results corresponding to the network text data. Specifically: The fused semantic features and the differential vector features are fused and concatenated to obtain the fused features; The fused features and corresponding true / false labels are input into a fully connected network, and the softmax function in the fully connected network is used to identify rumors, thus obtaining the rumor identification results corresponding to the news event.

4. A rumor detection system based on multi-feature fusion and sentiment difference analysis, characterized in that, This system is used in the rumor detection method based on multi-feature fusion and sentiment difference analysis as described in any one of claims 1-3, wherein the system comprises: The data acquisition module is used to acquire the online text data of news events and the corresponding online comment data. The feature extraction module is used to input the online text data and multiple online comment data into the BERT model for feature extraction, to obtain the semantic features and sentiment features of the online text data, and to obtain the semantic features and sentiment features of each online comment data. The feature processing module is used to fuse the semantic features of the main text with the semantic features of multiple comments to obtain fused semantic features, and to calculate the difference between the sentiment features of the main text and the sentiment features of each comment to obtain the difference vector feature. The rumor identification module is used to fuse semantic features and differential vector features to generate fused features. The fused features are then identified using a pre-trained fully connected network to obtain the rumor identification results corresponding to the news event.

5. A rumor detection system based on multi-feature fusion and sentiment difference analysis according to claim 4, characterized in that, The system also includes: The network optimization module is used to calculate the error between the rumor identification result and the true result using the cross-entropy loss function. Based on the error value, the parameters of the fully connected network are iteratively optimized until the loss function converges to obtain the optimized fully connected network.

6. A computer device comprising a system memory and a processor, wherein the system memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 3.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 3.

8. A computer program product containing instructions, characterized in that, When the instructions are executed by a cluster of computer devices, the cluster of computer devices causes the cluster of computer devices to perform the method as described in any one of claims 1 to 3.

Images