User generated content anti-addiction method and system

Abstract

The invention provides a user generated content anti-addiction method and system, and belongs to the technical field of data analysis, and the method comprises the steps: building a first label of current page content at the current time t, and a plurality of second labels corresponding to a plurality of historical page contents in a time period T before the current time t; comparing the first label with the plurality of second labels one by one, and if the similarity or distance value of the first label and the second label is greater than a preset threshold value, judging that the user enters an addiction state; if the user enters the addiction state, an alarm signal is sent out. According to the method, personalized intervention can be performed according to actual conditions such as interests, preferences and behavior modes of different users, so that anti-addiction measures are more accurate and effective, the safety of generating a content platform by the user is improved, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method and system for preventing user-generated content from becoming addicted. Background Art

[0002] Currently, UGC (User-generated-Content) has become a mainstream Internet service. Some Internet companies combine UGC with recommendation algorithms to push a large amount of content to users, which makes it easy for users to become addicted while using UGC. Users who are addicted to UGC content platforms may browse, create or interact for a long time continuously, and then neglect important matters in real life such as learning, work, and socializing. Long-term use may not only have a negative impact on the user's physical health, such as decreased vision and cervical pain, but may also cause psychological problems, such as anxiety, loneliness, and distraction. Therefore, how to effectively prevent users from becoming addicted to UGC content platforms has become an urgent problem to be solved.

[0003] To solve this problem, a variety of anti-addiction methods have been proposed and applied. For example, users’ online time and content access rights are limited by setting usage time limits, mandatory rest mechanisms, real-name authentication, and age verification. However, most of these methods are based on simple control of users’ online time or behavior patterns, and lack in-depth analysis and accurate identification of users’ content preferences. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for preventing addiction to user-generated content, which determines whether the user has entered an addictive state by analyzing and comparing the characteristic information of the user's current page content and the previous page content. The method can capture the user's interest preferences and behavior patterns, thereby providing the user with more personalized and accurate anti-addiction prompts and measures to solve at least one technical problem existing in the above-mentioned background technology.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for preventing user-generated content addiction, comprising: Get the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

[0006] As a further limitation of the first aspect of the present invention, the page content includes one or two different modal data, and the modal data is text data, image data or video data.

[0007] As a further limitation of the first aspect of the present invention, the first third-order tensor is converted into a character set and vectorized to obtain a first feature vector, and a first label is obtained based on the first feature vector, including: converting the first third-order tensor into a first character set, vectorizing the first character set to obtain the first feature vector; extracting feature information of the first feature vector; calculating the conditional probabilities of different independent features, and performing cluster analysis on the first feature information according to the conditional probability values ​​to obtain a first label.

[0008] As a further limitation of the first aspect of the present invention, character set conversion and vectorization are performed on multiple second-third-order tensors to obtain multiple second feature vectors, and multiple second labels are obtained based on the multiple second feature vectors, including: converting each of the second-third-order tensors into a second character set, vectorizing each of the second character sets to obtain a second feature vector; extracting feature information of the second feature vector, calculating conditional probabilities of different independent features, and obtaining the second label based on the conditional probability values ​​and performing cluster analysis.

[0009] As a further limitation of the first aspect of the present invention, the similarity or distance value between the first tag and the second tag is calculated one by one; if the similarity between the first tag and the second tag corresponding to a certain historical moment is higher than the set similarity threshold or the distance value is lower than the set distance threshold, it indicates that there is consistency or similarity between the current page content and the historical page content at that historical moment, and then it is judged that the user has entered an addicted state.

[0010] As a further limitation of the first aspect of the present invention, a naive Bayes classifier is used to calculate the prior probability and conditional probability of each category based on the training data set, and the posterior probability of the first feature information belonging to each category is calculated based on the input first feature information, and the category with the largest probability is selected as the prediction result, i.e., the first label.

[0011] In a second aspect, the present invention provides a user-generated content anti-addiction system, comprising: The acquisition module is used to acquire the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; A processing module is used to establish a first third-order tensor corresponding to the current page content according to the current page content, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; and to establish a plurality of second third-order tensors corresponding to historical pages at a plurality of different historical moments according to the historical page contents at a plurality of different historical moments, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; A conversion module, configured to perform character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtain a first label according to the first feature vector; perform character set conversion and vectorization on a plurality of the second third-order tensors to obtain a plurality of second feature vectors, and obtain a plurality of second labels according to the plurality of second feature vectors; A comparison module, used for comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, it is determined that the user has entered an addicted state; The alarm module is used to send out an alarm signal if the user enters an addicted state.

[0012] In a third aspect, the present invention provides a computer device comprising: a processor and a computer-readable storage medium; the processor is suitable for executing a computer program; the computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the user-generated content anti-addiction method as described in the first aspect is implemented.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program is suitable for being loaded by a processor and executing the user-generated content anti-addiction method as described in the first aspect.

[0014] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method for preventing addiction to user-generated content as described in the first aspect.

[0015] The beneficial effects of the present invention are as follows: first, the user's current page content and the page content within the previously set time are obtained, and then the characteristic information of these contents is extracted and cluster analysis is performed to obtain corresponding labels. Next, the similarity or distance value between the label of the current page content and the label of the previous page content is compared. If the similarity is higher than the set threshold (or the distance value is lower than the set threshold), it is judged that the user may enter an addicted state. Finally, an alarm signal or other preset anti-addiction measures are triggered to remind the user to pay attention to the usage time and content selection. It is possible to perform personalized intervention according to the actual conditions such as the interests, preferences and behavior patterns of different users, making the anti-addiction measures more accurate and effective, improving the security of the UGC content platform, and improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 This is a functional framework diagram of the user-generated content anti-addiction system described in an embodiment of the present invention.

[0018] Figure 2 The present invention is a block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation to the embodiments of the present invention.

[0020] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0021] Example 1

[0022] In this embodiment 1, a method for preventing user-generated content addiction is provided, comprising the following steps: Get the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

[0023] In this embodiment, the user's current page content and the page content within the previously set time are first obtained, and then the feature information of these contents is extracted and cluster analysis is performed to obtain the corresponding labels. Next, the similarity or distance value between the label of the current page content and the label of the previous page content is compared. If the similarity is higher than the set threshold (or the distance value is lower than the set threshold), it is judged that the user may enter an addicted state. Finally, an alarm signal or other preset anti-addiction measures are triggered to remind the user to pay attention to the usage time and content selection. This personalized analysis method makes the anti-addiction measures more accurate and can be customized according to the actual situation of different users. It helps to improve the user experience and security of the UGC content platform and promote the health of users.

[0024] In this embodiment, in the above-mentioned page content information collection step, in actual application, the server is used as the implementation subject, and includes the following steps: Through the API interface or crawler technology of the UGC content platform, the current page content browsed by the user at the current time t is captured in real time, a time window T (i.e., the time period T before time t) is set, and the historical page content browsed by the user at multiple different historical moments within the time window is extracted through the historical records or database of the platform. The page content described above includes but is not limited to multimedia data such as text, images, and videos. Although the page content collected in this step includes but is not limited to multimedia data of multiple modalities such as text, images, and videos, in the stage after this step, the current page content and the historical page content only include information of one or two modalities, such as only one of text, images, and videos, or only any two of text, images, and videos.

[0025] In actual applications, the user end is used as the implementation body. By taking screenshots, the page content browsed by the user at the current time t is captured in real time. A time window T is set. By taking screenshots, the page content browsed by the user at multiple different historical moments within the time window is extracted and retained.

[0026] It is understandable that, when the above-mentioned page content information collection step is implemented, conventional methods can be used to obtain the content of the current page, wherein the current page refers to the page displayed by the current terminal, usually the terminal is a display and / or a speaker and / or a touch display device, wherein the content of the current page refers to the content of various media such as text, images, etc. of the current page. In this embodiment, no matter in what form the current page content and historical page content are obtained, time is the main basis, and each page content is assigned a timestamp, so that the collected page content can be arranged according to the timestamp, thereby ensuring the correctness and consistency of the data source sorting, and providing support for subsequent time series analysis.

[0027] In this embodiment, it can be understood that the current time t is obtained, and a first third-order tensor including the current time t is established. The first third-order tensor is a vector containing three elements. In addition to the current time t information, the elements contained in the first third-order tensor also include the image sample of the current page and the color depth of the current page image. The time period T is set, and multiple second third-order tensors including historical moment information within the time period T before the current time t are established. The second third-order tensors also include historical moment image samples and color depth. Image sample and color depth information are applicable to pages containing image content.

[0028] In this embodiment, multiple historical moments are set within a time period T, which are counted as T1, T2, ..., Tn, respectively, and multiple third-order tensors including historical time information T1, T2, ..., Tn are obtained respectively, that is, the second page content is obtained at T1, T2, ..., Tn. It can also be understood that this embodiment also includes deleting duplicate page content, and only one of the multiple identical screenshots is retained. For the most widely used mobile devices, deleting duplicate content helps control storage size and reduce storage requirements. For UGC content, it can also reduce the deviation of analysis results.

[0029] In this embodiment, for the current page content, natural language processing (NLP) and computer vision (CV) technology are used to perform in-depth analysis to extract first feature information such as keywords, image features, and sentiment tendencies. Specifically, for the text content in the page content, after text recognition, the TF-IDF method, N-gram model or bag of words model content is converted into a first character set, and then the first character set is vectorized to extract the text data feature vector; and / or for the audio content in the page content, the voice conversion method is used to convert the audio content into text content, and then the TF-IDF method, N-gram model or bag of words model is used to extract the text data feature vector; and / or for the image content in the page, the traditional image feature extraction method is used to extract the feature vector of the image, and the traditional image feature extraction method includes SIFT method, SURF method, HOG method, ORB method, etc. This is common knowledge in the art and is not listed one by one here. It can be understood that this step depends on the computing module deployed on the server or the computing module deployed on the client.

[0030] For the historical page content, firstly adopt the method described in the first feature information extraction to extract features, and then use an unsupervised learning algorithm (such as K-means, DBSCAN, etc.) to perform cluster analysis to form multiple different content category labels, namely the second labels.

[0031] It is understandable that in one or more embodiments, the first feature is extracted currently, and the second feature is extracted and retained at a historical moment, and there is a time difference between the two. For the first page content, since it represents the latest data, only simple feature extraction is required to meet basic analysis needs. For the historical page content, since it contains historical data, more in-depth data mining and analysis are required, so cluster analysis is used to discover hidden patterns or trends in the data.

[0032] In this embodiment, the following steps can also be used to extract feature vectors: for the text content in the page content, after the text recognition, a convolutional neural network or a recurrent neural network is used to extract the text data feature vector; and / or for the audio content in the page content, a speech conversion method is used to convert the audio content into text content, and then a convolutional neural network or a recurrent neural network is used to extract the text data feature vector; and / or for the image content in the page, a convolutional neural network or a recurrent neural network is used to extract the image feature vector. It can be understood that this step depends on the computing module deployed on the server.

[0033] In this embodiment, a label generation algorithm is used to map the first feature information to a predefined label system to obtain a label of the current page content, i.e., a first label. In one embodiment, a naive Bayes classifier can be used to generate the first label, calculate the conditional probabilities of different independent features, classify the first feature information according to the conditional probability values, and obtain the first label according to the classification. More specifically, when using a naive Bayes classifier, first calculate the prior probability and conditional probability of each category according to the training data set, then calculate the posterior probability of belonging to each category according to the input first feature information, and select the category with the largest probability as the prediction result, i.e., the first label. It can be understood that the above-mentioned training data set can be a customized picture data set or text data set, which is a common technology in the field and will not be repeated here. In one embodiment, a support vector machine can also be used. First, a support vector machine model is trained on the training data set to find the optimal hyperplane parameters, and then the first feature information is input into the trained support vector machine model, and the category to which it belongs is determined according to the output value, i.e., the first label. In one embodiment, unsupervised or semi-supervised deep learning algorithms such as K-means clustering algorithm, DBSCAN clustering algorithm or label propagation algorithm may also be used, which will not be described in detail here.

[0034] In this embodiment, an empty data structure (i.e., a similarity list) is created in advance. Different data structures can be selected depending on the programming language and the amount of data. Lists or arrays are the most common choices because they can store multiple values ​​in sequence and can be quickly accessed through indexes. The similarity list is set to empty so that similarity values ​​can be added later. After initialization, the program will traverse each second tag, calculate the similarity or distance value between the first tag and each second tag, and add the result to the similarity or distance value list. The similarity or distance value list will be used in subsequent steps to determine whether there is a similarity value that exceeds a preset threshold or a distance value that is lower than a preset threshold, so as to determine whether the user is repeatedly browsing similar content and is at risk of addiction.

[0035] It can be understood that in the above steps, the core is to compare the first label with the second label one by one, and then for each pair of compared labels, calculate the similarity between them, and judge whether the user has entered an addicted state based on the calculated similarity or distance value and the preset threshold; if the similarity of a pair of labels is higher than the threshold or the distance value is lower than the threshold, it may indicate that there is a high consistency or similarity between the current page content and the previous page content, and then judge that the user has entered an addicted state.

[0036] In this embodiment, it is necessary to ensure that each first tag is compared with the corresponding second tag (based on time sequence or other association rules). For each pair of compared tags, the similarity or distance value between them is calculated. The calculation method of the similarity or distance value can be selected according to the actual application scenario, such as cosine similarity, Euclidean distance, Manhattan distance, etc. According to the calculated similarity or distance value, and the preset threshold, it is judged whether the user has entered an addicted state. If the similarity of a pair of tags is higher than the threshold (or the distance value is lower than the threshold, depending on the selected calculation method and the threshold setting method), it may indicate that the user has a high degree of consistency or similarity between the current page content and the previous page content, and then it may be judged that the user has entered an addicted state. After judging that the user has entered an addicted state, an alarm signal or other preset anti-addiction measures can be further triggered to remind the user to pay attention to the usage time and content selection, or to restrict the user's further access.

[0037] Example 2

[0038] like Figure 1 As shown, this embodiment 2 provides a user-generated content anti-addiction system, including: The acquisition module is used to acquire the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; A processing module is used to establish a first third-order tensor corresponding to the current page content according to the current page content, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; and to establish a plurality of second third-order tensors corresponding to historical pages at a plurality of different historical moments according to the historical page contents at a plurality of different historical moments, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; A conversion module, configured to perform character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtain a first label according to the first feature vector; perform character set conversion and vectorization on a plurality of the second third-order tensors to obtain a plurality of second feature vectors, and obtain a plurality of second labels according to the plurality of second feature vectors; A comparison module, used for comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, it is determined that the user has entered an addicted state; The alarm module is used to send out an alarm signal if the user enters an addicted state.

[0039] It is understandable that the above-mentioned modules can be separately or completely combined into one or several other modules to constitute, or one (some) of the modules can be further divided into multiple smaller functional units to constitute, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present invention. The above-mentioned modules are divided based on logical functions. In practical applications, the functions of one module can also be implemented by multiple modules, or the functions of multiple modules can be implemented by one module. In other embodiments of the present invention, the system may also include other modules. In practical applications, these functions can also be implemented with the assistance of other modules, and can be implemented by the collaboration of multiple modules.

[0040] According to another embodiment of the present invention, the system described in this embodiment can be constructed, and the method of Embodiment 1 of the present invention can be implemented by running a computer program (including program code) capable of executing the steps involved in the corresponding method described in Embodiment 1 on a general computing device such as a computer including processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM). The computer program can be recorded on, for example, a computer-readable recording medium, and loaded into the above-mentioned computing device through the computer-readable recording medium and run therein.

[0041] This embodiment is based on a UGC content platform (such as a short video platform) with a large number of users, and some users are overly addicted to UGC content. In order to solve this problem, the platform intends to apply the above system to conduct in-depth analysis of user page content through artificial intelligence technology, determine the user's addiction status, and conduct personalized intervention.

[0042] The platform collects page data of users browsing UGC content, including text, images, etc., and pre-processes the collected data, such as denoising and standardization, to improve the accuracy of subsequent analysis. Artificial intelligence technology is used to extract feature information of page content, such as keywords, image features, etc., and cluster analysis is performed on the extracted feature information to obtain corresponding labels, such as entertainment, education, technology, etc. The similarity or distance value between the current page content label and the previous page content label is calculated, which can be achieved through algorithms such as cosine similarity and Euclidean distance. A threshold is set. When the similarity or distance value exceeds the threshold, it is considered that the user may be repeatedly browsing similar content and there is a risk of addiction. According to the similarity or distance value and the preset threshold, it is judged whether the user has entered an addicted state. When the user is judged to be in an addicted state, an alarm signal is triggered, such as sending a reminder message to the user or popping up a warning window. In addition, time limits and rest reminders can be set. When a user browses UGC content continuously for more than a certain time, the content playback is automatically paused and the user is reminded to rest.

[0043] Example 3

[0044] like Figure 2 As shown, the present implementation provides an electronic device, which includes a processor, a communication interface, and a computer-readable storage medium. The processor, the communication interface, and the computer-readable storage medium may be connected via a bus or other means.

[0045] Among them, the communication interface is used to receive and send data, the computer-readable storage medium can be stored in the memory of the electronic device, the computer-readable storage medium is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer-readable storage medium.

[0046] A processor (or CPU (Central Processing Unit)) is the computing core and control core of an electronic device, which is suitable for implementing one or more instructions, and specifically suitable for loading and executing one or more instructions to implement corresponding method flows or corresponding functions.

[0047] The processor is configured to perform the following process: Get the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

[0048] Example 4

[0049] This implementation provides a computer-readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The computer-readable storage medium provides a storage space that stores the processing system of the electronic device.

[0050] In addition, the storage space also stores one or more instructions suitable for being loaded and executed by the processor, and these instructions may be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here may be a high-speed RAM memory, or a non-volatile memory, such as at least one disk memory; optionally, it may also be at least one computer-readable storage medium located away from the aforementioned processor.

[0051] In one embodiment, the computer-readable storage medium stores one or more instructions; the processor loads and executes the one or more instructions stored in the computer-readable storage medium to implement the following process: Get the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

[0052] Example 5

[0053] The present implementation provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the following process: Get the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

[0054] A person skilled in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0055] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by the computer or a data processing device such as a server, data center, etc. that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0056] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative work on the basis of the technical solution disclosed in the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for preventing user-generated content addiction, characterized in that: include: Get the current page content at the current time, as well as the historical page content at multiple different historical moments within a period of time before the current time; According to the current page content, a first third-order tensor corresponding to the current page content is established, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; according to the historical page contents at multiple different historical moments, multiple second third-order tensors corresponding to historical pages at multiple different historical moments are established, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; Performing character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtaining a first label according to the first feature vector; performing character set conversion and vectorization on the plurality of second third-order tensors to obtain a plurality of second feature vectors, and obtaining a plurality of second labels according to the plurality of second feature vectors; Comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, determining that the user has entered an addicted state; If the user enters an addicted state, an alarm signal will be issued.

2. The user-generated content anti-addiction method according to claim 1, characterized in that: The page content includes one or two different modal data, and the modal data is text data, image data or video data.

3. The user-generated content anti-addiction method according to claim 1, characterized in that: The first third-order tensor is converted into a character set and vectorized to obtain a first feature vector, and a first label is obtained according to the first feature vector, including: converting the first third-order tensor into a first character set, vectorizing the first character set to obtain the first feature vector; extracting feature information of the first feature vector; calculating the conditional probabilities of different independent features, and performing cluster analysis on the first feature information according to the conditional probability values ​​to obtain a first label.

4. The user-generated content anti-addiction method according to claim 1, characterized in that: Performing character set conversion and vectorization on multiple second-third-order tensors to obtain multiple second feature vectors, and obtaining multiple second labels based on the multiple second feature vectors, including: converting each of the second-third-order tensors into a second character set, vectorizing each of the second character sets to obtain a second feature vector; extracting feature information of the second feature vector, calculating conditional probabilities of different independent features, and obtaining the second label based on the conditional probability values ​​and performing cluster analysis.

5. The user-generated content anti-addiction method according to claim 1, characterized in that: The similarity or distance value between the first tag and the second tag is calculated one by one; if the similarity between the first tag and the second tag corresponding to a certain historical moment is higher than the set similarity threshold or the distance value is lower than the set distance threshold, it indicates that there is consistency or similarity between the current page content and the historical page content at the historical moment, and then it is judged that the user has entered an addicted state.

6. The user-generated content anti-addiction method according to claim 3, characterized in that: Using the naive Bayes classifier, the prior probability and conditional probability of each category are calculated according to the training data set. According to the input first feature information, the posterior probability of it belonging to each category is calculated, and the category with the largest probability is selected as the prediction result, that is, the first label.

7. A user-generated content anti-addiction system, characterized in that: include: The acquisition module is used to acquire the current page content at the current time t, and the historical page content at multiple different historical moments within a time period T before the current time t; A processing module is used to establish a first third-order tensor corresponding to the current page content according to the current page content, wherein the first third-order tensor includes current time information, current page image samples, and current page image color depth; and to establish a plurality of second third-order tensors corresponding to historical pages at a plurality of different historical moments according to the historical page contents at a plurality of different historical moments, wherein each of the second third-order tensors includes historical moment information, page image samples at the historical moment, and page image color depth at the historical moment; A conversion module, configured to perform character set conversion and vectorization on the first third-order tensor to obtain a first feature vector, and obtain a first label according to the first feature vector; perform character set conversion and vectorization on a plurality of the second third-order tensors to obtain a plurality of second feature vectors, and obtain a plurality of second labels according to the plurality of second feature vectors; A comparison module, used for comparing the first tag with the plurality of the second tags one by one, and if the similarity or distance value between the first tag and the second tag is greater than a preset threshold, it is determined that the user has entered an addicted state; The alarm module is used to send out an alarm signal if the user enters an addicted state.

8. A computer device, characterized in that: include: a processor and a computer readable storage medium; A processor, suitable for executing a computer program; the computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the user-generated content anti-addiction method as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which is suitable for being loaded by a processor and executing the user-generated content anti-addiction method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the user-generated content anti-addiction method as described in any one of claims 1 to 6.

Images