A power user public opinion risk early warning method and related device

Abstract

This invention provides a method and related apparatus for early warning of public opinion risks among power users. The method includes acquiring text data containing public opinion content within the power industry; identifying preset key texts from the text data and determining the risk assessment coefficient of the current text data based on the frequency of occurrence of the key texts; performing a preliminary risk assessment based on the risk assessment coefficients and issuing a first-type alarm signal when the preliminary risk assessment result indicates the occurrence of risk; inputting all risk assessment coefficients within the time period of the current text data into a prediction model, performing a deep risk assessment based on the predicted risk assessment coefficients, and issuing a second-type alarm signal based on the deep risk assessment result; and providing early warning of the current public opinion situation based on the alarm signals. This invention derives risk assessments through automated collection and analysis of public opinion data and performs further alarm processing based on the assessment results, enabling timely response to emerging public opinion threats.

Description

Technical Field

[0001] This invention belongs to the field of public opinion risk management technology, specifically relating to a method and related device for early warning of public opinion risks for power users. Background Technology

[0002] The power industry is in a phase of rapid development, and accurately grasping public opinion and managing public sentiment in the power sector presents new challenges. The widespread use of instant messaging tools such as Weibo has ushered in the "self-media" era for information dissemination and public communication. The rapid speed and wide reach of online communication have created new difficulties for power sector public opinion management.

[0003] Currently, the management of public opinion mainly relies on manual data collection and analysis. The process from data collection and analysis to risk assessment lacks effective automated tools, resulting in slow processing speed and an inability to respond promptly to emerging public opinion threats. Summary of the Invention

[0004] In view of this, the present invention aims to provide a method and related device for early warning of public opinion risks of power users. By automatically collecting and analyzing public opinion data to obtain risk assessment, and further alarm processing based on the assessment results, it can respond to newly emerging public opinion threats in a timely manner.

[0005] To achieve the above objectives, the technical solution provided by the present invention is as follows:

[0006] In a first aspect, the present invention provides a method for early warning of public opinion risks among electricity users, comprising the following steps:

[0007] Obtain text data containing public opinion content within the power industry;

[0008] The key texts are identified from the text data, and the risk assessment coefficient of the current text data is determined based on the frequency of occurrence of the key texts. The key texts are texts that reflect public opinion content, and the risk assessment coefficient is positively correlated with the frequency.

[0009] A preliminary risk assessment is conducted based on the risk assessment coefficient, and a first-class alarm signal is issued when the preliminary risk assessment result indicates that a risk has occurred.

[0010] Based on all risk assessment coefficients within the time period of the current text data, a risk assessment coefficient prediction sequence is generated and input into a pre-trained prediction model to obtain the risk assessment prediction coefficients.

[0011] Based on the risk assessment prediction coefficient, a deep risk assessment is conducted on the current risk, and a second type of alarm signal is issued to indicate the severity of the risk based on the results of the deep risk assessment.

[0012] Early warnings are issued based on the first and second types of alarm signals to assess the current public opinion situation.

[0013] Furthermore, predefined key text is identified from the text data, including:

[0014] Based on a basic recognition dictionary, key texts are identified from text data, and the frequency of all key texts appearing in the text data is counted. The basic recognition dictionary includes preset key texts and their corresponding weights.

[0015] Furthermore, the risk assessment coefficient of the current text data is determined based on the frequency of occurrence of key texts, including:

[0016] The weights corresponding to key texts are combined with their frequency of occurrence to obtain a comprehensive weight.

[0017] The risk assessment coefficient is obtained by summing the combined weights of all key texts.

[0018] Furthermore, identifying preset key text from text data also includes:

[0019] Divide the text data into several simple sentences;

[0020] Based on a comprehensive recognition dictionary, the predicate key text is identified from several simple sentences. The first part of the predicate key text in the simple sentence is taken as the subject to be identified, and the second part is taken as the object to be identified. The comprehensive recognition dictionary includes the subject key text, the predicate key text, and the object key text, as well as the corresponding weights.

[0021] Traverse the subject and object parts to be identified, filter out simple sentences that simultaneously contain subject key text, predicate key text, and object key text, and count the frequency of the corresponding key text occurrences.

[0022] Furthermore, the risk assessment coefficient of the current text data is determined based on the frequency of occurrence of key texts, including:

[0023] The weights and frequencies of the subject key text, predicate key text, and object key text in each selected simple sentence are combined to obtain the comprehensive weight of each simple sentence.

[0024] The risk assessment coefficient is obtained by summing the combined weights of all the simple sentences.

[0025] Furthermore, the preliminary risk assessment includes:

[0026] The risk assessment coefficient is compared with the preset risk threshold. If it exceeds the preset risk threshold, it is assessed as a risk.

[0027] Furthermore, in-depth risk assessment includes:

[0028] The risk assessment prediction coefficients are matched with preset risk levels to obtain risk prediction levels. Each risk level corresponds to a risk assessment prediction coefficient range and an alarm signal.

[0029] Secondly, the present invention provides a power user public opinion risk early warning device, including...

[0030] The public opinion data collection unit is used to acquire text data containing public opinion content in the power industry;

[0031] The topic extraction unit is used to identify preset key texts from text data and determine the risk assessment coefficient of the current text data based on the frequency of occurrence of key texts. Key texts are texts that reflect public opinion content, and the magnitude of the risk assessment coefficient is positively correlated with the frequency.

[0032] The risk assessment unit is used to conduct a preliminary risk assessment of the current risk based on the risk assessment coefficient;

[0033] The risk level prediction unit is used to generate a risk assessment coefficient prediction sequence based on all risk assessment coefficients within the time period of the current text data, and input it into the pre-trained prediction model to obtain the risk assessment prediction coefficients; it is also used to perform in-depth risk assessment of the current risk based on the risk assessment prediction coefficients.

[0034] The risk processing unit is used to issue a first-class alarm signal when the preliminary risk assessment result indicates that a risk has occurred; it is also used to issue a second-class alarm signal indicating the severity of the risk based on the results of the in-depth risk assessment.

[0035] The early warning unit is used to issue early warnings about the current public opinion situation based on the first type of alarm signal and the second type of alarm signal.

[0036] Accordingly, the present invention provides a computer device, the device including a processor and a memory:

[0037] The memory is used to store computer programs and send the instructions of the computer programs to the processor;

[0038] The processor executes a power user public opinion risk early warning method according to the instructions of the computer program, as described in the first aspect.

[0039] Accordingly, the present invention provides a computer-readable storage medium storing a computer program, which is executed by a processor as described in the first aspect: a method for early warning of public opinion risks among power users.

[0040] In summary, this invention provides a method and related apparatus for early warning of public opinion risks among power users. The method includes: acquiring text data containing public opinion content within the power industry; identifying preset key texts from the text data and determining the risk assessment coefficient of the current text data based on the frequency of occurrence of the key texts (key texts reflecting public opinion content, with a positive correlation between the magnitude and frequency of the risk assessment coefficient); conducting a preliminary risk assessment of the current risk based on the risk assessment coefficients, and issuing a first-type alarm signal when the preliminary risk assessment result indicates the occurrence of risk; generating a risk assessment coefficient prediction sequence based on all risk assessment coefficients within the time period of the current text data, and inputting it into a pre-trained prediction model to obtain risk assessment prediction coefficients; conducting a deep risk assessment of the current risk based on the risk assessment prediction coefficients, and issuing a second-type alarm signal indicating the severity of the risk based on the deep risk assessment result; and issuing an early warning of the current public opinion situation based on the first and second type alarm signals. This invention derives risk assessment through automated collection and analysis of public opinion data, and performs further alarm processing based on the assessment results, enabling timely response to emerging public opinion threats. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart of a power user public opinion risk early warning method provided in an embodiment of the present invention;

[0043] Figure 2 A flowchart for determining risk assessment coefficients based on a comprehensive identification dictionary is provided for embodiments of the present invention;

[0044] Figure 3 This is a block diagram illustrating the composition of a power user public opinion risk early warning device provided in an embodiment of the present invention;

[0045] Figure 4 This is a block diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0047] Please see Figure 1 This embodiment provides a method for early warning of public opinion risks among electricity users, including the following steps:

[0048] S11: Obtain text data containing public opinion content in the power industry.

[0049] It should be noted that this step collects textual data related to public opinion in the power industry. For example, textual data can be collected from the latest information released by the power industry, articles, and online discussions. Textual content can be directly copied from power industry websites and discussion websites (such as Weibo, forums, etc.) and stored to form the text to be processed.

[0050] S12: Identify preset key texts from the text data, and determine the risk assessment coefficient of the current text data based on the frequency of occurrence of the key texts. The key texts are texts that reflect public opinion content, and the magnitude of the risk assessment coefficient is positively correlated with the frequency.

[0051] It should be noted that this step identifies keywords or phrases related to public opinion and quantifies the risk level based on their frequency of occurrence. Natural language processing techniques (such as TF-IDF, word frequency statistics, etc.) can be used to analyze the text and identify key texts related to public opinion.

[0052] S13: Based on the risk assessment coefficient, conduct a preliminary risk assessment of the current risk, and issue a first-class alarm signal when the preliminary risk assessment result indicates that a risk has occurred.

[0053] It should be noted that this step involves a preliminary risk assessment based on a risk rating coefficient, quickly determining the level of public opinion risk based on the frequency of keyword occurrences, and making an immediate response. The first type of warning signal is characterized by rapid response, enabling timely action.

[0054] S14: Generate a risk assessment coefficient prediction sequence based on all risk assessment coefficients within the time period of the current text data, and input it into the pre-trained prediction model to obtain the risk assessment prediction coefficients.

[0055] It should be noted that this step utilizes historical data to predict future public opinion trends and prepare countermeasures in advance. This can be achieved by establishing a time interval and a time period T, repeating the aforementioned steps, and obtaining the risk assessment coefficients K for each fixed time interval within the time period T, thus obtaining a risk assessment coefficient prediction sequence. Based on time series algorithms, future risk assessment data trends are predicted to obtain the risk assessment prediction coefficient K. 预测 .

[0056] S15: Perform a deep risk assessment of the current risk based on the risk assessment prediction coefficient, and issue a second type of alarm signal to indicate the severity of the risk based on the deep risk assessment results.

[0057] It should be noted that this step further assesses the severity of public opinion risks based on the prediction results and issues more detailed alerts. The second type of alert signal contains specific risk levels, which can be matched with specific countermeasures, thereby achieving accurate risk management.

[0058] S16: Issue early warnings about the current public opinion situation based on the first type of alarm signal and the second type of alarm signal.

[0059] It should be noted that this step integrates information from real-time and predictive alerts to form a complete public opinion early warning mechanism. By combining real-time and predictive alert signals, the overall situation of public opinion risk can be comprehensively assessed, and early warning notices can be sent to relevant personnel accordingly.

[0060] This embodiment provides a method for early warning of public opinion risks among power users. It obtains risk assessments through automated collection and analysis of public opinion data, and performs further alarm processing based on the assessment results, which can respond promptly to newly emerging public opinion threats.

[0061] In some embodiments, identifying preset key text from text data includes:

[0062] Based on a basic recognition dictionary, key texts are identified from text data, and the frequency of all key texts appearing in the text data is counted. The basic recognition dictionary includes preset key texts and their corresponding weights.

[0063] It should be noted that the key text can be keywords or key phrases. Below is an example of a basic recognition dictionary:

[0064]

[0065] In a further embodiment, determining the risk assessment coefficient of the current text data based on the frequency of occurrence of key text includes:

[0066] S21: Combine the weights corresponding to key texts with their frequency of occurrence to obtain a comprehensive weight;

[0067] S22: Summing the combined weights of all key texts yields the risk assessment coefficient.

[0068] Below is an example (using excerpts from Weibo posts as an example):

[0069] "#PowerGridMaintenanceComplaints# Speechless! The power grid maintenance outages lately have been going on for way too long! The food in my fridge has all gone bad, and I have to rely on my phone for light at night. Maintenance work is important, but we innocent users shouldn't have to pay for these long power outages! I hope the power company will take this issue seriously and stop making us pay for these 'dark times'! @PowerCompany, don't be so unreasonable! Come and see these complaining netizens!"

[0070] Based on the keywords in the dictionary, the above keywords were identified and their frequencies were recorded for this blog post, as follows:

[0071]

[0072] The risk assessment coefficient is obtained by multiplying the frequency of keyword identification by the corresponding weight and then summing them up. The details are as follows:

[0073]

[0074] in, This represents the risk assessment coefficient. This represents the weight of the i-th keyword / keyword. This indicates the frequency of the i-th keyword / keyword.

[0075] In some embodiments, identifying preset key text from text data further includes:

[0076] S31: Text data can be divided into several simple sentences, which can be done by using punctuation marks as the dividing method.

[0077] S32: Based on a comprehensive recognition dictionary, identify the predicate key text from several simple sentences, and take the first part of the predicate key text in the simple sentence as the subject to be identified and the second part as the object to be identified. The comprehensive recognition dictionary includes the subject key text, the predicate key text and the object key text and their corresponding weights.

[0078] This step first identifies whether the simple sentence contains a predicate from the comprehensive identification dictionary. If it does, the part before the predicate is considered the subject to be identified, and the part after it is considered the object to be identified.

[0079] S33: Traverse the subject and object parts to be identified, filter out simple sentences that simultaneously contain subject key text, predicate key text, and object key text, and count the frequency of the corresponding key text.

[0080] This step involves subject and object identification. Specifically, it iterates through the subject and object parts to be identified, checking if the subject and object exist in the comprehensive identification dictionary. If both exist, the comprehensive weight of the simple sentence is calculated by summing the corresponding weights of the identified subject, predicate, and object. If at least one is not identified, the next simple sentence is identified. Each simple sentence is acquired and its subject, predicate, and object are identified until all simple sentences have been identified.

[0081] It should be noted that the comprehensive recognition dictionary includes three items: "subject," "predicate," and "object," and each item is assigned a corresponding weight; an example is shown below:

[0082]

[0083]

[0084]

[0085] The following simple example illustrates the steps above:

[0086] "Speechless! The power company has been cutting off our electricity for way too long lately! The food in our fridge has gone bad, and we have to rely on our phones for light at night. Maintenance is important, but we innocent users shouldn't have to pay for these long power outages! We hope the power grid will take this issue seriously! @PowerCompany, don't be so unreasonable! Come and see this scathing complaint from netizens!"

[0087] First, the above statement is divided into simple sentences using punctuation marks (this example uses exclamation marks, periods, question marks, semicolons, and ellipses as separators). The processed paragraph is shown below:

[0088] 1. Speechless!

[0089] 2. The power company has been cutting off power for way too long lately!

[0090] 3. All the food in the refrigerator at home has gone bad, and I have to rely on my phone for light at night.

[0091] 4. Maintenance work is important, but we innocent users shouldn't have to pay for long power outages!

[0092] 5. We hope the power grid (subject) will take seriously the problem of our community's power outages (predicate)!

[0093] 6. @Power company, don't go too far!

[0094] 7. Come and see what these netizens are complaining about!

[0095] Next, identify the subject, verb, and object in the above simple sentences. In the examples above, "electricity" and "power grid" are the subjects, "stop" is the verb, and "electricity" is the object. Therefore, simple sentences 2 and 5 are the sentences that need to be selected.

[0096] In a further embodiment, determining the risk assessment coefficient of the current text data based on the frequency of occurrence of key text includes:

[0097] S41: Combine the weights and frequencies of the subject key text, predicate key text, and object key text in each selected simple sentence to obtain the comprehensive weight of each simple sentence;

[0098] S42: Summing the combined weights of all simple sentences yields the risk assessment coefficient.

[0099] Based on the examples in the aforementioned embodiments, the weights of simple sentences 2 and 5 are calculated. The weight of a simple sentence is the sum of the subject weight, predicate weight, and object weight. Then, the weights of simple sentences 2 and 5 are added together to obtain the risk assessment coefficient. The process for determining the risk assessment coefficient based on the comprehensive recognition dictionary is as follows: Figure 2 As shown.

[0100] In some embodiments, the preliminary risk assessment includes:

[0101] The risk assessment coefficient is compared with the preset risk threshold. If it exceeds the preset risk threshold, it is assessed as a risk.

[0102] It should be noted that a risk threshold is set, and the risk assessment coefficient is compared with the risk threshold. If the risk exceeds the risk threshold, it is assessed as a current risk and the first alarm signal of "current risk" is issued.

[0103] In some embodiments, in-depth risk assessment includes:

[0104] The risk assessment prediction coefficients are matched with preset risk levels to obtain risk prediction levels. Each risk level corresponds to a risk assessment prediction coefficient range and an alarm signal.

[0105] It should be noted that, based on K 预测 For example, when predicting risk levels, the following is an example:

[0106]

[0107] The above , , The threshold for judgment can be designed and adjusted according to actual needs. There are no restrictions here. The risk levels of "low", "medium" and "high" are just examples. In actual applications, more levels can be set as needed, or corresponding judgment thresholds can be designed based on the levels.

[0108] Each risk prediction and identification level is assigned a separate alarm signal, namely the second alarm signal, the third alarm signal, ..., the m-th alarm signal, corresponding to the m-1 risk prediction levels (i.e., the second type of alarm signal). For example, if the risk levels are divided into "low", "medium", and "high" as described above, then the second alarm signal, the third alarm signal, and the fourth alarm signal are matched.

[0109] Based on the same inventive concept, this application also provides a power user public opinion risk early warning device for implementing the power user public opinion risk early warning method described above. The solution provided by this system is similar to the solution described in the above method; therefore, the specific limitations in the power user public opinion risk early warning device embodiments provided below can be found in the limitations of the power user public opinion risk early warning method described above, and will not be repeated here.

[0110] Please see Figure 3 This embodiment provides a power user public opinion risk early warning device, including...

[0111] The public opinion data collection unit is used to acquire text data related to public opinion within the power industry. Specifically, it collects the latest information released by the power industry, documents, and online discussions, primarily focusing on text data. This data can be directly copied from power industry websites and discussion websites (such as Weibo and forums) and stored accordingly.

[0112] The topic extraction unit is used to identify preset key texts from text data and determine the risk assessment coefficient of the current text data based on the frequency of occurrence of key texts. Key texts are texts reflecting public opinion content, and the magnitude of the risk assessment coefficient is positively correlated with the frequency. Specifically, this unit mainly outputs risk assessment coefficients, including the current risk assessment coefficient and historical risk assessment coefficients (both the current and historical risk assessment coefficients are obtained in the same way; the only difference between them is the time dimension. That is, the risk assessment coefficient obtained at the current time is the current risk assessment coefficient, and when the risk assessment coefficient at the next moment is output, the risk assessment coefficient at the previous moment and earlier is the historical risk assessment coefficient).

[0113] The extraction process for the topic extraction unit is as follows:

[0114] Construct an identification dictionary, which includes keywords, key terms, and their corresponding weights;

[0115] Get the text content, traverse the text content and identify keywords, key words and their corresponding frequencies.

[0116] The risk assessment unit is used to conduct a preliminary risk assessment of the current risk based on the risk assessment coefficient.

[0117] The risk level prediction unit is used to generate a risk assessment coefficient prediction sequence based on all risk assessment coefficients within the time period of the current text data, and input it into the pre-trained prediction model to obtain the risk assessment prediction coefficients; it is also used to perform in-depth risk assessment of the current risk based on the risk assessment prediction coefficients.

[0118] Specifically, this unit acquires real-time and historical output data from the risk assessment unit, predicts future trends in risk assessment data based on time series algorithms, and then predicts the risk level based on the predicted risk assessment data. Time series algorithms such as ARIMA and Prophet can be used.

[0119] The risk processing unit is used to issue a first-type alarm signal when the preliminary risk assessment indicates the presence of risk; it is also used to issue a second-type alarm signal indicating the severity of the risk based on the results of the in-depth risk assessment. This unit sets a risk processing method for each risk level, receives the risk level output from the risk level prediction unit, selects the appropriate risk processing method, including sending a first alarm signal to the terminal. In addition, it can receive the current risk judgment coefficient from the risk determination unit and compare it with a set threshold; if the threshold is exceeded, a second alarm signal is sent.

[0120] The early warning unit is used to issue early warnings about the current public opinion situation based on the first type of alarm signal and the second type of alarm signal.

[0121] In addition, the above-mentioned device may also include a data preprocessing unit and a terminal.

[0122] The data preprocessing unit serves as an intermediate unit between the public opinion data collection unit and the topic extraction unit, playing a role in further processing the data. Its main purpose is to make the data more standardized. This unit can be set as needed and can be omitted.

[0123] The terminal includes a management terminal and an execution terminal. The management terminal is used to access the aforementioned units and has impact data that modifies the process data of these units, such as judgment thresholds, weights, and identification dictionary contents. The execution terminal is mainly connected to the risk processing unit and is used to receive data (such as alarm signals) from the risk processing unit.

[0124] This embodiment completes data collection, processing, and analysis through a public opinion data collection unit, a data preprocessing unit, and a topic extraction unit. Finally, based on the output of the topic extraction unit, a risk assessment unit calculates a risk rating coefficient to assess the current existence of risk and predict whether future risks will occur. Specifically, the prediction of future risks is achieved through a time series algorithm in the risk level prediction unit. Furthermore, this invention includes a risk processing unit, providing response methods and strategies for each risk level, including sending alarm signals to terminals. Based on this, the technical solution of this invention is sufficient to solve the problem that public opinion control largely relies on manual data collection and analysis, resulting in slow processing speeds.

[0125] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0126] Reference Figure 4 The present invention also provides a computer device, including: a memory and a processor, and a computer program stored in the memory. When the computer program is executed on the processor, it implements the power user public opinion risk early warning method as described in any of the above methods.

[0127] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. This computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 4 The examples of computer devices are merely examples and do not constitute a limitation on computer devices. They may include more or fewer components than shown in the illustration, or combinations of certain components, or different components. For example, they may also include input / output devices, network access devices, etc.

[0128] The processor referred to can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (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. A general-purpose processor can be a microprocessor or any conventional processor.

[0129] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard drive or RAM. In other embodiments, the memory may be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory may include both internal and external storage units of the computer device. The memory is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory can also be used to temporarily store data that has been output or will be output.

[0130] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the power user public opinion risk early warning method as described in any of the above methods.

[0131] In this embodiment, if the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0132] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0133] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0134] In the embodiments disclosed in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0135] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for early warning of public opinion risks among electricity users, characterized in that, Includes the following steps: Obtain text data containing public opinion content within the power industry; Preset key texts are identified from the text data, and the risk assessment coefficient of the current text data is determined based on the frequency of occurrence of the key texts. The key texts are texts that reflect the public opinion content, and the magnitude of the risk assessment coefficient is positively correlated with the frequency. A preliminary risk assessment is performed on the current risk based on the aforementioned risk assessment coefficient, and a first-class alarm signal is issued when the preliminary risk assessment result indicates that a risk has occurred. Based on all risk assessment coefficients within the time period of the current text data, a risk assessment coefficient prediction sequence is generated and input into a pre-trained prediction model to obtain the risk assessment prediction coefficients. Based on the risk assessment prediction coefficient, a deep risk assessment is performed on the current risk, and a second type of alarm signal is issued to indicate the severity of the risk based on the deep risk assessment result; Based on the first type of alarm signal and the second type of alarm signal, an early warning is issued regarding the current public opinion situation; Identifying preset key text from the text data includes: The text data is divided into several simple sentences; Based on a comprehensive recognition dictionary, the predicate key text is identified from several simple sentences, and the first part of the predicate key text in the simple sentence is taken as the subject to be identified, and the second part is taken as the object to be identified. The comprehensive recognition dictionary includes the subject key text, the predicate key text and the object key text, as well as the corresponding weights. Traverse the subject to be identified and the object to be identified, filter out simple sentences that simultaneously contain the subject key text, the predicate key text and the object key text, and count the frequency of the corresponding key texts; The risk assessment coefficient of the current text data is determined based on the frequency of occurrence of the key text, including: The weights and frequencies of the subject key text, predicate key text and object key text in each of the selected simple sentences are combined to obtain the comprehensive weight of each simple sentence. The risk assessment coefficient is obtained by summing the combined weights of all the simple sentences.

2. The method for early warning of public opinion risks among electricity users according to claim 1, characterized in that, The preliminary risk assessment includes: The risk assessment coefficient is compared with a preset risk threshold. If the risk exceeds the preset risk threshold, it is assessed as a risk.

3. The method for early warning of public opinion risks among electricity users according to claim 1, characterized in that, The in-depth risk assessment includes: The risk assessment prediction coefficients are matched with preset risk levels to obtain risk prediction levels. Each risk level corresponds to a risk assessment prediction coefficient range and an alarm signal.

4. A power user public opinion risk early warning device, characterized in that, include: The public opinion data collection unit is used to acquire text data containing public opinion content in the power industry; The topic extraction unit is used to identify preset key texts from the text data and determine the risk assessment coefficient of the current text data based on the frequency of occurrence of the key texts. The key texts are texts that reflect the public opinion content, and the magnitude of the risk assessment coefficient is positively correlated with the frequency. The risk assessment unit is used to perform a preliminary risk assessment of the current risk based on the risk assessment coefficient. The risk level prediction unit is used to generate a risk assessment coefficient prediction sequence based on all risk assessment coefficients within the time period of the current text data, and input it into the pre-trained prediction model to obtain the risk assessment prediction coefficients. It is also used to conduct in-depth risk assessment of the current risk based on the aforementioned risk assessment prediction coefficient; The risk processing unit is used to issue a first-class alarm signal when the preliminary risk assessment result indicates that a risk has occurred. It is also used to issue a second type of alarm signal indicating the severity of the risk based on the results of in-depth risk assessment; The early warning unit is used to issue an early warning about the current public opinion situation based on the first type of alarm signal and the second type of alarm signal; Identifying preset key text from the text data includes: The text data is divided into several simple sentences; Based on a comprehensive recognition dictionary, the predicate key text is identified from several simple sentences, and the first part of the predicate key text in the simple sentence is taken as the subject to be identified, and the second part is taken as the object to be identified. The comprehensive recognition dictionary includes the subject key text, the predicate key text and the object key text, as well as the corresponding weights. Traverse the subject to be identified and the object to be identified, filter out simple sentences that simultaneously contain the subject key text, the predicate key text and the object key text, and count the frequency of the corresponding key texts; The risk assessment coefficient of the current text data is determined based on the frequency of occurrence of the key text, including: The weights and frequencies of the subject key text, predicate key text and object key text in each of the selected simple sentences are combined to obtain the comprehensive weight of each simple sentence. The risk assessment coefficient is obtained by summing the combined weights of all the simple sentences.

5. A computer device, characterized in that, The device includes a processor and a memory: The memory is used to store computer programs and send the instructions of the computer programs to the processor; The processor executes a power user public opinion risk early warning method according to any one of the instructions of the computer program as described in any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is executed by a processor as described in any one of claims 1-3, a method for early warning of public opinion risks among power users.

Images