A government hotline work order and public opinion correlation analysis method, device, equipment and storage medium thereof
By formatting and extracting entities from government hotline work orders and public opinion data, calculating similarity, and generating accurate correlation analysis results, the accuracy problem of correlation analysis between government hotline work orders and public opinion in existing technologies has been solved, and real-time correlation and risk warning have been achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGZHOU BAIYUN DISTRICT GOVERNMENT SERVICES & DATA ADMINISTRATION BUREAU
- Filing Date
- 2026-02-13
- Publication Date
- 2026-05-29
AI Technical Summary
The existing system for linking government hotline work orders with public opinion analysis cannot achieve real-time and accurate correlation, resulting in long-tail missed detections and potential risk omissions.
By acquiring government hotline work orders and public opinion data, preprocessing is performed using preset formatting rules, entities are extracted, the similarity between the addresses and entities involved is calculated, data pairs that meet preset association rules are selected, and association analysis results are generated based on the similarity of core events.
It achieves real-time and accurate correlation between government hotline work orders and public opinion, avoiding long-tail omissions and potential risk oversights, and can comprehensively cover data from multiple platforms and respond quickly.
Smart Images
Figure CN122113028A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of work order and public opinion correlation analysis technology, specifically to a method, device, equipment and storage medium for government hotline work order and public opinion correlation analysis. Background Technology
[0002] With the steady development of my country's economy, culture, and society, the public's willingness and ability to participate in social governance have significantly improved. On the one hand, the public's awareness of their rights has continued to strengthen, and their attention to social affairs has increased. The 12345 government hotline, as a bridge for direct communication between the government and the public, has become a core channel for reflecting social problems and expressing people's demands. Its work order data truthfully records the pain points and public opinions of social operation. On the other hand, the rapid development of Internet technology has restructured the information dissemination landscape. Online social media platforms, with their immediacy, interactivity, and diffusion, have become the main source of public opinion and the main battleground for its dissemination. Social emotions and demands are rapidly converging and fermenting in cyberspace through fragmented information.
[0003] In this context, the correlation analysis between public opinion and the 12345 government hotline work orders has significant strategic value: From a governance perspective, the two data complement each other—government hotline work orders reflect specific problems that have occurred, while public opinion data captures potential risk signals. The combination of the two can construct a full-chain intelligent governance system of "public opinion perception - risk warning - resource allocation - effect evaluation"; From a technical perspective, by deeply mining cross-platform data through technologies such as natural language processing and sentiment analysis, it is possible to identify the evolution patterns and dissemination paths of people's demands, providing accurate basis for government decision-making.
[0004] The current method of using keyword matching to analyze the correlation between government hotline work orders and public opinion is ineffective because the data heterogeneity, expression form and information density of public opinion and government hotline work order data are different. It cannot effectively identify the correlation implied by different expression forms of the same event, lacks the ability to correlate new events in real time, and causes long-tail missed detection and potential risk omission, resulting in low correlation accuracy and reliability. Summary of the Invention
[0005] In view of this, this application provides a method, device, equipment and storage medium for analyzing the correlation between government hotline work orders and public opinion, which solves the technical problem that existing government hotline work order and public opinion correlation analysis cannot perform real-time and accurate correlation analysis.
[0006] The first aspect of this application provides a method for analyzing the correlation between government hotline work orders and public opinion, including: In response to requests for correlation analysis of government hotline work orders and public opinion, obtain government hotline work order data and public opinion data; The government hotline work order data and the public opinion data are preprocessed using preset formatting rules to obtain government hotline work order information and public opinion information with uniform format. Entity extraction is performed on the government hotline work order information and the public opinion information to obtain the corresponding government hotline work order entity and public opinion entity; Based on the government hotline work order entity and the public opinion entity, filter out government hotline work order-public opinion data pairs that conform to the preset association rules; Based on the core event similarity of the aforementioned government hotline work order-public opinion data pairs, the correlation analysis results between government hotline work orders and public opinion are generated.
[0007] Furthermore, the step of filtering and matching government hotline work order-public opinion data pairs that conform to preset association rules based on the government hotline work order entity and the public opinion entity specifically includes: Based on the government hotline work order entity and the public opinion entity, calculate the similarity of the involved address and the similarity of the involved subject between the government hotline work order entity and the public opinion entity; The system filters and matches government hotline work orders and public opinion data pairs whose address similarity is greater than or equal to a preset address matching threshold or whose subject similarity is greater than or equal to a preset subject matching threshold.
[0008] Furthermore, the step of calculating the similarity of the involved address and the similarity of the involved entities based on the government hotline work order entity and the public opinion entity specifically includes: Extract the work order-related address elements of the government hotline work order entity and the public opinion-related address elements of the public opinion entity. If the work order-related address elements and the public opinion-related address elements satisfy the intersection element rule or the basic element rule, then calculate the text similarity between several work order-related address elements and public opinion-related address elements of the same element type. The maximum value of the text similarity is the feature type similarity; the weighted average of all feature type similarities is used to obtain the address similarity. The entities involved in the work orders of the government hotline work order entity are converted into work order entity vectors, and the entities involved in the public opinion entity are converted into public opinion entity vectors, resulting in several pairs of entity vectors. The cosine similarity of the several pairs of entity vectors is then calculated. The maximum value of the cosine similarity is the entity similarity.
[0009] Furthermore, the intersection element rule is as follows: The intersection of the work order-related address element and the public opinion-related address element includes any one of the elements of street, road, and community, and also includes any other element besides street, road, and community; or the intersection of the work order-related address element and the public opinion-related address element includes the point of interest element or the development zone element. The basic element rules are as follows: The intersection of the work order-related address elements and the public opinion-related address elements includes three basic elements: street, road, and community.
[0010] Furthermore, before extracting the work order-related address element of the government hotline work order entity and the public opinion-related address element of the public opinion entity, the process further includes: The work order address of the government hotline work order entity and the public opinion address of the public opinion entity are processed into structured elements to obtain the work order address element and the public opinion address element.
[0011] Furthermore, the generation of correlation analysis results between government hotline work orders and public opinion based on the core event similarity of the government hotline work order-public opinion data pair specifically includes: Calculate the core event similarity of the government hotline work order-public opinion data pair, wherein the core event similarity is calculated based on the core event semantic similarity between the government hotline work order entity and the public opinion entity in the government hotline work order-public opinion data pair; Based on the government hotline work order-public opinion data pairs whose similarity to the core events is greater than the preset reliability threshold, the correlation analysis results between government hotline work orders and public opinion are generated.
[0012] Furthermore, the preset formatting rules include preset body formatting rules and preset address formatting rules; The step of preprocessing the government hotline work order data and the public opinion data using preset formatting rules to obtain government hotline work order information and public opinion information with uniform format specifically includes: Based on preset subject formatting rules, the government hotline work order data and the public opinion data are formatted to obtain work order subjects and public opinion subjects with uniform format. Based on preset address formatting rules, the government hotline work order data and the public opinion data are formatted to obtain uniformly formatted work order-related addresses and public opinion-related addresses.
[0013] Furthermore, the step of extracting entities from the government hotline work order information and the public opinion information to obtain the corresponding government hotline work order entities and public opinion entities specifically includes: Based on the work order entity information in the government hotline work order entity, the corresponding government hotline work order entity is obtained by extracting the entity through preset work order extraction rules. Based on the public opinion entity information in the public opinion entity, the corresponding public opinion entity is obtained by extracting the public opinion entity through preset public opinion extraction rules.
[0014] A second aspect of this application provides a device for analyzing the correlation between government hotline work orders and public opinion, the device comprising: The data acquisition module responds to requests for correlation analysis between government hotline work orders and public opinion by acquiring government hotline work order data and public opinion data.
[0015] The data processing module uses preset formatting rules to preprocess the government hotline work order data and the public opinion data to obtain government hotline work order information and public opinion information with uniform format.
[0016] The entity extraction module extracts entities from the government hotline work order information and the public opinion information respectively, to obtain the corresponding government hotline work order entities and public opinion entities.
[0017] The data matching module filters out government hotline work order-public opinion data pairs that conform to preset association rules based on the government hotline work order entity and the public opinion entity.
[0018] The results generation module generates correlation analysis results between government hotline work orders and public opinion based on the core event similarity of the government hotline work order-public opinion data pair.
[0019] A third aspect of this application provides a computer device, including: a processor and a memory; the processor is connected to the memory, wherein the memory is used to store a computer program, and the processor is used to call the computer program so that the computer device executes the government hotline work order and public opinion correlation analysis method.
[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program adapted to be loaded and executed by a processor, such that a computer device having the processor performs the government hotline work order and public opinion correlation analysis method.
[0021] One of the above technical solutions has the following advantages and effects: One of the methods for analyzing the correlation between government hotline work orders and public opinion in the above scheme includes: In response to requests for correlation analysis between government hotline work orders and public opinion, obtain government hotline work order data and public opinion data; The government hotline work order data and the public opinion data are preprocessed using preset formatting rules to obtain government hotline work order information and public opinion information with uniform format. Entity extraction is performed on the government hotline work order information and the public opinion information to obtain the corresponding government hotline work order entity and public opinion entity; Based on the government hotline work order entity and the public opinion entity, filter out government hotline work order-public opinion data pairs that conform to the preset association rules; Based on the core event similarity of the aforementioned government hotline work order-public opinion data pairs, the correlation analysis results between government hotline work orders and public opinion are generated.
[0022] As can be seen from the above scheme, the aforementioned method for analyzing the correlation between government hotline work orders and public opinion preprocesses government hotline work order data and public opinion data using preset formatting rules, resulting in government hotline work order information and public opinion information with unified formats. This unifies the expression format and information density of public opinion data and government hotline work order data, solving the problems of heterogeneity, inconsistent expression formats, and inconsistent information density between government hotline work order data and public opinion data. Furthermore, by extracting entities from government hotline work order information and public opinion information, the method obtains government hotline work order entities and public opinion entities. The system then filters and matches government hotline work order entities and public opinion entities to obtain government hotline work order-public opinion data pairs that conform to preset association rules, and calculates the core event similarity of the government hotline work order-public opinion data pairs. Based on the core event similarity of the government hotline work order-public opinion data pairs, it generates association analysis results between government hotline work orders and public opinion, thereby achieving accurate association between government hotline work orders and public opinion, avoiding long-tail omissions and potential risk omissions. Through the association analysis results between government hotline work orders and public opinion, the association between public opinion and government hotline work orders can be understood. Attached Figure Description
[0023] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating a method for analyzing the correlation between government hotline work orders and public opinion, as provided in this application. Figure 2 This is a schematic diagram of the structure of a government hotline work order and public opinion correlation analysis device provided in this application; Figure 3 This is a schematic diagram of the structure of a government hotline work order and public opinion correlation analysis device provided in this application. Detailed Implementation
[0024] This application provides a method, apparatus, equipment, and storage medium for analyzing the correlation between government hotline work orders and public opinion, which solves the technical problems of existing government hotline work order and public opinion correlation analysis being unable to achieve real-time and accurate correlation, and having long-tail missed detections and potential risk omissions.
[0025] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this application.
[0026] In the description of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0027] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.
[0028] Please see Figure 1 This is a flowchart illustrating a method for analyzing the correlation between government hotline work orders and public opinion, as proposed in this application.
[0029] Depend on Figure 1 As can be seen, the method for analyzing the correlation between government hotline work orders and public opinion in this embodiment includes: 100. Respond to requests for correlation analysis between government hotline work orders and public opinion, and obtain government hotline work order data and public opinion data.
[0030] It is understandable that the government hotline work order data can originate from the 12345 government hotline information system, obtained through API interfaces or database exports, or from telephone, websites, mini-programs, official accounts, apps, the State Council website, provincial websites, letters, and visits; the public opinion data can originate from public online platforms, such as social media, news portals, forums / Baidu Tieba, etc. However, public opinion data from public online platforms is often unstructured data, which is heterogeneous to the structured government hotline work order data.
[0031] 200. Using preset formatting rules, preprocess the government hotline work order data and the public opinion data to obtain government hotline work order information and public opinion information with uniform format.
[0032] It can be understood that the acquired public opinion data can be either denoised or undenoised. If it is undenoised data, then in some embodiments, the public opinion data may contain some noise information, such as irrelevant symbols, emoticons, advertising links, etc. Preprocessing the public opinion data, including cleaning it, is necessary to remove this noise information and obtain valid public opinion text information. Optionally, this noise information can be matched and deleted using regular expressions. Regular expressions are a powerful tool for matching, finding, replacing, or validating string patterns; by defining pattern rules, they can efficiently handle complex string operations.
[0033] It should be noted that in this implementation, the preset formatting rules include preset body formatting rules and preset address formatting rules.
[0034] Step 200 specifically includes: S201. Based on preset subject formatting rules, perform subject formatting on the government hotline work order data and the public opinion data to obtain work order subjects and public opinion subjects with unified format.
[0035] S202. Based on preset address formatting rules, the government hotline work order data and the public opinion data are formatted to obtain uniform work order addresses and public opinion addresses.
[0036] Among them, the preset subject formatting rules are used to format the subjects involved in the government hotline work order data and public opinion data (such as enterprises, companies, and shops) to ensure that the information of the subjects involved is presented in a unified and standardized format. For example, different expressions for the same shop such as "XX (a certain plaza store)", "XX·a certain plaza store", and "XX a certain store" are unified as "XX (a certain plaza store)".
[0037] The preset address formatting rules aim to establish standardized specifications for address data, accurately identify and standardize the writing of information at the province, city, district, and street levels in the address, and eliminate data inconsistencies caused by differences in address expression (such as "No. 283, Tongsha Road, Tonghe Street, Baiyun District, Guangzhou" versus "No. 283, Tongsha Road (near Tonghe Metro Station), Baiyun District, Guangzhou").
[0038] Specifically, the address data involved in the government hotline work order information and public opinion information follow the "China Place Name and Address Coding Rules" (GB / T 18521) and are arranged in the hierarchy of "province / municipality / autonomous region-city / county-district / town-street / road-house number / village name-building / unit / floor / room number".
[0039] For example, if the government hotline work order data includes "A consumer dispute occurred at 3 pm today at a restaurant on the third floor of Wanda Plaza, Yuncheng East Road, Baiyun District, Guangzhou," then the entity involved in the work order is "the restaurant," and the address of the work order should be formatted as: Restaurant on the third floor of Wanda Plaza, Yuncheng East Road, Sanyuanli Street, Baiyun District, Guangzhou, Guangdong Province.
[0040] It is understandable that in some embodiments, public opinion data may contain non-standard expressions such as colloquialisms and abbreviations used by netizens. By preprocessing the public opinion data, these non-standard expressions can be formatted. For example, the abbreviation of the address involved in the public opinion data, "Baiyun Wanda", can be formatted as "Wanda Plaza, Yuncheng East Road, Sanyuanli Street, Baiyun District, Guangzhou City, Guangdong Province".
[0041] It should be noted that the government hotline work order information includes, but is not limited to: work order number, work order acceptance time, work order content, work order incident time, work order involved parties, and work order involved address. The public opinion information includes, but is not limited to: public opinion link, public opinion release time, public opinion involved parties, public opinion release content, public opinion incident time, public opinion release author, and public opinion involved address. Specifically, the work order content includes the core event involved in the work order, such as the type of problem, problem description, and impact consequences; the public opinion release content includes the core event involved in the public opinion, such as the type of problem, problem description, and impact consequences; the work order time includes the work order acceptance time and the work order incident time; and the public opinion time includes the public opinion release time and the public opinion incident time.
[0042] Specifically, the types of problems include noise pollution, food safety, and street vending; the descriptions of the problems include continuous loudspeaker noise at night, foreign objects in food, and fruit stalls occupying sidewalks; and the consequences include residents being unable to sleep, multiple customers experiencing diarrhea, and pedestrians experiencing inconvenience.
[0043] 300. Entity extraction is performed on the government hotline work order information and the public opinion information respectively to obtain the corresponding government hotline work order entity and public opinion entity.
[0044] In this embodiment, based on the work order entity information in the government hotline work order entity, the corresponding government hotline work order entity is obtained by extracting the entity through preset work order extraction rules. Based on the public opinion entity information in the public opinion entity, the corresponding public opinion entity is obtained by extracting the public opinion entity through preset public opinion extraction rules.
[0045] It should be noted that entity extraction of the government hotline work order information and the public opinion information mainly involves extracting the core entities of the government hotline work order information and the public opinion information, such as the subject involved in the work order, the content involved in the work order, the time involved in the work order, the time of acceptance of the work order, the address involved in the work order, the subject involved in the public opinion, the content of the public opinion release, the time of release of the public opinion, the address involved in the public opinion, and the time involved in the public opinion.
[0046] The entities involved in a work order are generally personal names, store names, organization names, or company names; the content of the work order includes the type of problem, a description of the problem, and its impact and consequences; and the address of the work order is the location where the event involved in the work order occurred.
[0047] Specifically, based on the public opinion entity information in the public opinion entity, entity extraction is performed on the public opinion entity using preset public opinion extraction rules to obtain the corresponding public opinion entity, including: Based on preset subject extraction rules, the subjects involved in the public opinion incident are extracted from the public opinion information; based on preset content extraction rules, the content of the public opinion release is extracted from the public opinion information; based on preset time extraction rules, the time of the public opinion release is extracted from the public opinion information; based on preset address extraction rules, the addresses involved in the public opinion incident are extracted from the public opinion information.
[0048] It is understandable that the extraction of the government hotline work order entity and the public opinion entity can be obtained through deep semantic parsing using a large language model (such as DeepSeek). A large language model is an artificial intelligence model based on deep learning technology, capable of processing and understanding natural language. By simulating human language processing mechanisms, it possesses powerful contextual understanding capabilities and multilingual support.
[0049] 400. Based on the government hotline work order entity and the public opinion entity, filter out government hotline work order-public opinion data pairs that conform to the preset association rules.
[0050] In this embodiment, the similarity of the involved addresses and the similarity of the involved entities are first calculated based on the government hotline work order entity and the public opinion entity. Specifically, this includes the following steps: S401. Extract the work order-related address element of the government hotline work order entity and the public opinion-related address element of the public opinion entity. If the work order-related address element and the public opinion-related address element satisfy the intersection element rule or the basic element rule, then calculate the text similarity between several work order-related address elements and public opinion-related address elements of the same element type.
[0051] It should be noted that the address elements involved in the work order and the address elements involved in public opinion include both the element type and the corresponding text. The maximum text similarity is the element type similarity; a weighted average of all element type similarities is used to obtain the address similarity.
[0052] Understandably, before extracting the work order-related address element of the government hotline work order entity and the public opinion-related address element of the public opinion entity, the following steps are also included: The work order address of the government hotline work order entity and the public opinion address of the public opinion entity are processed into structured elements to obtain work order address elements and public opinion address elements.
[0053] In some embodiments, the mgeo geographic feature annotation model is used to decompose the address into structured features, resulting in feature types such as province, city, district, street, road or community and their corresponding text.
[0054] To reduce interference caused by data heterogeneity and inconsistent expression, standard stop words in addresses will be removed. For example, the three high-level administrative division elements of province, city, and district will be removed, such as "Guangdong Province, Guangzhou City, Baiyun District" or "Guangzhou City, Baiyun District". For street, road, or community type elements, their text will be further cleaned up, and common suffixes such as "street", "town", "township", "community residents' committee", "residents' committee", "office", "road", "number", and "community" will be removed, retaining only the core name.
[0055] Furthermore, the intersection element rule is as follows: The intersection of the work order-related address elements and the public opinion-related address elements includes any one of the elements of street, road, or community, and also includes any other element besides street, road, or community; or the intersection of the work order-related address elements and the public opinion-related address elements includes the element of point of interest or the element of development zone.
[0056] Among these, any of the other elements includes province, city, district, county, township, village, house number, building number, and landmark building name; points of interest refer to landmark entities with specific names and spatial locations, such as schools, hospitals, and shopping malls; development zones refer to specific areas with clear boundaries in terms of administrative jurisdiction or functional division, such as economic and technological development zones, high-tech industrial development zones, and industrial parks. Compared with elements such as streets, roads, and communities, point of interest and development zone elements have higher spatial reference precision or specific administrative / functional attributes, and can effectively solve matching ambiguities caused by duplicate road names or excessively long road segments.
[0057] The basic element rules are as follows: The intersection of the address elements involved in the work order and the address elements involved in public opinion includes three basic elements: street, road and community.
[0058] Specifically, the text similarity between several work order-related address elements and public opinion-related address elements of the same element type is calculated. This involves identifying the text of the same element type among the work order-related and public opinion-related address elements, and then calculating the text similarity for that element type. For example, if both the work order-related and public opinion-related addresses contain street and road elements, the text under the street element of both addresses is retrieved, and the street element text similarity is calculated. Similarly, the text under the road element of both addresses is retrieved, and the road element text similarity is calculated. It is understood that if the work order-related and public opinion-related addresses also include community elements, the community element text similarity is calculated in the same way, which will not be elaborated further here.
[0059] It should be noted that there may be multiple text similarities for street elements, road elements, and community elements. The maximum text similarity for a given element is then taken as the element type similarity. A weighted average of all element type similarities is then calculated to obtain the address similarity. Each element type has an equal weight.
[0060] S402. Convert the entities involved in the work order in the government hotline work order entity into work order entity vectors, and convert the entities involved in the public opinion in the public opinion entity into public opinion entity vectors to obtain several pairs of entity vectors. Calculate the cosine similarity of the several pairs of entity vectors; where the maximum value of the cosine similarity is the entity similarity.
[0061] Understandably, before converting the entities involved in government hotline work orders into vectors, redundant whitespace, punctuation, and common stop words such as "Guangzhou Baiyun District," "street," "road," and "number" are removed from the entities involved in the work orders. Similarly, the same operation is performed on the entities involved in public opinion incidents to avoid interference caused by data heterogeneity and inconsistent expression, which could affect accuracy.
[0062] Specifically, in some embodiments, the Word2Vec model of text2vec can be used to convert the entities involved in work orders and the entities involved in public opinion into vectors.
[0063] After calculating the similarity of the address involved and the similarity of the subject involved, the government hotline work order-public opinion data pairs with the address similarity greater than or equal to the preset address matching threshold or the subject similarity greater than or equal to the preset subject matching threshold are selected and matched.
[0064] In some embodiments, the preset address matching threshold is set to 0.7-0.9, preferably 0.75; the preset subject matching threshold is set to 0.8-0.95, preferably 0.9.
[0065] It is understandable that the government hotline work order-public opinion data pair can be one or more pairs.
[0066] 500. Based on the core event similarity of the aforementioned government hotline work order-public opinion data pair, generate the correlation analysis results between government hotline work orders and public opinion.
[0067] In this embodiment, the core event similarity is calculated based on the semantic similarity of the core events of the government hotline work order entity and the public opinion entity in the government hotline work order-public opinion data pair.
[0068] The core event similarity focuses on determining whether the content of the work order and the content of the public opinion event belong to the same event. In this embodiment, the content of the work order includes the problem type, problem description, and impact consequences described by the government hotline work order entity, while the content of the public opinion event includes the problem type, problem description, and impact consequences described by the public opinion event entity. The semantic similarity of the core event is obtained by weighted fusion of the semantic similarity of the work order content in the government hotline work order data and the problem type, problem description, and impact consequences described by the public opinion content in the public opinion data.
[0069] For example: In May 2023, a barbecue restaurant operating late at night in a commercial street in Tonghe Subdistrict, Baiyun District, Guangzhou City, generated noise and fumes, and simultaneously caused: Government hotline work order data: Problem type: Noise pollution.
[0070] Problem description: The barbecue restaurant uses loudspeakers to attract customers from 1 a.m. to 3 a.m. every day, accompanied by loud noises from diners.
[0071] Consequences: Residents in surrounding communities have been unable to get proper rest for a long time, with many reporting insomnia.
[0072] Public opinion data: Problem type: restaurant noise and oil fume pollution.
[0073] Problem description: A barbecue restaurant near Tonghe subway station is playing music late at night, and customers are playing drinking games and making a lot of noise. The range hood is also making a loud humming sound.
[0074] Consequences: The upstairs residents couldn't sleep all night, and the children were listless at school the next day.
[0075] Based on the above government hotline work order data and public opinion data, a semantic big data model was used to calculate the semantic similarity between the content of the work orders and the content of the public opinion, resulting in: The semantic similarity of the problem type is Sim1=1.0, the semantic similarity of the problem description is Sim2=0.98, and the semantic similarity of the impact and consequences is Sim3=0.95.
[0076] If the weight of semantic similarity of question type is set to 0.5, the weight of semantic similarity of question description is set to 0.3, and the weight of semantic similarity of impact consequence is set to 0.2, then the core event similarity of the government hotline work order-public opinion data pair is (0.5*1.0+0.3*0.98+0.2*0.95)=0.98.
[0077] In this embodiment, after calculating the core event similarity of the government hotline work order-public opinion data pair, the core event similarity of the government hotline work order-public opinion data pair is compared with a preset reliability threshold. If the core event similarity is greater than the preset reliability threshold, the data of the government hotline work order-public opinion data pair is subjected to association analysis to generate the association analysis results of government hotline work orders and public opinion.
[0078] As shown above, if the preset confidence threshold is 0.8, the core event similarity calculation result of the above government hotline work order-public opinion data pair is 0.98, which is higher than the preset confidence threshold of 0.8. Therefore, the government hotline work order-public opinion data pair is judged as a high confidence association pair, thereby generating the association analysis result between government hotline work orders and public opinion.
[0079] It should be noted that the core event similarity reflects the strength of the correlation between government hotline work order data and public opinion data. The higher the core event similarity, the stronger the correlation between government hotline work order data and public opinion data.
[0080] In this implementation, the results of the correlation analysis between government hotline work orders and public opinion include the associated government hotline work order information, associated public opinion information, and the similarity of the involved addresses, involved entities, core events, semantic similarity, and public opinion heat of the government hotline work order-public opinion data pair. Specifically, the associated government hotline work order information includes, but is not limited to, work order number, work order processing time, work order incident time, work order incident content, work order involved entity, work order involved address, and work order spatiotemporal information; the associated public opinion information includes, but is not limited to, public opinion link, public opinion release time, public opinion release content, public opinion involved entity, public opinion involved address, public opinion release author, and public opinion spatiotemporal information.
[0081] The government hotline work order and public opinion correlation analysis method described in this embodiment preprocesses government hotline work order data and public opinion data to obtain standardized government hotline work order information and public opinion information, thereby unifying the expression form and information density of government hotline work order data and public opinion data. This solves the problem of cross-source matching difficulties caused by the heterogeneity, inconsistent expression form, and inconsistent information density of government hotline work order and public opinion data. Furthermore, by extracting entities from government hotline work order information and public opinion information, government hotline work order entities and public opinion entities are obtained, based on... The system filters and matches government hotline work order entities and public opinion entities to obtain government hotline work order-public opinion data pairs that conform to preset association rules. It then calculates the core event similarity of these data pairs and generates association analysis results based on this similarity. This achieves real-time and accurate association between government hotline work orders and public opinion, avoiding long-tail omissions and potential risk oversights. The association analysis results provide insight into the relationship between public opinion and government hotline work orders. Furthermore, the government hotline work order-public opinion association analysis method described in this invention can perform association analysis on government hotline work order data from multiple platforms and public opinion data from public network platforms. This avoids the inefficiency caused by manual cross-validation due to the large scale of cross-platform data, achieving comprehensive coverage and rapid response across platforms.
[0082] The above is an embodiment of a method for analyzing the correlation between government hotline work orders and public opinion provided in this application. The following is a schematic diagram of a device for analyzing the correlation between government hotline work orders and public opinion provided in this application. Please refer to [link / reference]. Figure 2 .
[0083] Please see Figure 2 This application provides a schematic diagram of the structure of a government hotline work order and public opinion correlation analysis device. The device includes: The data acquisition module 10 is used to acquire government hotline work order data and public opinion data in response to requests for correlation analysis between government hotline work orders and public opinion data.
[0084] The data processing module 20 uses preset formatting rules to preprocess the government hotline work order data and the public opinion data to obtain government hotline work order information and public opinion information with uniform format.
[0085] The entity extraction module 30 extracts entities from the government hotline work order information and the public opinion information respectively to obtain the corresponding government hotline work order entities and public opinion entities.
[0086] The data matching module 40 filters out government hotline work order-public opinion data pairs that conform to preset association rules based on the government hotline work order entity and the public opinion entity.
[0087] The result generation module 50 generates the correlation analysis results between government hotline work orders and public opinion based on the core event similarity of the government hotline work order-public opinion data pair.
[0088] It is understood that, in another embodiment, the data processing module 20 includes: The subject formatting submodule formats the government hotline work order data and the public opinion data based on preset subject formatting rules to obtain work order subjects and public opinion subjects with uniform format.
[0089] The address formatting submodule formats the addresses of the government hotline work order data and the public opinion data based on preset address formatting rules, so as to obtain uniform work order addresses and public opinion addresses.
[0090] In this embodiment of the invention, the entity extraction module 30 includes: The government hotline work order entity extraction submodule extracts the corresponding government hotline work order entity based on the work order entity information in the government hotline work order entity and through preset work order extraction rules.
[0091] The public opinion entity extraction submodule extracts the public opinion entity based on the public opinion entity information in the public opinion entity and through preset public opinion extraction rules to obtain the corresponding public opinion entity.
[0092] In this embodiment of the invention, the data matching module 40 includes: The subject similarity calculation submodule calculates the similarity of the involved address and the similarity of the involved subject between the government hotline work order entity and the public opinion entity based on the government hotline work order entity and the public opinion entity. The public opinion data filtering submodule filters and matches government hotline work orders and public opinion data pairs whose similarity to the address involved is greater than or equal to a preset address matching threshold or whose similarity to the subject involved is greater than or equal to a preset subject matching threshold.
[0093] In this embodiment of the invention, the result generation module 50 includes: The core event similarity calculation submodule is used to calculate the core event similarity of the government hotline work order-public opinion data pair; The association analysis result generation submodule generates association analysis results between government hotline work orders and public opinion data pairs based on the core event similarity exceeding a preset reliability threshold. The core event similarity is calculated by weighted fusion of the semantic similarity of the core events of the government hotline work order entity and the public opinion entity within the government hotline work order-public opinion data pair.
[0094] In this embodiment, the government hotline work order and public opinion correlation analysis device uses a data processing module to acquire government hotline work order data and public opinion data. The data processing module preprocesses the government hotline work order data and public opinion data to obtain government hotline work order information and public opinion information in a unified format. This unifies the expression form and information density of the government hotline work order data and public opinion data, solving the problem of cross-source matching difficulties caused by the heterogeneity, inconsistent expression form, and inconsistent information density of government hotline work order and public opinion data. The entity extraction module extracts entities from the government hotline work order information and public opinion information to obtain... The system identifies government hotline work order entities and public opinion entities. Based on these entities, it filters and matches data pairs that conform to preset association rules, calculating the core event similarity of each pair. Based on this similarity, it generates association analysis results between government hotline work orders and public opinion, achieving real-time and accurate association between public opinion data and government hotline work order data. This avoids long-tail omissions and potential risk oversights. The association analysis results provide a deeper understanding of the relationship between government hotline work orders and public opinion. Furthermore, the government hotline work order and public opinion association analysis device described in this embodiment can perform association analysis on government hotline work order data from multiple platforms and public opinion data from public network platforms. This avoids the inefficiency caused by manual cross-validation due to the large scale of cross-platform data, achieving comprehensive coverage and rapid response of cross-platform data.
[0095] Please see Figure 3 This application also provides a device for analyzing the correlation between government hotline work orders and public opinion, the device including a processor A and a memory B; Memory B is used to store program code and transfer the program code to processor A.
[0096] Processor A is used to execute the steps of the government hotline work order and public opinion correlation analysis method in the aforementioned embodiments according to the instructions in the program code.
[0097] This application also provides a computer device, including a processor and a memory; the processor is connected to the memory, wherein the memory is used to store a computer program, and the processor is used to call the computer program so that the computer device executes the government hotline work order and public opinion correlation analysis method.
[0098] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0099] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0100] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0101] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.
Claims
1. A method for analyzing the correlation between government hotline work orders and public opinion, characterized in that, The correlation analysis method includes: In response to requests for correlation analysis between government hotline work orders and public opinion, obtain government hotline work order data and public opinion data; The government hotline work order data and the public opinion data are preprocessed using preset formatting rules to obtain government hotline work order information and public opinion information with uniform format. Entity extraction is performed on the government hotline work order information and the public opinion information to obtain the corresponding government hotline work order entity and public opinion entity; Based on the government hotline work order entity and the public opinion entity, filter and match government hotline work order-public opinion data pairs that conform to preset association rules; Based on the core event similarity of the aforementioned government hotline work order-public opinion data pairs, the correlation analysis results between government hotline work orders and public opinion are generated.
2. The method for analyzing the correlation between government hotline work orders and public opinion according to claim 1, characterized in that, The step of filtering and matching government hotline work order-public opinion data pairs that conform to preset association rules based on the government hotline work order entity and the public opinion entity specifically includes: Based on the government hotline work order entity and the public opinion entity, calculate the similarity of the involved address and the similarity of the involved subject between the government hotline work order entity and the public opinion entity; The system filters and matches government hotline work orders and public opinion data pairs whose address similarity is greater than or equal to a preset address matching threshold or whose subject similarity is greater than or equal to a preset subject matching threshold.
3. The method for analyzing the correlation between government hotline work orders and public opinion according to claim 2, characterized in that, The calculation of the similarity between the involved addresses and the involved entities based on the government hotline work order entity and the public opinion entity specifically includes: Extract the work order-related address elements of the government hotline work order entity and the public opinion-related address elements of the public opinion entity. If the work order-related address elements and the public opinion-related address elements satisfy the intersection element rule or the basic element rule, then calculate the text similarity between several work order-related address elements and public opinion-related address elements of the same element type. The maximum value of the text similarity is the feature type similarity; the weighted average of all feature type similarities is used to obtain the address similarity. The entities involved in the work orders of the government hotline work order entity are converted into work order entity vectors, and the entities involved in the public opinion entity are converted into public opinion entity vectors, resulting in several pairs of entity vectors. The cosine similarity of the several pairs of entity vectors is then calculated. The maximum value of the cosine similarity is the entity similarity.
4. The method for analyzing the correlation between government hotline work orders and public opinion according to claim 3, characterized in that, The rules for the intersection elements are as follows: The intersection of the work order-related address element and the public opinion-related address element includes any one of the elements of street, road, and community, and also includes any other element besides street, road, and community; or the intersection of the work order-related address element and the public opinion-related address element includes the point of interest element or the development zone element. The basic element rules are as follows: The intersection of the work order-related address elements and the public opinion-related address elements includes three basic elements: street, road, and community.
5. The method for analyzing the correlation between government hotline work orders and public opinion according to claim 3, characterized in that, Before extracting the work order-related address elements of the government hotline work order entity and the public opinion-related address elements of the public opinion entity, the method further includes: The work order address of the government hotline work order entity and the public opinion address of the public opinion entity are processed into structured elements to obtain the work order address element and the public opinion address element.
6. The method for analyzing the correlation between government hotline work orders and public opinion according to any one of claims 1 to 5, characterized in that, The generation of correlation analysis results between government hotline work orders and public opinion based on the core event similarity of the government hotline work order-public opinion data pairs specifically includes: Calculate the core event similarity of the government hotline work order-public opinion data pair, wherein the core event similarity is calculated based on the core event semantic similarity between the government hotline work order entity and the public opinion entity in the government hotline work order-public opinion data pair; Based on the government hotline work order-public opinion data pairs whose similarity to the core events is greater than the preset reliability threshold, the correlation analysis results between government hotline work orders and public opinion are generated.
7. The method for analyzing the correlation between government hotline work orders and public opinion according to claim 1, characterized in that, The preset formatting rules include preset body formatting rules and preset address formatting rules; The process of preprocessing the government hotline work order data and the public opinion data using preset formatting rules to obtain government hotline work order information and public opinion information with uniform formatting specifically includes: Based on preset subject formatting rules, the government hotline work order data and the public opinion data are formatted to obtain work order subjects and public opinion subjects with uniform format. Based on preset address formatting rules, the government hotline work order data and the public opinion data are formatted to obtain uniformly formatted work order-related addresses and public opinion-related addresses.
8. A device for analyzing the correlation between government hotline work orders and public opinion, characterized in that, The device includes: The data acquisition module is used to obtain government hotline work order data and public opinion data; The data processing module preprocesses the government hotline work order data and the public opinion data to obtain government hotline work order information and public opinion information in a standardized format. The work order extraction module extracts entities from the government hotline work order information to obtain government hotline work order entities; The public opinion extraction module extracts entities from the public opinion information to obtain public opinion entities; The data matching module filters and matches government hotline work order-public opinion data pairs that conform to preset association rules based on the government hotline work order entity and the public opinion entity. The results generation module generates correlation analysis results between government hotline work orders and public opinion based on the core event similarity of the government hotline work order-public opinion data pair.
9. A computer device, characterized in that, include: Processor and memory; The processor is connected to a memory, wherein the memory is used to store a computer program, and the processor is used to invoke the computer program to cause the computer device to perform the method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any one of claims 1-7.