An artificial intelligence-based public service platform system

Abstract

The application provides an artificial intelligence-based public service platform system, aiming to realize accurate service recommendation for user demand through multiple modules. The system includes a user interaction module for receiving user text, voice, image and other input request data, and combining time and location information processing. The data acquisition module is responsible for collecting multi-source heterogeneous city public service data. Based on the collected data, the knowledge graph construction module constructs a dynamic knowledge graph containing public service entities, user entities and their attributes. The intelligent analysis module uses the knowledge graph for time series analysis, spatial analysis and cross-domain correlation analysis, generating service demand prediction and spatial demand distribution results. The personalized recommendation module provides the user with the most suitable target service and associated service combination through demand matching, location optimization and service combination recommendation units. The system can improve the response efficiency of public services and provide personalized, multi-field service recommendations for users.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a public service platform system based on artificial intelligence. Background Technology

[0002] In existing technologies, public service platform systems typically provide services through manual operation or basic data processing methods. These systems collect basic user information and service needs, and provide corresponding public services according to preset rules. With the development of information technology, some systems have begun to use simpler algorithms for data analysis and processing to improve service efficiency and accuracy. However, most of these systems are based on data from a single source, and their response to user needs is limited to specific service areas, failing to effectively integrate information from different data sources.

[0003] However, existing public service platforms face several limitations in addressing the increasingly complex demands of urban services. First, user service requests are often diverse and dynamic, making it difficult for existing systems to fully understand these complex needs. Second, existing systems rely heavily on single or fixed data sources, hindering the integration of multi-source, heterogeneous data and resulting in low accuracy in service recommendations. Furthermore, these systems typically lack cross-domain collaborative analysis when providing services, failing to fully explore the connections between different public services.

[0004] Therefore, it is necessary to develop a new public service platform system based on artificial intelligence. Summary of the Invention

[0005] This application provides an artificial intelligence-based public service platform system that offers users personalized, multi-domain, and precise service recommendations.

[0006] This application provides a public service platform system based on artificial intelligence, including:

[0007] The user interaction module is used to receive user input request data, wherein the input request data includes text, voice, images, and time and location information of the user input.

[0008] The data acquisition module, connected to the user interaction module, is used to collect multi-source heterogeneous urban public service data, including historical service record data, user demographic data, and urban event and policy change data.

[0009] A knowledge graph construction module, connected to the data acquisition module, is used to collect urban public service data and construct a public service knowledge graph. The public service knowledge graph includes public service entities and their attributes, user entities and their attributes, time and geographical location information of public service entities and user entities, and the relationship and interaction history between public service entities and user entities.

[0010] An intelligent analysis module, connected to the knowledge graph construction module, is used to generate service demand analysis results by combining user input request data and the constructed public service knowledge graph; wherein, the intelligent analysis module includes:

[0011] The time series analysis unit is used to perform time series analysis based on historical data and user input request data in the constructed public service knowledge graph, and generate service demand prediction results.

[0012] The spatial analysis unit is used to perform spatial analysis based on geographic location information in the constructed public service knowledge graph and time information at the time of user input, generating spatial demand distribution results; and

[0013] The cross-domain association analysis unit is used to perform cross-domain association analysis based on the potential associations between different public service entities in the constructed public service knowledge graph, and generate cross-domain service combination recommendation strategies.

[0014] A personalized recommendation module, connected to the intelligent analysis module, is used to provide users with accurate service recommendations based on the results of time-series analysis, spatial analysis, and cross-domain correlation analysis performed by the intelligent analysis module; wherein, the personalized recommendation module includes:

[0015] The demand matching unit is used to filter out the target service that best matches the user's input request from the service demand prediction results generated by the time series analysis unit, and to recommend the target service first.

[0016] The location optimization unit, based on the spatial demand distribution results generated by the spatial analysis unit and combined with the user's geographic location information, prioritizes recommending relevant services in or near the user's location; and

[0017] The service composition recommendation unit is used to recommend multiple related public service combinations to users based on the service composition strategy generated by the cross-domain association analysis unit.

[0018] The technical solution provided in this application has the following beneficial effects:

[0019] (1) Improve tag accuracy: By combining time and geographic location information through the user interaction module, the intelligent analysis module can perform time-series analysis, spatial analysis, and cross-domain correlation analysis based on knowledge graphs, providing users with accurate service recommendations that meet their personalized needs, significantly improving the targeting of services and user experience. (2) The data collection module can collect various information such as historical service records, population statistics, urban events, and policy changes from multi-source heterogeneous data, ensuring that the system can analyze user needs from different dimensions and make effective responses, enhancing the comprehensiveness and reliability of data analysis. (3) Through time-series and spatial analysis, the intelligent analysis module can predict service needs in specific times and regions in the future, helping the system to optimize resource allocation and service delivery methods in advance, avoiding resource waste, and improving the efficiency and response speed of public services. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of a public service platform system based on artificial intelligence provided in the first embodiment of this application. Detailed Implementation

[0021] Many specific details are set forth in the following description to provide a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this application; therefore, this application is not limited to the specific embodiments disclosed below.

[0022] The first embodiment of this application provides a public service platform system based on artificial intelligence. Please refer to... Figure 1 This figure is a schematic diagram of the first embodiment of this application. The following is in conjunction with... Figure 1 The first embodiment of this application provides a detailed description of a public service platform system based on artificial intelligence.

[0023] The system includes a user interaction module 101, a data acquisition module 102, a knowledge graph construction module 103, an intelligent analysis module 104, and a personalized recommendation module 105.

[0024] User interaction module 101 is used to receive user input request data, wherein the input request data includes text, voice, image input by the user, as well as time information and location information when the user inputs.

[0025] Furthermore, the user interaction module is specifically used for:

[0026] It receives various forms of input request data from users, including text input, voice input, and image input; and uses natural language processing technology to recognize and parse text and voice, and uses image recognition technology to parse images or scanned documents uploaded by users to generate structured request information.

[0027] While the user is inputting a request, the system captures and records the user's time and location information, and transmits it along with the input request data to the data processing module.

[0028] The user interaction module 101 receives various forms of user input, including text, voice, and images. To process these different input formats, the module employs Natural Language Processing (NLP) and image recognition technologies. For text and voice input, the module first parses the input using NLP. Text input directly extracts key information, such as the type, details, and time of the service request. For voice input, the system first uses speech recognition to convert speech into text, then further parses the text using NLP to generate structured service request information. For image input, the module processes user-uploaded images or scanned documents using image recognition technology. This technology automatically identifies key information in images, such as forms, documents, and tickets, and extracts relevant content, transforming unstructured image data into structured request information.

[0029] Furthermore, the user interaction module 101 automatically captures and records the time and location information of the user's input request. The time information is typically generated automatically by the system's timestamp, accurate to the second. The location information is obtained through the user's device's GPS function or network location services. The user's time and location information are packaged together with the user's input request data and transmitted to the data processing module for subsequent data analysis and service recommendations. This information helps the system understand the context of the user's service request, such as when and where the request was made, thus enabling more accurate matching of appropriate services.

[0030] Through the multimodal data processing of the user interaction module 101, the system can comprehensively process user requests in different forms, such as text, voice, and images, and automatically capture the user's time and location during input to ensure data accuracy and integrity. This enables the system to understand user needs from multiple dimensions and provide more personalized and precise services.

[0031] The data acquisition module 102 is connected to the user interaction module and is used to collect multi-source heterogeneous urban public service data, including historical service record data, user population statistics data, and urban event and policy change data.

[0032] Furthermore, the data acquisition module is specifically used for:

[0033] Data related to urban public services is collected from multiple data sources, including historical service record databases, user demographic databases, and systems for government policies and urban events.

[0034] Identify and process heterogeneous data from different data sources, including structured data, semi-structured data, and unstructured data;

[0035] Through data transformation and standardization steps, the collected data is converted into a unified format to ensure that all data can be effectively used by the system.

[0036] The data acquisition module 102 plays a crucial role in the AI-based public service platform system, responsible for collecting data related to urban public services from multiple diverse data sources. This module connects to the user interaction module 101, initiating the data acquisition process by receiving user input and automated demand identification. The data sources are very broad, typically including but not limited to historical service record databases, user demographic databases, government policy update systems, and urban event management systems. To ensure effective coverage of all dimensions of urban services, the data acquisition module needs robust multi-source heterogeneous data processing capabilities.

[0037] In its implementation, the data acquisition module first collects data from various data sources. The historical service record database provides long-term information on public service usage, such as the types of services requested by citizens over a past period, service completion times, and service satisfaction levels. By analyzing these historical records, the system can understand the usage trends and demand distribution of different public services. The user demographic database provides demographic data on urban residents, including age, gender, occupation, and income levels. This data plays a crucial role in accurately providing personalized public services. The government policy system and the urban event management system provide the latest policy changes, emergencies, and other background information related to public services. This data reflects the dynamic changes in the urban environment in real time, helping the system make corresponding adjustments.

[0038] The data acquisition module also has the capability to process data in different formats. Data acquired from various sources is often heterogeneous, meaning it may exist in structured, semi-structured, or unstructured forms. Structured data, for example, comes from traditional database systems and has clearly defined fields and formats. Semi-structured data, such as XML files or JSON formatted data, while having some structure, cannot be used directly like relational databases. Unstructured data may come from social media or documents containing text, images, videos, etc., and usually lacks a fixed format. The data acquisition module, through built-in recognition algorithms and data transformation tools, can effectively identify these different forms of data and transform them into a unified format that the system can process. This transformation step includes not only simple data format conversion but also complex data parsing and standardization processes to ensure that all data can be used uniformly within the system.

[0039] To ensure data consistency and integrity, the data acquisition module standardizes all collected data. Through standardization, the module integrates heterogeneous data into a unified format, removes redundant information, and ensures that all information can be effectively processed and analyzed by the system within the same framework. For example, historical service records and user demographic data may use different time formats or geographic information representation methods. The data acquisition module converts this inconsistent information into a unified time and location format, ensuring effective matching and association of content from different data sources during subsequent intelligent analysis. Furthermore, the data acquisition module ensures data integrity. If missing or incomplete information is found in some data sources during the data acquisition process, the module will automatically supplement the missing data using data completion algorithms or from other relevant data sources.

[0040] Finally, the data acquisition module transmits the transformed and standardized data to the knowledge graph construction module 103. The latter then uses this data to construct a knowledge graph containing user entities, public service entities, and their relationships, serving as the foundation for intelligent analysis. Through this multi-layered, multi-source heterogeneous data acquisition and processing, the system can dynamically analyze and intelligently recommend urban public service needs on a comprehensive and accurate basis.

[0041] The knowledge graph construction module 103 is connected to the data acquisition module and is used to collect urban public service data to construct a public service knowledge graph. The public service knowledge graph includes public service entities and their attributes, user entities and their attributes, time and geographical location information of public service entities and user entities, and the relationship and interaction history between public service entities and user entities.

[0042] Furthermore, the knowledge graph construction module is specifically used for:

[0043] Based on urban public service data collected from the data acquisition module, public service entities and user entities are identified;

[0044] The identified public service entities and user entities are assigned corresponding attributes. The attributes of public service entities include service type, service provider, service description, timeliness, and availability. The attributes of user entities include demographic information, service preferences, and interaction history.

[0045] Based on historical interaction data between user entities and public service entities, relationships between entities are constructed; wherein, the relationships include user service requests, user service participation, user feedback, service delivery timeliness, and service availability.

[0046] The main function of the knowledge graph construction module 103 is to construct a knowledge graph that accurately reflects the relationship between public service entities and user entities based on the urban public service data obtained from the data acquisition module 102. This module first identifies two core entities in the system: public service entities and user entities. Public service entities refer to various services provided in the city, such as transportation, healthcare, education, and government services; each service is considered an entity. User entities represent residents of the city or individuals who use these public services.

[0047] Once these entities are identified, the system assigns them their respective attributes. For public service entities, the knowledge graph construction module can assign attributes including service type (such as medical services, education services, etc.), service provider (such as a specific government department or third-party organization), service description (detailing the service content), timeliness (whether the service is valid within a specific time), and service availability (the current state of the service, whether it can be provided). These attribute settings enable the knowledge graph to comprehensively and clearly display the characteristics and functions of each public service entity.

[0048] For user entities, the knowledge graph construction module assigns attributes to each user based on demographic information, service usage preferences, and interaction history obtained from the data collection module. Demographic information includes the user's age, gender, occupation, and place of residence; service preferences are information derived from the analysis of the user's historical service request records, helping the system understand which types of public services the user tends to use; and interaction history records all past interactions between the user and public service entities, such as when the user used which services, the completion status of the services, and the user's feedback.

[0049] Next, the knowledge graph construction module will build relationships between user entities and public service entities based on historical interaction data. These relationships include not only simple records of user service requests, but also the process of user participation in services, such as the start time, duration, and end status of a service. The module will also record user feedback on services, which can be based on user ratings, comments, or complaints about service quality. The timeliness and availability of service provision are also key relationship elements, indicating whether the service can be provided on time when requested by the user, and whether the service is available or when it will be restored to an available state.

[0050] Through these processes, the knowledge graph construction module gradually builds a dynamic and updatable graph, demonstrating how urban public services interact with users. This knowledge graph not only includes each entity and its attributes but also shows the complex relationships between entities, such as which services are most popular during specific time periods, which users tend to use certain types of services, and user satisfaction or feedback on services. This enables the module to provide a solid data foundation for other modules in the system (such as the intelligent analysis module 104) to support subsequent time-series analysis, spatial analysis, and cross-domain correlation analysis.

[0051] In summary, the knowledge graph construction module 103 visualizes various public services in the city and their relationships with users in a structured and dynamic manner. This ensures that the system can monitor the usage of public services in real time and provides detailed and accurate data support for subsequent intelligent analysis and precise service recommendations. The working principle of this module ensures the integrity and reliability of the knowledge graph, enabling the public service platform to effectively address complex urban service needs.

[0052] The intelligent analysis module 104, connected to the knowledge graph construction module, is used to combine user input request data and the constructed public service knowledge graph to generate service demand analysis results; wherein, the intelligent analysis module includes:

[0053] The time series analysis unit is used to perform time series analysis based on historical data and user input request data in the constructed public service knowledge graph, and generate service demand prediction results.

[0054] The spatial analysis unit is used to perform spatial analysis based on geographic location information in the constructed public service knowledge graph and time information at the time of user input, generating spatial demand distribution results; and

[0055] The cross-domain association analysis unit is used to perform cross-domain association analysis based on the potential associations between different public service entities in the constructed public service knowledge graph, and generate cross-domain service combination recommendation strategies.

[0056] Furthermore, the timing analysis unit is specifically used for:

[0057] Based on historical service data in the constructed public service knowledge graph, the temporal trends and cyclical changes in service demand are identified;

[0058] By combining user input request data, including user service type requirements, request time, and geographical location information, the current user request data is matched with historical data in the knowledge graph to identify the temporal distribution characteristics of similar service requests; by analyzing the historical performance of similar requests at different time points, the trend of service demand changes in the current time point and future time periods can be predicted.

[0059] By combining historical service trends and user request patterns, service demand forecasts for a specific future time period are generated; the service demand forecasts include the expected service demand, peak demand periods, and geographical distribution.

[0060] Furthermore, the spatial analysis unit is specifically used for:

[0061] Based on the geographic location information in the constructed public service knowledge graph, the geographic distribution of different public service entities is identified, including the location of public service providers, the coverage of service facilities, and the location of user service requests; the historical geographic data recorded in the knowledge graph is matched with the current user's geographic location information to identify the historical service demand patterns related to the current service request and geographic location.

[0062] By combining the time information input by the user, the distribution of service demand in different geographical areas within a specific time period can be identified; by analyzing the service demand of the same geographical location in similar time periods in historical data, regional demand distribution patterns can be generated to predict the hot spots of service demand in the future within that time period.

[0063] By combining the time information input by the user with the corresponding geographical location data, the spatial demand distribution result for the current time period is generated. The spatial demand distribution result includes the predicted peak service demand areas, demand density, and potential service shortage areas.

[0064] Furthermore, the cross-domain association analysis unit is specifically used for:

[0065] Based on the constructed public service knowledge graph, potential relationships between different public service entities are identified;

[0066] By conducting correlation analysis on service entities across different public service sectors, and combining the co-occurrence of service entities, user preferences, and interdependencies within and outside the sectors, the rationality of the service portfolio is assessed.

[0067] The association strength of service entities is quantified, and a recommendation strategy for cross-domain service combinations is generated based on the user's current request, historical needs, and association relationships.

[0068] The core function of the intelligent analysis module 104 is to use a series of technical means, combining user input request data and public service knowledge graphs, to achieve time-series analysis, spatial analysis, and cross-domain correlation analysis, in order to generate service demand prediction results and cross-domain service recommendation strategies. The following will describe in detail the implementation methods of each analysis unit from a technical perspective to ensure that those skilled in the art can understand and implement these technologies.

[0069] First, the time-series analysis unit is technically implemented based on time-series analysis models and machine learning algorithms. This unit relies on historical data extracted from knowledge graphs, typically in the form of timestamps marking the usage of each service entity at different points in time. To identify the temporal trends and cyclical changes in service demand, the time-series analysis unit can use time-series prediction models such as the Autoregressive Integral Moving Average (ARIMA) model or Long Short-Term Memory (LSTM) network. These models can learn the patterns of service demand changes from historical service records, automatically capturing long-term trends, cyclical fluctuations, and short-term fluctuations. Furthermore, the time-series analysis unit also combines user input data, including service type, request time, and geographic location information, using similarity matching algorithms (such as K-Nearest Neighbors (KNN)) to match the current user request with similar requests in historical service data. After matching, based on the performance of historical data, the system can generate service demand prediction results for a specific time period. These prediction results include demand volume, peak demand periods, etc., and are output to the personalized recommendation module for further service recommendations.

[0070] To improve prediction accuracy, the time series analysis unit also uses cross-validation to validate and adjust the model, ensuring that it can provide effective predictions for different types of services. Furthermore, based on the model's prediction results, the time series analysis unit dynamically adjusts the model according to actual service demand data to address potential nonlinear changes or unexpected events.

[0071] Secondly, the spatial analysis unit uses Geographic Information System (GIS) and spatial statistical techniques, combined with geographic location information extracted from the knowledge graph, to perform spatial demand distribution analysis. The core technology of this unit is spatial data processing and analysis, mainly including geocoding, spatial interpolation, and hotspot analysis. First, the spatial analysis unit needs to match the user's current geographic location information with historical geographic data in the knowledge graph. This process can be achieved through geocoding technology. Geocoding technology converts the user's geographic location (such as address or GPS coordinates) into standardized geographic entities, enabling matching with historical service data in the knowledge graph.

[0072] After matching is complete, the spatial analysis unit uses spatial interpolation algorithms (such as Kriging interpolation or inverse distance weighted interpolation) to predict historical service demand in different regions. These interpolation methods can predict service demand in areas that have not yet been collected based on known historical data. By performing spatial analysis on service requests within the same time period in the past, the system can generate service demand distribution maps for the current time point and a future time period, including areas of high service demand and potential service shortage areas. To identify peak demand areas, the spatial analysis unit can also use hotspot analysis technology to identify locations with abnormally concentrated service demand within a specific area, thereby providing the system with a basis for targeted resource allocation.

[0073] Finally, the cross-domain association analysis unit analyzes the potential associations between different public service entities using entity relationships in the knowledge graph and data mining algorithms, and generates cross-domain service combination recommendation strategies. The core technology of cross-domain association analysis is graph theory algorithms and collaborative filtering. A knowledge graph is essentially a complex relational network where each service entity and user entity exists as a node, and the relationships between nodes are connected by edges. To identify potential associations between different service entities, the cross-domain association analysis unit can use graph embedding algorithms (such as DeepWalk or Node2Vec) to embed service entities into a vector space, thereby quantifying the similarity and association strength between each entity. Through this embedding, the system can analyze the frequency of co-occurrence of service entities in different service domains and the similarity of user preferences, identifying cross-domain service combinations.

[0074] Furthermore, the cross-domain association analysis unit can also utilize collaborative filtering technology to generate service combination recommendation strategies by analyzing users' historical service usage behavior. Specifically, collaborative filtering technology can identify collaborative relationships between different service entities by recognizing common behaviors of multiple users across different service domains. For example, the system can analyze which users simultaneously used transportation and medical services and provide cross-domain service combination recommendations to other users based on these patterns.

[0075] In practical applications, the cross-domain association analysis unit combines the association strength of service entities, the user's historical needs, preferences, and current requests to generate cross-domain service combination recommendation strategies. This strategy is not only based on the user's single needs, but also provides users with more comprehensive, cross-domain service combinations through association analysis to meet their diverse needs.

[0076] The entire intelligent analysis module 104 works collaboratively through time-series analysis, spatial analysis, and cross-domain correlation analysis, ultimately transmitting the analysis results to the personalized recommendation module to generate accurate service recommendations. These technologies ensure that the system can provide users with efficient and personalized public service solutions based on the complex relationships between time, geographical location, and services.

[0077] The following is a reference implementation code for the intelligent analysis module.

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] A personalized recommendation module 105, connected to the intelligent analysis module, is used to provide users with accurate service recommendations based on the results of time-series analysis, spatial analysis, and cross-domain correlation analysis performed by the intelligent analysis module; wherein, the personalized recommendation module includes:

[0086] The demand matching unit is used to filter out the target service that best matches the user's input request from the service demand prediction results generated by the time series analysis unit, and to recommend the target service first.

[0087] The location optimization unit, based on the spatial demand distribution results generated by the spatial analysis unit and combined with the user's geographic location information, prioritizes recommending relevant services in or near the user's location; and

[0088] The service composition recommendation unit is used to recommend multiple related public service combinations to users based on the service composition strategy generated by the cross-domain association analysis unit.

[0089] Furthermore, the demand matching unit is specifically used for:

[0090] Identify the services most relevant to user input requests from the service demand prediction results generated by the time series analysis unit;

[0091] The selected candidate services are prioritized based on the strength of the match, and the target service that best matches the user's request is determined and recommended to the user first.

[0092] Furthermore, the position optimization unit is specifically used for:

[0093] Get the user's current location P u (t), Location of the matched service provider P s (t), and acquire historical performance data H of the service provider through the data acquisition module. s ;

[0094] Calculate the recommendation score for the candidate service provider according to Formula 1 below:

[0095]

[0096] Among them, f opt d(P) represents the recommendation score of the candidate service provider. u (t),P s (t) represents the user's location P. u (t) and the location P of the service provider s The distance between (t); T wait Indicates the service provider's estimated waiting time; c(P) u (t),P s (t) represents the degree of traffic congestion, obtained based on real-time traffic data; w represents the user preference weight; H s This represents the performance score of the service provider; α1, α2, α3 are weight parameters; ∈, γ, δ are smoothing parameters;

[0097] The calculated recommendation scores of candidate service providers are sorted, and the service provider with the highest recommendation score is obtained and displayed to the user.

[0098] Furthermore, the service combination recommendation unit is specifically used for:

[0099] Calculate the optimization score (SCO) of user u for service composition S at time t using Formula 2 as follows:

[0100] SCO=ω1·C(S)+ω2·R(u,S,t)+ω3·E(u,S) (2)

[0101] Where ω1, ω2, ω3 are weight coefficients; C(S) is the internal synergy function; R(u,S,t) is the user relevance function; and E(u,S) is the expected efficiency function.

[0102] The internal synergy function is calculated using the following formula 3:

[0103]

[0104] Where |S| is the number of services in service composition S; sim(s i ,s j ) indicates service s i and s j The semantic similarity is calculated using the Word2Vec model. Word2Vec is a model based on word vector representation. It converts the descriptive text of a service into vectors, and then calculates the cosine similarity between these vectors to obtain the semantic similarity between services. The cosine similarity ranges from [-1, 1], and the closer the value is to 1, the more similar the two services are semantically.

[0105] comp(s i ,s j ) indicates service s i and s j The complementarity score is obtained from historical data using a frequent itemset mining algorithm. This algorithm analyzes which service combinations users frequently use simultaneously; frequent patterns in these combinations indicate complementarity. The complementarity score measures the frequency with which two services co-occur in real-world applications, thus assessing whether they complementarily meet the user's needs.

[0106] The user relevance function is calculated using the following formula 4:

[0107]

[0108] Among them, cos(V u,t V s,t V represents the cosine similarity of the feature vectors of user u and service s at time t. User feature vector V u,t Service feature vector V s,t It is constructed based on user behavior data and historical service attribute data. User feature vectors typically include user preferences and historical usage patterns, while service feature vectors are based on service attributes and descriptive information. Cosine similarity can be used to measure the degree of matching between users and services; a higher cosine similarity value indicates a higher degree of matching between the user's preferences and the service.

[0109] I u,sThis represents the historical number of interactions between user u and service s. This metric reflects the interaction history between the user and the service, smoothing the impact of historical data on current decisions through the logarithm of the interaction count. If a user has frequently used a service in the past, the relevance score of that service will be higher.

[0110] The expected efficiency function is calculated using the following formula 5:

[0111]

[0112] Among them, T s,u This represents the expected benefit that user u will gain from using service s. Expected benefit can be estimated based on the user's historical ratings of the service or the actual utility of the service, reflecting the benefits the user will obtain from using the service.

[0113] D s This represents the expected time cost of service s. Expected time cost refers to the time required for a user to use a service; the system estimates the average duration of a service based on historical data.

[0114] D max This represents the maximum total time cost acceptable to the user. This is typically a user-defined threshold, indicating the maximum time the user is willing to spend on the service mix. If the time cost of services in the mix exceeds this threshold, the system will penalize the mix's score accordingly.

[0115] H(S) represents the entropy of the service composition S.

[0116] To calculate the entropy of the service combination S, the Shannon entropy formula, which is relevant to information theory, can typically be used. The specific calculation process is as follows:

[0117] 1. First, the services in the service portfolio S need to be categorized. Each service can be divided according to its type, function, category, and other characteristics. For example, services can be divided into transportation services, medical services, and education services.

[0118] 2: For each category in the service portfolio S, calculate the proportion of services in that category to the total number of services in the portfolio. This can be expressed as:

[0119]

[0120] Where p i It is the probability of category i.

[0121] 3: For each category in the service portfolio S, calculate the entropy of portfolio S using Shannon's entropy formula:

[0122]

[0123] Where, pi It is the probability of the i-th type of service; log(p) i Typically, the natural logarithm or base-2 logarithm is used; n is the total number of service categories.

[0124] The optimization scores (SCO) of multiple service combinations (S) are compared, and the service combination with the highest score is recommended to the user.

[0125] The core task of the personalized recommendation module 105 is to provide users with the most suitable personalized public service recommendations by integrating the results of time-series analysis, spatial analysis, and cross-domain correlation analysis, combined with the user's specific needs, geographical location, and historical interaction data. This module specifically consists of a demand matching unit, a location optimization unit, and a service combination recommendation unit, with each unit responsible for the service recommendation process across different dimensions.

[0126] The demand matching unit first uses the service demand prediction results generated by the time series analysis unit to filter out the services most relevant to the user's input request. This process relies on the time trend analysis results of historical data and the current user's service request data. During the filtering process, the system determines which services are most likely to meet the user's current needs based on the service type, the user's needs, and the degree of matching between historical service requests. To achieve this matching, the system uses similarity measurement technology to evaluate the degree of matching by calculating the similarity between the user's current request and historical requests (such as the cosine similarity between feature vectors). Based on the strength of the matching, the demand matching unit prioritizes the filtered services and determines the target service that best meets the user's needs, recommending that service first.

[0127] The location optimization unit, based on the spatial demand distribution results generated by the spatial analysis unit and combined with the user's current geographic location information, prioritizes recommending relevant services in or near the user's location. The technical implementation of location optimization is based on Formula 1, which involves the calculation of multiple parameters, such as the user's current location P. u (t) and the location P of the service provider s The distance d(P) between (t) u (t),P s (t)), and the degree of traffic congestion c(P) calculated from real-time traffic data. u (t),P s (t)). In addition, the system also considers the service provider's waiting time T. wait and performance rating H sThese data are dynamically acquired through the data acquisition module 102. Based on Formula 1, the system calculates a recommendation score for each candidate service provider and sorts them by score, prioritizing the recommendation of the service provider with the highest score to the user. This process considers not only geographical location but also service quality, waiting time, and traffic conditions to ensure that users receive a convenient and efficient service experience.

[0128] The service composition recommendation unit further expands the recommendation dimensions by providing users with cross-domain service composition recommendations through service composition strategies generated by the cross-domain association analysis unit. This unit uses Formula 2 to calculate the optimization score SCO of a user for a service composition S at time t, which includes three core indicators: internal synergy, user relevance, and expected efficiency. The internal synergy function C(S) is calculated based on the semantic similarity and complementarity scores between services. Semantic similarity uses the Word2Vec model, calculating cosine similarity through word vector representations of service description texts. The complementarity score is calculated based on the frequent itemset mining algorithm, identifying which service compositions have complementary characteristics from historical data, and calculating it based on the frequent occurrence patterns of the compositions. The user relevance function R(u,S,t) is measured by calculating the cosine similarity between the user's and the service's feature vectors. The feature vectors are constructed from user behavior data and historical service attribute data, with the number of interactions I... u,s This is also included in the calculation to reflect the historical intensity of user interactions with the service. The expected efficiency function E(u,S) evaluates the overall efficiency of the user's use of the service portfolio, expressed as the expected benefit T of the user's use of the service. s,u And expected time cost D s The system performs calculations and considers the complexity of the service composition (represented by the entropy H(S) of the composition). Using these techniques, the system scores multiple service compositions and ultimately selects the highest-scoring composition to recommend to the user.

[0129] The entire personalized recommendation process is multi-dimensional, taking into account user needs, geographical location, service delivery efficiency, and the synergy of service combinations. The personalized recommendation module can not only recommend single services but also provide users with a more comprehensive and integrated service experience through cross-domain service combinations. This recommendation model ensures the system's intelligence and flexibility, enabling dynamic adjustments based on changes in user needs and real-time data, thereby providing users with optimal public services.

[0130] The following is the reference implementation code for the personalized recommendation module 105.

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

Claims

1. A public service platform system based on artificial intelligence, characterized in that, include: The user interaction module is used to receive user input request data, wherein the input request data includes text, voice, images, and time and location information of the user input. The data acquisition module, connected to the user interaction module, is used to collect multi-source heterogeneous urban public service data, including historical service record data, user demographic data, and urban event and policy change data. A knowledge graph construction module, connected to the data acquisition module, is used to construct a public service knowledge graph using the collected urban public service data. The public service knowledge graph includes public service entities and their attributes, user entities and their attributes, time and geographical location information of public service entities and user entities, and the relationship and interaction history between public service entities and user entities. An intelligent analysis module, connected to the knowledge graph construction module, is used to generate service demand analysis results by combining user input request data and the constructed public service knowledge graph; wherein, the intelligent analysis module includes: The time-series analysis unit is used to match the user's current request data with the historical data in the public service knowledge graph based on the historical data in the constructed public service knowledge graph, and to identify the time distribution characteristics of similar service requests. The user's input request data includes the user's service type requirements, request time, and geographical location information. By analyzing the historical performance of similar service requests at different time points, the unit predicts the trend of service demand changes in the current time point and in future time periods. The spatial analysis unit is used to identify the geographical distribution of different public service entities based on the geographical location information in the constructed public service knowledge graph and the time information of user input, including the location of public service providers, the coverage of service facilities, and the location of user service requests; match the historical geographical data recorded in the public service knowledge graph with the current user's geographical location information to identify historical service demand patterns related to the current service request and geographical location; and generate spatial demand distribution results, wherein the spatial demand distribution results include predicted service demand peak areas, demand density, and potential service shortage areas; and the cross-domain association analysis unit is used to perform cross-domain association analysis based on the potential associations between different public service entities in the constructed public service knowledge graph to generate cross-domain service combination recommendation strategies. A personalized recommendation module, connected to the intelligent analysis module, is used to provide users with accurate service recommendations based on the results of time-series analysis, spatial analysis, and cross-domain correlation analysis performed by the intelligent analysis module; wherein, the personalized recommendation module includes: The demand matching unit is used to filter out the target service that best matches the user's input request from the service demand prediction results generated by the time series analysis unit, and to recommend the target service first. The location optimization unit is used to prioritize recommending relevant services in or near the user's location based on the spatial demand distribution results generated by the spatial analysis unit and the user's geographic location information; and the service combination recommendation unit is used to recommend multiple related public service combinations to the user based on the service combination strategy generated by the cross-domain association analysis unit.

2. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The user interaction module is specifically used for: It receives various forms of input request data from users, including text input, voice input, and image input; and uses natural language processing technology to recognize and parse text and voice, and uses image recognition technology to parse images or scanned documents uploaded by users to generate structured request information. While the user is inputting a request, the system captures and records the user's time and location information, and transmits it along with the input request data to the data processing module.

3. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The data acquisition module is specifically used for: Data related to urban public services is collected from multiple data sources, including historical service record databases, user demographic databases, and systems for government policies and urban events. Identify and process heterogeneous data from different data sources, including structured data, semi-structured data, and unstructured data; Through data transformation and standardization steps, the collected data is converted into a unified format to ensure that all data can be effectively used by the system.

4. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The knowledge graph construction module is specifically used for: Based on urban public service data collected from the data acquisition module, public service entities and user entities are identified; The identified public service entities and user entities are assigned corresponding attributes. The attributes of public service entities include service type, service provider, service description, timeliness, and availability. The attributes of user entities include demographic information, service preferences, and interaction history. Based on historical interaction data between user entities and public service entities, relationships between entities are constructed; wherein, the relationships include user service requests, user service participation, user feedback, service delivery timeliness, and service availability.

5. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The cross-domain association analysis unit is specifically used for: Based on the constructed public service knowledge graph, potential relationships between different public service entities are identified; By conducting correlation analysis on service entities across different public service sectors, and combining the co-occurrence of service entities, user preferences, and interdependencies within and outside the sectors, the rationality of the service portfolio is assessed. The association strength of service entities is quantified, and a recommendation strategy for cross-domain service combinations is generated based on the user's current request, historical needs, and association relationships.

6. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The demand matching unit is specifically used for: Identify the services most relevant to user input requests from the service demand prediction results generated by the time series analysis unit; The selected candidate services are prioritized based on the strength of the match, and the target service that best matches the user's request is determined and recommended to the user first.

7. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The position optimization unit is specifically used for: Get the user's current location Location of matching service providers It also obtains historical performance data from service providers through the data acquisition module. ; Calculate the recommendation score for the candidate service provider according to Formula 1 below: ； in, This indicates the recommendation score for the candidate service provider; Indicates user location Location of service provider The distance between them; Indicates the estimated waiting time from the service provider; Indicates the degree of traffic congestion, obtained based on real-time traffic data; Indicates user preference weights; This indicates the service provider's performance rating, which can be obtained through user satisfaction feedback; These are weight parameters; It is a smoothing parameter; The calculated recommendation scores of candidate service providers are sorted, and the service provider with the highest recommendation score is obtained and displayed to the user.

8. The public service platform system based on artificial intelligence according to claim 1, characterized in that, The service combination recommendation unit is specifically used for: Calculate the user's information according to Formula 2 below. In time For service portfolio Optimized score : ； in, These are weighting coefficients; It is an internal coordinating function; It is a user relevance function; It is the expected efficiency function; The internal synergy function is calculated using the following formula 3: ； in, Indicates service and The semantic similarity is calculated using the Word2Vec model based on the word vector representation of the service description text. The cosine similarity between services is then used as the service similarity score. and semantic similarity; Indicates service and The complementarity score is obtained by identifying complementary relationships from historical usage data through a frequent itemset mining algorithm and calculating based on the frequent occurrence patterns of service composition. The user relevance function is calculated using the following formula 4: ； in, Indicates user Services In time The cosine similarity of the feature vectors is calculated based on the feature vectors of users and services. The feature vectors are constructed from historical data of user behavior and service attributes. Indicates user Services The number of historical interactions; The expected efficiency function is calculated using the following formula 5: ； in, Indicates user Use service The expected returns obtained; Indicates service The expected time cost; The maximum total time cost acceptable to the user; Indicates service composition Entropy; For multiple service combinations Optimized score The system compares the services and selects the highest-scoring combination to recommend to the user.

Images