Urban area rating method, system and device and storage medium

Abstract

The invention relates to the technical field of data analysis and real estate assessment, in particular to an urban area rating method, system and device and a storage medium, and the method comprises the steps: obtaining urban multi-dimensional data (including urban population, industry, supporting facilities and real estate market supply and demand relationship data), carrying out comprehensive analysis processing (including dynamic weight adjustment and deep neural network association analysis integration), calculating an urban real estate market support degree index, and calculating a regional real estate market health index after dividing the city into regions. According to the invention, urban multi-dimensional data are comprehensively considered, and accurate rating and real-time tracking of the real estate market are realized through dynamic weight adjustment and the deep neural network.

Description

Technical Field

[0001] The present application relates to the technical field of data analysis and real estate evaluation, and in particular to a method, system, device and storage medium for rating urban areas. Background Art

[0002] In the current economic environment, the real estate market plays a vital role in the development of cities and regions. Accurately evaluating the real estate market conditions in cities and regions is of great significance to corporate investment and personal home purchase decisions. However, traditional real estate market evaluation methods often rely on a single indicator or limited data dimensions, which makes it difficult to fully and accurately reflect the true situation of the market. For example, some evaluation methods only consider the supply and demand relationship in the real estate market, while ignoring the impact of important factors such as urban population, industrial development, and supporting facilities on the real estate market. In addition, existing evaluation methods usually adopt a static analysis model, which is unable to track market changes in real time and adjust the evaluation results in a timely manner.

[0003] Therefore, based on the above problems, the prior art still needs to be improved. Summary of the invention

[0004] The purpose of this application is to provide a city area rating method, system, device and storage medium, aiming to solve the problems that traditional real estate market evaluation methods are single and static, cannot fully consider multi-dimensional urban data, and are difficult to track market changes in real time.

[0005] The first purpose of this application is to provide a city area rating method, including: Acquire multi-dimensional data of the city, wherein the multi-dimensional data of the city includes city population data, city industry data, city supporting facilities data and city real estate market supply and demand relationship data; Comprehensive analysis and processing of the multi-dimensional data of the city is performed, including: quantification and standardization of dynamic weight adjustment, real-time adjustment of the weight coefficient of the multi-dimensional data of the city based on market trends and data sensitivity; association analysis and integration using a deep neural network model to learn the complex nonlinear relationship between the multi-dimensional data of the city; By comprehensively analyzing the multi-dimensional data of the city and the first index model, the city real estate market support index is calculated, and the weight coefficient of the first index model is dynamically adjusted; The city is divided into several regions, and regional multi-dimensional data is obtained. The regional real estate market health index is calculated through comprehensive analysis of the regional multi-dimensional data and the second index model, and the weight coefficient of the second index model is dynamically adjusted.

[0006] By adopting the above technical solutions, we can comprehensively obtain multi-dimensional data of cities, including urban population, industry, supporting facilities and real estate market supply and demand data, provide a rich and comprehensive information basis for rating, and avoid the one-sidedness of a single data source. Dynamic weight adjustment combines market trends and data sensitivity to adapt to market changes in real time, making the weights of data in each dimension more scientific and reasonable, and accurately reflecting the actual impact of various factors on the support of urban real estate markets and regional real estate market health indexes in different periods. The deep neural network model effectively learns the complex nonlinear relationship between multi-dimensional data in cities, improves the accuracy and depth of analysis, and provides strong technical support for accurate calculation of indexes. It realizes accurate rating and real-time tracking of the real estate market, ensuring that the rating results are always consistent with the actual market conditions, and providing a reliable basis for corporate investment and personal decision-making. At the same time, dividing cities into regions for separate evaluation can not only grasp the overall real estate market conditions of the city, but also deeply understand the differences in characteristics of each region, and provide more detailed and targeted references.

[0007] In a possible implementation of the present application, the steps of using a deep neural network model for association analysis and integration include: Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the city, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the city real estate market support index; The neural network is trained by using a back propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted city real estate market support index and the actual index until a preset convergence standard is reached; When a city is divided into several regions, the city multi-dimensional data is divided to generate the regional multi-dimensional data; Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the region, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the real estate market health index of the region; The neural network is trained using a back-propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted real estate market health index of the region and the actual index until a preset convergence standard is reached.

[0008] By adopting the above technical solutions, a specific neural network structure is constructed. The input layer corresponds to the multi-dimensional data of the city or the multi-dimensional data of the region, which can make full use of the rich data sources for analysis, making the analysis results more comprehensive and accurate. The number of nodes in the hidden layer is adjusted according to the complexity of the data and the training effect to ensure that the model can adapt to data sets in different situations and improve the flexibility and adaptability of the model. The output layer corresponds to the urban real estate market support index and the regional real estate market health index, respectively, which provides clear indicators for accurately evaluating the real estate market conditions in cities and regions. The back propagation algorithm is used for training, and the weights and thresholds are adjusted by continuously minimizing the error between the predicted index and the actual index, so that the model can gradually approach the actual situation and improve the accuracy of the prediction. Training for cities and regions separately can better capture the characteristics of the real estate market as a whole and in each region, and provide strong support for accurate rating. It is helpful to achieve a comprehensive and in-depth analysis of urban and regional real estate markets, provide a scientific basis for corporate investment and personal home purchases, and promote the stable development and reasonable planning of the real estate market.

[0009] In a possible implementation manner of the present application, the quantification and standardization processing of dynamic weight adjustment, and the step of adjusting the weight coefficient of the multi-dimensional data of the city in real time based on market trends and data sensitivity, include: The specific values ​​of market trends and data sensitivity are converted into fuzzy language variables, including: for market trends, multiple key indicators are determined for evaluation; each key indicator is divided into several intervals, and the several intervals are corresponded to fuzzy language variables; for data sensitivity, the influence of the multi-dimensional data of the city on the support index of the city real estate market or the health index of the regional real estate market is analyzed; according to the degree of sensitivity, it is divided into high sensitivity, medium sensitivity and low sensitivity and corresponds to fuzzy language variables; Establish fuzzy inference rules to determine the weight adjustment strategy of the multi-dimensional data of the city under different market trends and data sensitivity combinations, and optimize and adjust the fuzzy inference rules based on historical data simulation and machine learning methods; According to the fuzzified market trends and data sensitivity, the weight coefficients of the multi-dimensional data of the city are dynamically adjusted.

[0010] By adopting the above technical solution, the specific values ​​of market trends and data sensitivity are converted into fuzzy language variables. By determining multiple key indicators to evaluate market trends and divide the intervals corresponding to fuzzy language variables, and at the same time dividing and corresponding fuzzy language variables according to the sensitivity, the evaluation of complex market trends and data sensitivity is more intuitive and easy to understand, and the importance of market dynamics and data of each dimension can be grasped comprehensively and meticulously. On the other hand, fuzzy reasoning rules are established and optimized and adjusted based on historical data simulation and machine learning methods, ensuring that the weight adjustment strategy of urban multi-dimensional data is more scientific and reasonable under different market trends and data sensitivity combinations, improving the accuracy of weight adjustment, enhancing the adaptability and flexibility of the method, and being able to better cope with various complex market situations. Finally, the weight coefficients of urban multi-dimensional data are dynamically adjusted according to the fuzzified market trends and data sensitivity, achieving the purpose of real-time adaptation to market changes, making the calculation of urban real estate market support index and regional real estate market health index more accurately reflect the actual market situation, providing a more reliable basis for the decision-making of enterprises and individuals, and helping to promote the stable development and scientific planning of the real estate market.

[0011] In a possible implementation manner of the present application, the method further includes: Rating the urban real estate market based on the urban real estate market support index, and rating the regional real estate market based on the regional real estate market health index; Obtain users' decision-making information, historical data, and trend analysis data; Combining the urban real estate market support index, regional real estate market health index, historical data and trend analysis data, a deep learning model is used to generate forecast suggestions, which include market trend judgments, potential risk warnings and corresponding response strategies.

[0012] By adopting the above technical solutions, the urban and regional real estate markets are rated based on the urban real estate market support index and the regional real estate market health index, providing clear quantitative standards for market evaluation and making the market situation clear at a glance. Obtain user decision information, historical data and trend analysis data, and combine these data and the two indexes to generate forecast suggestions using a deep learning model, including market trend judgment, potential risk warnings and response strategies, providing users with a comprehensive and forward-looking decision-making reference. This not only helps users make more informed decisions, such as reasonable planning by the government, precise investment by enterprises, and rational purchase of houses by individuals, but also can warn of potential risks in advance, so that all parties can take timely response measures, thereby promoting the stable and healthy development of the real estate market and improving the market's risk resistance and sustainable development level.

[0013] In a possible implementation of the present application, the steps of using a deep learning model to generate prediction suggestions in combination with the city real estate market support index, the regional real estate market health index, historical data and trend analysis data include: Cleaning and standardizing the urban real estate market support index, the regional real estate market health index, and the historical data to remove outliers and noise so that the data have a unified format and scale. For the trend analysis data, use a time series analysis method to extract the main trend features; Build a multi-layer deep learning model, including an input layer, multiple hidden layers, and an output layer. The input layer receives preprocessed data, the hidden layer learns complex patterns and relationships in the data through different neuron combinations and activation functions, and the output layer generates forecast recommendations, including market trend judgments, potential risk warnings, and corresponding response strategies; Using a large amount of the historical data to train the deep learning model, using optimization algorithms such as stochastic gradient descent to adjust the weights and parameters of the model by minimizing the error between the predicted suggestions and the actual results. During the training process, using cross-validation and other techniques to prevent overfitting and improve the generalization ability of the model; Input the current urban real estate market support index, the regional real estate market health index and the trend analysis data into the trained deep learning model to obtain market trend judgment, potential risk warnings and corresponding response strategies; Evaluate the generated forecast suggestions and analyze the accuracy and reliability of the forecasts by comparing them with actual market information; Adjust and optimize the deep learning model based on the evaluation results; Obtain user feedback on the prediction suggestions to further adjust and optimize the deep learning model.

[0014] By adopting the above technical solutions, the urban real estate market support index, regional real estate market health index, and historical data are cleaned and standardized to remove outliers and noise, ensure the quality and uniformity of the data, and provide a reliable basis for subsequent analysis. The time series analysis method is used to extract the main trend characteristics of trend analysis data, which can better grasp the market dynamics. A multi-layer deep learning model is constructed, and through different levels of neuron combinations and activation functions, the complex patterns and relationships in the data are effectively learned, and accurate forecasting suggestions are output to provide users with comprehensive market trend judgments, potential risk warnings, and response strategies. A large amount of historical data is used for training, and optimization algorithms and cross-validation techniques are used to improve the accuracy and generalization ability of the model, so that it can adapt to different market conditions. After the current data is input into the trained model to obtain the prediction results, it is evaluated by comparing with the actual market situation. According to the evaluation results and user feedback, the model is continuously adjusted and optimized, which further improves the accuracy and reliability of the prediction. It provides a scientific decision-making basis for participants in the real estate market, helps to reduce risks, improve decision-making efficiency, and promote the stability and sustainable development of the real estate market.

[0015] In a possible implementation manner of the present application, the method further includes: Visually present the city and its divided areas in the form of a map; Regions with different ratings are distinguished by different colors, and the city real estate market support index and the regional real estate market health index are displayed in the corresponding area in the form of a floating window; When the user selects a specific area, the real estate market health index of the area and the proportion of multi-dimensional data of the area are provided; Get the time node information selected by the user; Based on the time node information, the changing trends of the urban and regional real estate markets are displayed, and the changes of the multi-dimensional data of the city and the multi-dimensional data of the region over time are presented in the form of dynamic charts.

[0016] By adopting the above technical solution, the city and its divided regions are intuitively presented in the form of a map, so that users can quickly and clearly understand the overall layout and regional division of the city. Regions with different ratings are distinguished by different colors, and the urban real estate market support index and regional real estate market health index are displayed in the form of floating windows, which enhances the visualization of information and facilitates users to intuitively understand the real estate market conditions in various regions. When the user selects a specific area, the regional real estate market health index and the proportion of regional multi-dimensional data are provided, providing users with more detailed regional information and helping users to deeply understand the characteristics of specific areas. The time node information selected by the user is obtained, and based on this, the changing trends of the urban and regional real estate markets are displayed, and the changes in multi-dimensional data over time are presented in the form of dynamic charts, so that users can intuitively observe the dynamic changes in the market, providing a strong basis for users to make decisions, and improving the scientificity and accuracy of decision-making.

[0017] In a possible implementation manner of the present application, the subsequent steps include: Obtain the time node information and specific area information specified by the user; Based on the urban real estate market support index, regional real estate market health index and regional multi-dimensional data corresponding to the time node information, the market conditions at the corresponding time point are analyzed, the area corresponding to the specific regional information is rated, and a detailed real estate market rating report is generated.

[0018] By adopting the above technical solution, the user's personalized needs can be accurately met by obtaining the time node information and specific area information specified by the user's needs. Market situation analysis is performed based on the urban real estate market support index, regional real estate market health index and regional multi-dimensional data corresponding to the time node information, ensuring an in-depth understanding of a specific area at a specific time point. Rating specific areas and generating detailed real estate market rating reports provide users with a comprehensive and professional decision-making reference. This not only helps users accurately grasp the real estate market dynamics in a specific area at a specific time, but also provides strong support for decisions such as corporate investment and personal home purchases, improves the scientificity and accuracy of decision-making, and promotes the stable and healthy development of the real estate market.

[0019] The second purpose of this application is to provide an urban area rating system, which includes: City multi-dimensional data acquisition module: acquires city multi-dimensional data, including city population data, city industry data, city supporting facilities data and city real estate market supply and demand relationship data; Comprehensive analysis and processing module: Comprehensive analysis and processing of the multi-dimensional data of the city, including: quantification and standardization of dynamic weight adjustment, real-time adjustment of the weight coefficient of the multi-dimensional data of the city based on market trends and data sensitivity; using a deep neural network model for correlation analysis and integration, and learning the complex nonlinear relationship between the multi-dimensional data of the city; Urban real estate market support index calculation module: calculates the urban real estate market support index through comprehensive analysis of the multi-dimensional data of the city and the first index model, and dynamically adjusts the weight coefficient of the first index model; Regional real estate market health index calculation module: divide the city into several regions, obtain regional multi-dimensional data, calculate the regional real estate market health index through comprehensive analysis of the regional multi-dimensional data and the second index model, and dynamically adjust the weight coefficient of the second index model.

[0020] By adopting the above technical solutions, we can comprehensively obtain multi-dimensional data of cities, including urban population, industry, supporting facilities and real estate market supply and demand data, provide a rich and comprehensive information basis for rating, and avoid the one-sidedness of a single data source. Dynamic weight adjustment combines market trends and data sensitivity to adapt to market changes in real time, making the weights of data in each dimension more scientific and reasonable, and accurately reflecting the actual impact of various factors on the support of urban real estate markets and regional real estate market health indexes in different periods. The deep neural network model effectively learns the complex nonlinear relationship between multi-dimensional data in cities, improves the accuracy and depth of analysis, and provides strong technical support for accurate calculation of indexes. It realizes accurate rating and real-time tracking of the real estate market, ensuring that the rating results are always consistent with the actual market conditions, and providing a reliable basis for corporate investment and personal decision-making. At the same time, dividing cities into regions for separate evaluation can not only grasp the overall real estate market conditions of the city, but also deeply understand the differences in characteristics of each region, and provide more detailed and targeted references.

[0021] The third object of this application is to provide a city area rating device, which includes: A memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned urban area rating method.

[0022] The fourth objective of this application is to provide a storage medium.

[0023] The fourth application objective of this application is achieved through the following technical solutions: A storage medium stores a computer program that can be loaded by a processor and execute the above-mentioned urban area rating method.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: 1. Comprehensively acquire multi-dimensional urban data to provide a rich and accurate information basis for urban area ratings, avoid the one-sidedness of a single data source, and at the same time achieve accurate rating and real-time tracking of the real estate market through dynamic weight adjustment and deep neural network models.

[0025] 2. The use of deep neural networks for correlation analysis and integration can effectively learn the complex nonlinear relationships among multi-dimensional urban data, improve the accuracy and depth of analysis, and provide strong technical support for the accurate calculation of the urban real estate market support index and the regional real estate market health index.

[0026] 3. By converting market trends and data sensitivity into fuzzy language variables and establishing fuzzy reasoning rules, scientific and reasonable dynamic weight adjustments can be achieved to better adapt to market changes.

[0027] 4. Combine multiple data and use deep learning models to generate forecast suggestions, including market trend judgment, potential risk warnings and response strategies, to provide users with comprehensive decision-making references and improve the scientificity and reliability of decision-making.

[0028] 5. Present city and regional information in an intuitive map format to enhance information visualization, making it easier for users to understand the real estate market conditions in different regions. At the same time, generate detailed real estate market rating reports based on user needs to meet personalized needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a flow chart of a city area rating method provided in an embodiment of the present application; Figure 2 It is a virtual structural diagram of a city area rating system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0031] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0032] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0033] This application embodiment provides a city area rating method, referring to Figure 1 , the main process of the method is described as follows: S1: Acquire multi-dimensional data of a city, wherein the multi-dimensional data of a city includes city population data, city industry data, city supporting facilities data, and city real estate market supply and demand relationship data; Among them, by connecting with the database of the local statistical department, the city's permanent population, registered population, population growth rate and other urban population data are regularly obtained. At the same time, the data interface provided by the mobile operator is used to analyze the flow of the population, for example, through the positioning information of the mobile phone signal, the distribution and flow trend of the population in different regions are determined. In terms of collecting urban industrial data, GDP, output value of various industries, industrial structure ratio and other information are extracted from the city's economic statistical database, and the local industrial and commercial departments and industry associations are interacted with to obtain the registration information, operating conditions, industry distribution and other data of the enterprise. In addition, big data analysis technology is used to mine news, industry reports, corporate official websites and other information on the Internet to obtain the latest dynamics and trends about the city's industry. For urban supporting facilities data, the city's infrastructure planning and construction, including the distribution and construction progress of transportation facilities such as roads, bridges, subways and buses, are obtained from the database of the urban planning department. The computer is also connected to the information system of the education department and the school to collect the city's educational supporting facilities such as the number, distribution, and teaching staff of the school, and cooperate with the medical department to obtain the number, grade, and number of beds of the hospital and other medical supporting facilities data. In addition, the API of map software and location service platform is used to obtain the distribution and scale of commercial supporting facilities such as commercial centers, shopping centers, supermarkets, etc. in the city. When obtaining the supply and demand relationship data of the urban real estate market, the information of real estate development projects in the city is obtained from the database of the real estate management department, including the location, scale, construction progress, sales, etc. of the project. At the same time, the online platform data interface of the real estate agency is used to understand the supply and demand situation of the second-hand housing market, housing price trends and other information. Data is also collected from the real estate information platform on the Internet to obtain the listing information, transaction price and other data of new and second-hand houses. In addition, by conducting market research, designing a questionnaire survey system and conducting interview record analysis, the demand and willingness of consumers to buy houses, as well as the supply plan and market expectations of real estate developers, are understood.

[0034] S2: Comprehensively analyzing and processing the multi-dimensional data of the city, including: quantifying and standardizing the dynamic weight adjustment, adjusting the weight coefficient of the multi-dimensional data of the city in real time based on market trends and data sensitivity; using a deep neural network model to perform correlation analysis and integration, and learning the complex nonlinear relationship between the multi-dimensional data of the city; Among them, the specific values ​​of market trends and data sensitivity are first converted into fuzzy language variables. For market trends, multiple key indicators are determined for evaluation, such as economic growth rate, population inflow rate, industrial development speed, etc. Each key indicator is divided into several intervals, for example, the economic growth rate is divided into high-speed growth interval, medium-speed growth interval, low-speed growth interval, etc., and these intervals are corresponded to fuzzy language variables, such as "high growth trend", "medium growth trend" and "low growth trend". For data sensitivity, the impact of multi-dimensional urban data on the urban real estate market support index or the regional real estate market health index is analyzed. Through a large amount of historical data simulation and analysis, the weight range of the impact of different data on the index is determined. Then, according to the degree of sensitivity, it is divided into high sensitivity, medium sensitivity and low sensitivity and corresponds to fuzzy language variables, such as "high sensitive data", "medium sensitive data" and "low sensitive data".

[0035] Next, fuzzy inference rules are established to determine the weight adjustment strategy for urban multidimensional data under different combinations of market trends and data sensitivity. For example, when the market is in a high growth trend and a certain data is highly sensitive data, a higher weight is given to the data. Fuzzy inference rules are optimized and adjusted based on historical data simulation and machine learning methods. By continuously inputting new market data and analysis results, the computer automatically adjusts the parameters and logic in the fuzzy inference rules to improve the accuracy and adaptability of the rules. Finally, according to the fuzzified market trends and data sensitivity, the computer dynamically adjusts the weight coefficients of urban multidimensional data.

[0036] Construct a deep neural network structure consisting of an input layer, a hidden layer, and an output layer. The input layer nodes correspond to multi-dimensional urban data, including urban population data, urban industry data, urban supporting facilities data, and urban real estate market supply and demand data. The number of hidden layer nodes is adjusted according to data complexity and training effect. Through a large number of experiments and optimizations, the appropriate number of hidden layer nodes is determined to ensure that the model can fully learn the complex relationships in the data. The output layer nodes correspond to the urban real estate market support index.

[0037] The neural network is trained using the back propagation algorithm. A large amount of known data is input into the neural network to calculate the error between the predicted urban real estate market support index and the actual index. Then, according to the principle of error back propagation, the weights and thresholds of the neural network are continuously adjusted to gradually reduce the error. The neural network is continuously trained until the preset convergence standard is reached, such as the error is less than a certain threshold or after a certain number of training times.

[0038] When a city is divided into several regions, the city multi-dimensional data is divided according to the regions to generate regional multi-dimensional data. Then, a new neural network structure is constructed, in which the input layer nodes correspond to the regional multi-dimensional data. The settings of the hidden layer and the output layer are similar to those of the neural network at the city level, and the output layer nodes correspond to the regional real estate market health index. Similarly, the back propagation algorithm is used to train the regional neural network. By minimizing the error between the predicted regional real estate market health index and the actual index, the weights and thresholds of the neural network are continuously adjusted until the preset convergence standard is reached.

[0039] S3: Calculate the city real estate market support index by comprehensively analyzing the multi-dimensional data of the city and the first index model, and dynamically adjust the weight coefficient of the first index model; Among them, the first index model is constructed first, which can be a multi-factor weighted comprehensive model. For example, the urban real estate market support index is determined by multiple factors such as urban population factors, urban industry factors, urban supporting facilities factors, and urban real estate market supply and demand factors, and each factor has a corresponding weight coefficient.

[0040] Extract population-related data from the multi-dimensional data of the city after comprehensive analysis, such as population size, population growth rate, population structure, etc. Use specific algorithms to convert these data into quantifiable indicators. For example, different scores can be given according to the population growth rate, and a high proportion of young people in the population structure can be given a higher score. Finally, a numerical value representing the city's population factors is obtained.

[0041] Extract industry-related data, including industry output value, industry development speed, proportion of emerging industries, etc. Also, convert these data into indicators through quantitative algorithms. For example, industries with high output value, fast development speed, and large proportion of emerging industries are given higher scores. Obtain the value of urban industrial factors.

[0042] Extract supporting facilities data such as transportation facilities, educational facilities, medical facilities, and commercial facilities. Give different scores based on the degree of perfection and coverage of the facilities. For example, areas with convenient transportation, high-quality schools, advanced hospitals, and developed commerce will be given higher scores. Determine the value of the city's supporting facilities factors.

[0043] Analyze the supply data of the real estate market, such as the number of newly built houses and land supply, and demand data, such as the demand for buying houses and renting houses. Different scores are given according to the tension between supply and demand. High scores are given when supply is tight and demand is large, and low scores are given when supply is excessive and demand is small. Obtain the value of the supply and demand relationship factors in the urban real estate market.

[0044] Monitor changes in market trends and data sensitivity in real time. According to the dynamic weight adjustment method described above, when market trends change, such as when the economic situation improves and population inflow increases, the weight coefficients of each factor are adjusted accordingly. If the economic situation improves, the importance of urban industrial factors may increase, so the weight coefficient of industrial factors will be increased. If population inflow increases, the weight coefficient of urban population factors may increase. At the same time, adjustments are made according to changes in data sensitivity. If the data sensitivity of a factor increases, it means that the factor has an increased impact on the urban real estate market support index, and the weight coefficient of the factor will be increased. For example, when it is found that the improvement of urban supporting facilities has an increased impact on the real estate market, the weight coefficient of the urban supporting facilities factor is increased. The weight coefficient is constantly adjusted according to new data and changes to ensure that the calculation of the urban real estate market support index can accurately reflect the current market conditions. Through the above steps, the urban real estate market support index is calculated using the multi-dimensional urban data after comprehensive analysis and processing and the dynamically adjusted first index model.

[0045] S4: Divide the city into several regions, obtain regional multi-dimensional data, calculate the regional real estate market health index through comprehensive analysis of the regional multi-dimensional data and the second index model, and dynamically adjust the weight coefficient of the second index model.

[0046] There are many ways to divide a city into several areas. For example, the city can be divided into different districts and counties based on administrative divisions; it can also be divided based on geographical features, such as natural boundaries such as rivers and mountains; it can also be divided based on urban functional zoning, such as commercial areas, residential areas, industrial areas, etc. After determining the division method, the city map is gridded, and each grid represents a region for subsequent data processing and analysis.

[0047] Using demographic data and geographic information system (GIS) technology, determine the number of permanent residents, population density, population structure and other information in each region. By cooperating with mobile operators, mobile phone signal positioning data can be obtained to further analyze the population flow in the region. Combined with business registration information and economic census data, determine the number of enterprises, industry types, output value and other information in each region. Analyze the industrial concentration and industrial development trends in the region, such as whether there is a concentration of emerging industries, transformation and upgrading of traditional industries, etc. Using urban planning data and GIS technology, determine the distribution and perfection of supporting facilities such as transportation facilities, educational facilities, medical facilities, and commercial facilities in each region. For example, calculate indicators such as the density of bus routes in the region, the number and level of schools and hospitals, and the scale of commercial centers. Collect real estate transaction data and land transfer data to determine the number of newly built houses, the number of second-hand housing transactions, and housing price trends in each region. Analyze the supply and demand relationship of the real estate market in the region, such as the characteristics of regions with oversupply or strong demand.

[0048] The second index model is constructed to calculate the regional real estate market health index. Similar to the first index model, the second index model can be a multi-factor weighted comprehensive model. Let the regional real estate market health index be, which is jointly determined by multiple factors such as regional population factors, regional industrial factors, regional supporting facilities factors, and regional real estate market supply and demand factors. The basic form of the model can be expressed as, where,,, are the weight coefficients of each regional factor respectively.

[0049] Extract population-related data from regional multi-dimensional data, such as the number of regional permanent residents, population growth rate, population structure, etc. Use specific algorithms to convert these data into quantifiable indicators, such as high population growth rates and high proportions of young people. Finally, a value representing regional population factors is obtained. Extract industry-related data in the region, including the number of enterprises, industrial output value, industrial development speed, etc. Also use quantitative algorithms to convert these data into indicators, such as high scores for regions with a large number of enterprises, high industrial output value, and fast development speed. Obtain the value of regional industrial factors. Extract supporting facilities data such as transportation facilities, educational facilities, medical facilities, and commercial facilities in the region. Different scores are given according to the degree of perfection and coverage of the facilities. For example, regions with convenient transportation, high-quality schools, advanced hospitals, and developed commerce are given higher scores. Determine the value of regional supporting facilities factors. Analyze the supply data of the real estate market in the region, such as the number of newly built houses and land supply; demand data, such as demand for buying houses and demand for renting houses. Different scores are given according to the degree of tension between supply and demand. Regions with tight supply and large demand are given higher scores, and regions with oversupply and small demand are given lower scores. Obtain the values ​​of regional real estate market supply and demand factors.

[0050] Real-time monitoring of changes in regional market trends and data sensitivity. Similar to the first index model, when regional market trends change, such as the rapid development of emerging industries and a large influx of population in a certain region, the weight coefficients of various regional factors are adjusted accordingly. If a region has a rapid development of emerging industries, the importance of regional industrial factors may increase, so the weight coefficient of regional industrial factors will be increased. If a large influx of population occurs in a certain region, the weight coefficient of regional population factors may increase. At the same time, adjustments are made according to changes in data sensitivity. If the data sensitivity of a regional factor increases, it means that the factor has a greater impact on the regional real estate market health index, and the weight coefficient of the regional factor will be increased. For example, when it is found that the improvement of supporting facilities in a certain region has a greater impact on the real estate market, the weight coefficient of the supporting facilities factor in the region is increased. The weight coefficient is constantly adjusted according to new data and changes to ensure that the calculation of the regional real estate market health index can accurately reflect the current market conditions in each region. Through the above steps, the real estate market health index of each region is calculated using the regional multi-dimensional data after comprehensive analysis and processing and the dynamically adjusted second index model.

[0051] Specifically, in some possible embodiments, the steps of using a deep neural network model to perform association analysis and integration include: Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the city, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the city real estate market support index; The neural network is trained by using a back propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted city real estate market support index and the actual index until a preset convergence standard is reached; When a city is divided into several regions, the city multi-dimensional data is divided to generate the regional multi-dimensional data; Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the region, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the real estate market health index of the region; The neural network is trained using a back-propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted real estate market health index of the region and the actual index until a preset convergence standard is reached.

[0052] Among them, first determine the input layer nodes, which correspond to the city's multi-dimensional data, including urban population data, urban industry data, urban supporting facilities data, and urban real estate market supply and demand data. For example, urban population data can include specific indicators such as population size, population growth rate, and population structure, and each indicator is used as an input node. Then determine the hidden layer. The determination of the number of hidden layer nodes needs to consider the complexity of the data and the training effect. You can first set an initial number of hidden layer nodes, and then find the most suitable number through multiple trials and adjustments. During the training process, if the model is found to be overfitting or underfitting, you can appropriately increase or decrease the number of hidden layer nodes. For example, if the data complexity is high, more hidden layer nodes may be needed to capture the complex relationship in the data; if the training effect is not good, you can try to increase the number of hidden layer nodes to improve the expression ability of the model. The output layer node corresponds to the urban real estate market support index. This index is a single value that reflects the overall situation of the urban real estate market.

[0053] When the city is divided into several regions, the city multi-dimensional data is divided according to the region to generate regional multi-dimensional data. The input layer nodes correspond to these regional multi-dimensional data, which also include regional population data, regional industrial data, regional supporting facilities data, and regional real estate market supply and demand relationship data. The method for determining the number of hidden layer nodes is similar to that of the neural network at the city level, and is adjusted according to the complexity of the regional data and the training effect. The data complexity of different regions may be different, so separate analysis and adjustment are required for each region. The output layer nodes correspond to the regional real estate market health index, which reflects the health of the real estate market in each region.

[0054] A large amount of known multi-dimensional data of cities is input into the constructed neural network, and the predicted urban real estate market support index is obtained through forward propagation calculation. Then, the predicted index is compared with the actual urban real estate market support index, and the error between the two is calculated. The back propagation algorithm is used. According to the principle of error back propagation, the weights and thresholds of the neural network are adjusted in sequence from the output layer so that the error gradually decreases. Specifically, the weights and thresholds are updated by gradient descent according to the error to minimize the difference between the predicted value and the actual value. This process is repeated until the preset convergence standard is reached. The convergence standard can be that the error is less than a certain threshold, or the error no longer decreases significantly after a certain number of training times.

[0055] For each region, the regional multi-dimensional data is input into the corresponding neural network, and the forward propagation calculation is performed to obtain the predicted regional real estate market health index. Similarly, the predicted index is compared with the actual regional real estate market health index to calculate the error. The weights and thresholds of the neural network are adjusted using the back propagation algorithm to gradually reduce the error. The training process is repeated until the preset convergence standard is reached to ensure that the neural network in each region can accurately predict the regional real estate market health index.

[0056] Specifically, in some possible embodiments, the quantification and standardization processing of dynamic weight adjustment, and the step of adjusting the weight coefficient of the multi-dimensional data of the city in real time based on market trends and data sensitivity include: The specific values ​​of market trends and data sensitivity are converted into fuzzy language variables, including: for market trends, multiple key indicators are determined for evaluation; each key indicator is divided into several intervals, and the several intervals are corresponded to fuzzy language variables; for data sensitivity, the influence of the multi-dimensional data of the city on the support index of the city real estate market or the health index of the regional real estate market is analyzed; according to the degree of sensitivity, it is divided into high sensitivity, medium sensitivity and low sensitivity and corresponds to fuzzy language variables; Establish fuzzy inference rules to determine the weight adjustment strategy of the multi-dimensional data of the city under different market trends and data sensitivity combinations, and optimize and adjust the fuzzy inference rules based on historical data simulation and machine learning methods; According to the fuzzified market trends and data sensitivity, the weight coefficients of the multi-dimensional data of the city are dynamically adjusted.

[0057] Among them, first determine multiple key indicators to evaluate market trends. For example, you can choose indicators such as economic growth rate, population inflow rate, and industrial development speed. Then, divide each key indicator into several intervals. For example, for economic growth rate, it can be divided into high-speed growth intervals (such as growth rate greater than 8%), medium-speed growth intervals (growth rate between 4%-8%), and low-speed growth intervals (growth rate less than 4%). Correspond these intervals to fuzzy language variables. For example, the high-speed growth interval corresponds to "high growth trend", the medium-speed growth interval corresponds to "medium growth trend", and the low-speed growth interval corresponds to "low growth trend".

[0058] Analyze the impact of multi-dimensional urban data on the urban real estate market support index or the regional real estate market health index. Through a large amount of historical data simulation and analysis, determine the weight range of the impact of different data on the index. According to the degree of sensitivity, it is divided into high sensitivity, medium sensitivity and low sensitivity. For example, if the change of a certain data has a greater impact on the index, it is classified as high sensitivity; if the impact is moderate, it is classified as medium sensitivity; if the impact is small, it is classified as low sensitivity. Correspond high, medium and low sensitivity to fuzzy language variables, such as "high sensitivity data", "medium sensitivity data" and "low sensitivity data".

[0059] According to different market trends and data sensitivity combinations, determine the weight adjustment strategy of the city's multi-dimensional data. For example, if the market is in a high growth trend and a certain data is highly sensitive data, give the data a higher weight. A series of fuzzy reasoning rules can be formulated, such as "If the market trend is a high growth trend and the data sensitivity is highly sensitive data, then the data weight is high" and "If the market trend is a medium growth trend and the data sensitivity is medium sensitive data, then the data weight is medium".

[0060] Use a large amount of historical data for simulation, input historical market trends and data sensitivity into fuzzy inference rules, and observe the effect of weight adjustment strategy. Through machine learning methods, such as genetic algorithms and particle swarm optimization algorithms, continuously adjust the parameters and logic in fuzzy inference rules to improve the accuracy and adaptability of the rules. For example, you can adjust the definition interval of fuzzy language variables and optimize the weight distribution of fuzzy inference rules.

[0061] Monitor changes in market trends and data sensitivity in real time. The current status of market trends and data sensitivity can be determined by regularly collecting new economic data, population data, industry data, etc., and analyzing the changing trends of the data. According to the fuzzified market trends and data sensitivity, the weight coefficients of the city's multi-dimensional data are dynamically adjusted using fuzzy reasoning rules. For example, if the current market trend is a medium growth trend and the sensitivity of a certain data is medium sensitive data, the weight of the data is adjusted to a medium level according to the fuzzy reasoning rules. As the market changes, this process is repeated continuously to ensure that the weight coefficients of the city's multi-dimensional data can reflect the dynamic changes of the market in real time, so that the calculation of the city's real estate market support index and the regional real estate market health index can more accurately reflect the actual market conditions.

[0062] Specifically, in some possible embodiments, the method further includes: Rating the urban real estate market based on the urban real estate market support index, and rating the regional real estate market based on the regional real estate market health index; Obtain users' decision-making information, historical data, and trend analysis data; Combining the urban real estate market support index, regional real estate market health index, historical data and trend analysis data, a deep learning model is used to generate forecast suggestions, which include market trend judgments, potential risk warnings and corresponding response strategies.

[0063] Among them, different rating intervals are determined according to the calculated urban real estate market support index. For example, the index can be divided into three levels: high, medium and low, corresponding to different market conditions. If the urban real estate market support index is high, it means that the urban real estate market has strong support and can be given a "good" rating; if the index is at a medium level, it can be given a "general" rating; if the index is low, it means that the market is facing certain pressure and can be given a "poor" rating. Similarly, the real estate market in each region is rated according to the regional real estate market health index. Multiple rating levels can be set, such as "very healthy", "healthy", "general", "unhealthy", "very unhealthy", etc. The specific rating standards can be adjusted according to actual conditions, such as considering factors such as the level of economic development, population density, and the degree of perfection of supporting facilities in the region.

[0064] The decision-making information of users can be obtained through questionnaires filled out by users, online surveys, etc. For example, the user's intention to buy a house, investment needs, preferences for different regions, etc. The transaction records and investment behaviors of users can also be obtained from real estate trading platforms, financial institutions and other channels to understand the actual decision-making of users. Collect historical data of urban and regional real estate markets, including housing price trends, changes in transaction volume, land supply, etc. These historical data can be obtained from real estate management departments, statistical agencies, research institutions, etc., and sorted and stored. Use data analysis tools and techniques to analyze the trends of the real estate market. For example, through time series analysis, regression analysis and other methods, predict the future trend of housing prices, changes in market supply and demand, etc. It is also possible to combine macroeconomic data, policy changes and other factors to conduct a comprehensive analysis of the trends of the real estate market.

[0065] The urban real estate market support index, regional real estate market health index, historical data and trend analysis data are cleaned and standardized. Outliers and noise are removed to make the data have a unified format and scale. For trend analysis data, time series analysis methods are used to extract the main trend features to better reflect the dynamic changes of the market. A multi-layer deep learning model is constructed, including an input layer, multiple hidden layers and an output layer. The input layer receives pre-processed data, including the urban real estate market support index, regional real estate market health index, historical data and trend analysis data. The hidden layer learns complex patterns and relationships in the data through different combinations of neurons and activation functions. The appropriate hidden layer structure and activation function can be selected according to the characteristics of the data and the requirements of the task. The output layer generates forecast recommendations, including market trend judgments, potential risk warnings and corresponding response strategies. The deep learning model is trained using a large amount of historical data. Optimization algorithms such as stochastic gradient descent are used to adjust the weights and parameters of the model by minimizing the error between the forecast recommendations and the actual results. During the training process, cross-validation and other techniques are used to prevent overfitting and improve the generalization ability of the model. The historical data can be divided into training set, validation set and test set, and the model parameters can be adjusted continuously to optimize the performance of the model on the validation set. The current urban real estate market support index, regional real estate market health index and trend analysis data are input into the trained deep learning model to obtain market trend judgment, potential risk warning and corresponding response strategies. For example, if the model predicts that the market trend is rising, it may prompt potential risks, such as high housing prices and oversupply, and give corresponding response strategies, such as cautious investment and attention to policy changes. Evaluate the generated forecast suggestions. Analyze the accuracy and reliability of the forecast by comparing with the actual market situation. Adjust and optimize the deep learning model according to the evaluation results. The model structure, parameter settings, training data, etc. can be adjusted to improve the performance of the model. Obtain user feedback on the forecast suggestions and further adjust and optimize the deep learning model. User feedback can help the model better understand user needs and improve the pertinence and practicality of the forecast.

[0066] Specifically, in some possible embodiments, the steps of using a deep learning model to generate prediction suggestions in combination with the city real estate market support index, the regional real estate market health index, historical data, and trend analysis data include: Cleaning and standardizing the urban real estate market support index, the regional real estate market health index, and the historical data to remove outliers and noise so that the data have a unified format and scale. For the trend analysis data, use a time series analysis method to extract the main trend features; Build a multi-layer deep learning model, including an input layer, multiple hidden layers, and an output layer. The input layer receives preprocessed data, the hidden layer learns complex patterns and relationships in the data through different neuron combinations and activation functions, and the output layer generates forecast recommendations, including market trend judgments, potential risk warnings, and corresponding response strategies; Using a large amount of the historical data to train the deep learning model, using optimization algorithms such as stochastic gradient descent to adjust the weights and parameters of the model by minimizing the error between the predicted suggestions and the actual results. During the training process, using cross-validation and other techniques to prevent overfitting and improve the generalization ability of the model; Input the current urban real estate market support index, the regional real estate market health index and the trend analysis data into the trained deep learning model to obtain market trend judgment, potential risk warnings and corresponding response strategies; Evaluate the generated forecast suggestions and analyze the accuracy and reliability of the forecasts by comparing them with actual market information; Adjust and optimize the deep learning model based on the evaluation results; Obtain user feedback on the prediction suggestions to further adjust and optimize the deep learning model.

[0067] Among them, first identify the outliers in the data. Data points that exceed a certain range can be regarded as outliers through statistical methods, such as calculating the mean and standard deviation. For example, if a data point deviates from the mean by more than three standard deviations, it can be considered an outlier. After removing the outliers, the data is standardized. Common standardization methods, such as Z-score standardization, can be used to convert the data into a distribution with a mean of 0 and a standard deviation of 1. This can make different data have a unified scale, which is convenient for subsequent analysis and processing. For the noise that may exist in historical data, filtering technology can be used to remove it. For example, using a moving average filter can smooth the data and reduce the impact of random fluctuations. Use time series analysis methods to extract the main trend features. Time series data can be modeled using methods such as the autoregressive moving average model (ARIMA). By analyzing the autocorrelation and trend of the data, the appropriate model parameters can be determined. For example, the ARIMA model can be used to decompose time series data into trend components, seasonal components, and random components. The trend component is extracted as the main trend feature for subsequent deep learning model input.

[0068] Construct a multi-layer deep learning model, including an input layer, multiple hidden layers, and an output layer. The number of nodes in the input layer is determined according to the dimensions of the preprocessed data. For example, if the processed urban real estate market support index, regional real estate market health index, historical data, and trend analysis data have a total of features, then the input layer has nodes. The number of hidden layers and the number of nodes in each hidden layer can be adjusted according to the complexity of the data and the requirements of the task. In general, you can start with a smaller number of hidden layers and nodes and gradually increase them through experiments to find the most suitable model structure. The number of nodes in the output layer is determined according to the type of forecast recommendations. For example, if you want to generate forecast recommendations in three aspects: market trend judgment, potential risk warning, and corresponding response strategies, the output layer can have three nodes. In the hidden layer, different combinations of neurons and activation functions can be selected to learn complex patterns and relationships in the data. Common neuron combinations include fully connected layers, convolutional layers, and recurrent layers. For real estate market data, you can choose a suitable combination based on the characteristics of the data. For example, if the data has time series characteristics, you can use a recurrent neural network (RNN) or a long short-term memory network (LSTM). The activation function can be sigmoid function, tanh function, ReLU function, etc. Different activation functions have different characteristics. You can compare the effects of different activation functions through experiments and select the most suitable activation function.

[0069] Collect a large amount of historical data, including the city real estate market support index, regional real estate market health index, historical housing price data, transaction volume data, etc. Divide the historical data into training set, validation set and test set according to a certain ratio. For example, 70% of the data can be used as the training set, 20% of the data as the validation set, and 10% of the data as the test set. Use optimization algorithms such as stochastic gradient descent to train the deep learning model. Stochastic gradient descent minimizes the error between the predicted suggestions and the actual results by continuously updating the weights and parameters of the model. During the training process, the error between the predicted value and the actual value is calculated according to the loss function, and the weights and parameters of the model are updated according to the error back propagation. Common loss functions include mean square error (MSE), cross entropy loss, etc. In order to prevent overfitting and improve the generalization ability of the model, cross-validation and other techniques are used in the training process. For example, K-fold cross validation can be used to divide the training set into parts, and one of the parts is selected as the validation set each time, and the rest is used as the training set for training. After multiple training and validation, the average model performance evaluation result is obtained. You can also use regularization techniques, such as L1 regularization and L2 regularization, to constrain the weights of the model to prevent overfitting due to excessive weights.

[0070] The current urban real estate market support index, regional real estate market health index and trend analysis data are organized into a format suitable for model input. Ensure that the data has undergone the same preprocessing steps as the training data and has a unified format and scale. Input the organized data into the trained deep learning model and obtain forecast recommendations through forward propagation calculation. The output layer of the model will generate forecast results in three aspects: market trend judgment, potential risk warning and corresponding response strategies.

[0071] Collect information about actual market conditions, including current housing price trends, changes in transaction volume, policy adjustments, etc. Compare the generated forecast suggestions with the actual market conditions to analyze the accuracy and reliability of the forecast. Some evaluation indicators, such as accuracy, recall, F1 value, etc., can be used to quantitatively evaluate the quality of the forecast results. Invite experts or practitioners in the real estate field to manually evaluate the forecast suggestions. They can evaluate the rationality and practicality of the forecast results based on their experience and expertise.

[0072] Based on the evaluation results, adjust and optimize the deep learning model. If the prediction results are inaccurate, you can adjust the model's structure, parameter settings, or choose a different optimization algorithm. For example, you can increase the number of hidden layers, adjust the way neurons are connected, change the learning rate, etc. You can also further analyze and process the data to find out the factors that may affect the prediction results and make corresponding adjustments. Get user feedback on the prediction suggestions. Users can provide feedback through questionnaires, online comments, etc. Based on user feedback, further adjust and optimize the deep learning model. For example, if the user thinks the prediction results are too optimistic or pessimistic, you can adjust the model's parameters to make it more in line with the user's expectations. Continue to collect user feedback and continuously optimize the model to improve the quality and practicality of the prediction suggestions.

[0073] Specifically, in some possible embodiments, the method further includes: Visually present the city and its divided areas in the form of a map; Regions with different ratings are distinguished by different colors, and the city real estate market support index and the regional real estate market health index are displayed in the corresponding area in the form of a floating window; When the user selects a specific area, the real estate market health index of the area and the proportion of multi-dimensional data of the area are provided; Get the time node information selected by the user; Based on the time node information, the changing trends of the urban and regional real estate markets are displayed, and the changes of the multi-dimensional data of the city and the multi-dimensional data of the region over time are presented in the form of dynamic charts.

[0074] Among them, obtain the city's geographic information data, including administrative divisions, roads, rivers, etc. At the same time, obtain the results of the city's regional division and determine the boundaries and scope of each area. Associate the city's real estate market support index and the regional real estate market health index with the corresponding area for display on the map. Use geographic information system (GIS) software or development tools to draw a city map. Clearly mark the boundaries and names of each area on the map. Different colors or patterns can be used to distinguish different areas to make the map more intuitive.

[0075] According to the rating standards of the urban real estate market support index and the regional real estate market health index, determine the colors corresponding to different ratings. For example, green can be used to represent high-rated areas, yellow to represent medium-rated areas, and red to represent low-rated areas. Make sure that the color selection is clearly distinguishable so that users can quickly identify areas with different ratings. On the map, for each area, display the urban real estate market support index and regional real estate market health index of the area in the form of a floating window. The floating window can contain the specific value of the index and a brief description so that users can understand the meaning of the index.

[0076] On the map interface, set up user interaction functions to allow users to select specific areas by clicking, hovering the mouse, etc. When a user selects an area, the corresponding event is triggered to display detailed information about the area. When a user selects a specific area, the real estate market health index of the area is displayed. At the same time, provide the proportion of multi-dimensional data in the area, such as the proportion of population data, industry data, supporting facilities data, etc. in the area. This information can be displayed in the form of charts or text, so that users can have a deeper understanding of the characteristics of the area.

[0077] Design a time selection interface to allow users to select a specific time node. You can use calendar controls, time sliders, etc. to facilitate users to select time. When the user selects a time node, the information is passed to the computer for subsequent processing. Store and manage city multi-dimensional data and regional multi-dimensional data in chronological order. Ensure that data at a specific time node can be quickly obtained.

[0078] Based on the time node information selected by the user, the changing trends of the urban and regional real estate markets at different time points are analyzed. Indicators such as the rate of change of the index and the growth rate of multi-dimensional data can be calculated to reflect the dynamic changes of the market. Using data visualization tools, dynamic charts are drawn to show the changes of urban multi-dimensional data and regional multi-dimensional data over time. Different types of charts such as line charts, bar charts, and area charts can be used to select the appropriate chart format according to the characteristics of the data. In dynamic charts, animation effects can be set to make the data change process more intuitive. Users can control the display of charts through operations such as play, pause, and fast forward. Through the above specific implementation methods, cities and their divided regions can be presented in an intuitive map form, which is convenient for users to understand the overall situation and regional differences of the urban real estate market. At the same time, by distinguishing ratings, displaying indexes, and providing detailed information by different colors, as well as displaying changing trends and dynamic charts, users are provided with comprehensive and in-depth real estate market analysis and decision-making support.

[0079] Specifically, in some possible embodiments, the steps after presenting the changes of the city multi-dimensional data and the regional multi-dimensional data over time in the form of dynamic charts include: Obtain the time node information and specific area information specified by the user; Based on the urban real estate market support index, regional real estate market health index and regional multi-dimensional data corresponding to the time node information, the market conditions at the corresponding time point are analyzed, the area corresponding to the specific regional information is rated, and a detailed real estate market rating report is generated.

[0080] Among them, design a user-friendly interface to allow users to enter time node information and specific area information. You can use text boxes, drop-down menus, map selections, etc. to facilitate users to specify the required information. Provide clear instructions and prompts on the interface to guide users to enter the information correctly. After the user submits the requirements, receive the time node information and specific area information entered by the user. Verify and process this information to ensure that its format is correct and within a reasonable range. If the information entered by the user is incomplete or incorrect, provide the user with a corresponding error prompt and ask the user to re-enter.

[0081] According to the time node information specified by the user, the data at the corresponding time point is extracted from the stored urban multi-dimensional data and regional multi-dimensional data. Including the urban real estate market support index, the regional real estate market health index and the multi-dimensional data of the specific region. Analyze the extracted data to understand the market conditions at the corresponding time point. The values ​​of various indicators can be calculated, such as the housing price growth rate, the transaction volume change rate, the population inflow rate, etc. Combined with market trends and historical data, a comprehensive assessment of the market situation is made. Analyze the supply and demand relationship, price trend, development trend and other aspects of the market. Rating the region based on the real estate market health index and multi-dimensional data of a specific region. The same rating standards as before can be used to divide the region into different levels, such as high, medium and low. In the rating process, the comprehensive influence of multiple factors is considered, including population, industry, supporting facilities, real estate market supply and demand, etc. Ensure that the rating results objectively and accurately reflect the real estate market conditions in the region.

[0082] Design a detailed real estate market rating report template, including the report title, table of contents, introduction, market situation analysis, regional rating results, conclusions and suggestions. Use clear charts, tables and text descriptions in the report to make the report easy to read and understand. According to the analysis results, fill the market conditions at the corresponding time point and the rating results of the specific area into the report template. In the market situation analysis section, describe in detail the changes and trends of various indicators and their impact on the market. In the regional rating results section, clearly give the rating level of the specific area and explain the basis and reasons for the rating. In the conclusion and suggestion section, summarize the main characteristics and problems of the market, and put forward corresponding suggestions and countermeasures to provide users with decision-making references. Generate a detailed real estate market rating report based on the filled report template. The report can be saved in PDF, Word and other formats for users to download and print. At the same time, provide a link or preview function of the report on the user interface so that users can directly view the report content. If users have any questions about the report or need further explanation, online consultation or customer service support can be provided. It can accurately analyze the market conditions at the corresponding time point, rate specific areas, and generate a detailed real estate market rating report according to user needs. Provide users with personalized real estate market analysis and decision support to help users better understand market dynamics and make informed decisions.

[0083] Another embodiment of the present application provides a city area rating system, wherein, referring to Figure 2 , a city-region rating system, comprising: City multi-dimensional data acquisition module 100: acquires city multi-dimensional data, wherein the city multi-dimensional data includes city population data, city industry data, city supporting facilities data and city real estate market supply and demand relationship data; Comprehensive analysis and processing module 200: performs comprehensive analysis and processing on the multi-dimensional data of the city, including: quantification and standardization of dynamic weight adjustment, real-time adjustment of the weight coefficient of the multi-dimensional data of the city based on market trends and data sensitivity; uses a deep neural network model to perform correlation analysis and integration, and learns the complex nonlinear relationship between the multi-dimensional data of the city; The city real estate market support index calculation module 300 calculates the city real estate market support index by comprehensively analyzing the processed city multi-dimensional data and the first index model, and the weight coefficient of the first index model is dynamically adjusted; Regional real estate market health index calculation module 400: divide the city into several regions, obtain regional multi-dimensional data, calculate the regional real estate market health index through comprehensive analysis of the regional multi-dimensional data and the second index model, and dynamically adjust the weight coefficient of the second index model.

[0084] The city area rating system provided in this embodiment can implement the various steps of the aforementioned embodiment due to the functions of its modules themselves and the logical connections between each other, and can therefore achieve the same technical effects as the aforementioned embodiment. The principle analysis can be found in the relevant description of the steps of the aforementioned city area rating method, which will not be repeated here.

[0085] An embodiment of the present application further provides a city area rating device, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the above-mentioned city area rating method.

[0086] An embodiment of the present application further provides a storage medium, which stores a computer program that can be loaded by a processor and execute the above-mentioned urban area rating method.

[0087] The storage medium provided in this embodiment can achieve the same technical effects as the aforementioned embodiments because the computer program therein implements the various steps of the aforementioned embodiments after being loaded and run on the processor. The principle analysis can be found in the relevant description of the aforementioned method steps, which will not be repeated here.

[0088] The storage medium includes, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0089] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0090] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0091] In addition, the term "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined, and is only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0092] Thus, any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0093] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A city area rating method, characterized in that: include: Acquire multi-dimensional data of the city, wherein the multi-dimensional data of the city includes city population data, city industry data, city supporting facilities data and city real estate market supply and demand relationship data; Comprehensive analysis and processing of the multi-dimensional data of the city is performed, including: quantification and standardization of dynamic weight adjustment, real-time adjustment of the weight coefficient of the multi-dimensional data of the city based on market trends and data sensitivity; association analysis and integration using a deep neural network model to learn the complex nonlinear relationship between the multi-dimensional data of the city; By comprehensively analyzing the multi-dimensional data of the city and the first index model, the city real estate market support index is calculated, and the weight coefficient of the first index model is dynamically adjusted; The city is divided into several regions, and regional multi-dimensional data is obtained. The regional real estate market health index is calculated through comprehensive analysis of the regional multi-dimensional data and the second index model, and the weight coefficient of the second index model is dynamically adjusted.

2. A city area rating method according to claim 1, characterized in that: The steps of using deep neural network model for association analysis and integration include: Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the city, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the city real estate market support index; The neural network is trained by using a back propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted city real estate market support index and the actual index until a preset convergence standard is reached; When a city is divided into several regions, the city multi-dimensional data is divided to generate the regional multi-dimensional data; Constructing a neural network structure including an input layer, a hidden layer and an output layer, wherein the input layer nodes correspond to the multi-dimensional data of the region, the number of hidden layer nodes is adjusted according to the data complexity and training effect, and the output layer nodes correspond to the real estate market health index of the region; The neural network is trained using a back-propagation algorithm, and the weights and thresholds of the neural network are continuously adjusted by minimizing the error between the predicted real estate market health index of the region and the actual index until a preset convergence standard is reached.

3. A city area rating method according to claim 2, characterized in that: The quantification and standardization of dynamic weight adjustment, and the steps of adjusting the weight coefficient of the multi-dimensional data of the city in real time based on market trends and data sensitivity, include: The specific values ​​of market trends and data sensitivity are converted into fuzzy language variables, including: for market trends, multiple key indicators are determined for evaluation; each key indicator is divided into several intervals, and the several intervals are corresponded to fuzzy language variables; for data sensitivity, the influence of the multi-dimensional data of the city on the support index of the city real estate market or the health index of the regional real estate market is analyzed; according to the degree of sensitivity, it is divided into high sensitivity, medium sensitivity and low sensitivity and corresponds to fuzzy language variables; Establish fuzzy inference rules to determine the weight adjustment strategy of the multi-dimensional data of the city under different market trends and data sensitivity combinations, and optimize and adjust the fuzzy inference rules based on historical data simulation and machine learning methods; According to the fuzzified market trends and data sensitivity, the weight coefficients of the multi-dimensional data of the city are dynamically adjusted.

4. A method for rating urban areas according to claim 1, characterized in that: The method also includes: Rating the urban real estate market based on the urban real estate market support index, and rating the regional real estate market based on the regional real estate market health index; Obtain users' decision-making information, historical data, and trend analysis data; Combining the urban real estate market support index, regional real estate market health index, historical data and trend analysis data, a deep learning model is used to generate forecast suggestions, which include market trend judgments, potential risk warnings and corresponding response strategies.

5. A city area rating method according to claim 4, characterized in that: Combining the city real estate market support index, regional real estate market health index, historical data and trend analysis data, the steps to generate forecast recommendations using a deep learning model include: Cleaning and standardizing the urban real estate market support index, the regional real estate market health index, and the historical data to remove outliers and noise so that the data have a unified format and scale. For the trend analysis data, use a time series analysis method to extract the main trend features; Build a multi-layer deep learning model, including an input layer, multiple hidden layers, and an output layer. The input layer receives preprocessed data, the hidden layer learns complex patterns and relationships in the data through different neuron combinations and activation functions, and the output layer generates forecast recommendations, including market trend judgments, potential risk warnings, and corresponding response strategies; Using a large amount of the historical data to train the deep learning model, using optimization algorithms such as stochastic gradient descent to adjust the weights and parameters of the model by minimizing the error between the predicted suggestions and the actual results. During the training process, using cross-validation and other techniques to prevent overfitting and improve the generalization ability of the model; Input the current urban real estate market support index, the regional real estate market health index and the trend analysis data into the trained deep learning model to obtain market trend judgment, potential risk warnings and corresponding response strategies; Evaluate the generated forecast suggestions and analyze the accuracy and reliability of the forecasts by comparing them with actual market information; Adjust and optimize the deep learning model based on the evaluation results; Obtain user feedback on the prediction suggestions to further adjust and optimize the deep learning model.

6. A city area rating method according to claim 1, characterized in that: The method also includes: Visually present the city and its divided areas in the form of a map; Regions with different ratings are distinguished by different colors, and the city real estate market support index and the regional real estate market health index are displayed in the corresponding area in the form of a floating window; When the user selects a specific area, the real estate market health index of the area and the proportion of multi-dimensional data of the area are provided; Get the time node information selected by the user; Based on the time node information, the changing trends of the urban and regional real estate markets are displayed, and the changes of the multi-dimensional data of the city and the multi-dimensional data of the region over time are presented in the form of dynamic charts.

7. A method for rating urban areas according to claim 6, characterized in that: The steps after presenting the changes of the city multi-dimensional data and the regional multi-dimensional data over time in the form of dynamic charts include: Obtain the time node information and specific area information specified by the user; Based on the urban real estate market support index, regional real estate market health index and regional multi-dimensional data corresponding to the time node information, the market conditions at the corresponding time point are analyzed, the area corresponding to the specific regional information is rated, and a detailed real estate market rating report is generated.

8. A city area rating system, characterized in that: include: City multi-dimensional data acquisition module: acquires city multi-dimensional data, including city population data, city industry data, city supporting facilities data and city real estate market supply and demand relationship data; Comprehensive analysis and processing module: Comprehensive analysis and processing of the multi-dimensional data of the city, including: quantification and standardization of dynamic weight adjustment, real-time adjustment of the weight coefficient of the multi-dimensional data of the city based on market trends and data sensitivity; using a deep neural network model for correlation analysis and integration, and learning the complex nonlinear relationship between the multi-dimensional data of the city; Urban real estate market support index calculation module: calculates the urban real estate market support index through comprehensive analysis of the multi-dimensional data of the city and the first index model, and dynamically adjusts the weight coefficient of the first index model; Regional real estate market health index calculation module: divide the city into several regions, obtain regional multi-dimensional data, calculate the regional real estate market health index through comprehensive analysis of the regional multi-dimensional data and the second index model, and dynamically adjust the weight coefficient of the second index model.

9. A city area rating device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any one of the urban area rating methods according to claims 1-7.

10. A storage medium, characterized in that: The computer program is stored which can be loaded by a processor and execute any one of the urban area rating methods according to claims 1-7.

Images