Real estate cycle prediction method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a real estate cycle prediction method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring characteristic data influencing periodic rotation of the real estate; seasonal state judgment is conducted on the feature data, initial seasonal labels and confidence coefficients of the real estate are obtained, and the initial seasonal labels represent flowing period information of the real estate; correcting the initial season label based on the regional level information of the real estate to obtain a corrected season label; obtaining a real estate cycle conversion probability of the region based on the corrected season label and the confidence coefficient; and based on the real estate period transition probability, obtaining a real estate period prediction result. Through the method, the real estate cycle fluctuation rule can be deeply analyzed in combination with the macroscopic regulation and control factor and the microscopic behavior data, and the precision of real estate cycle prediction is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a real estate cycle prediction method and system, an electronic device, and a storage medium. BACKGROUND

[0002] Due to significant differences in market responses between different levels of cities, especially the phase difference of up to 9 months in market cycles between first-tier and third- and fourth-tier cities, traditional analysis models usually use a unified index system and fail to fully consider the dynamic signals of the real market, resulting in insufficient accuracy in capturing the real state of the market. Traditional methods often rely on single-dimensional indicators (such as housing price indexes) for judgment, which can easily misjudge the short-term market recovery caused by regional adjustments as a long-term upturn cycle, thus causing misinterpretation of market trends. SUMMARY

[0003] Embodiments of the present application aim to at least partially address one of the technical problems in the related art. To this end, embodiments of the present application propose a real estate cycle prediction method and system, an electronic device, and a storage medium.

[0004] Embodiments of the present application provide a real estate cycle prediction method, which includes: obtaining feature data affecting real estate cycle rotation; performing seasonal state determination on the feature data to obtain an initial seasonal label and a confidence level of the real estate, wherein the initial seasonal label represents real estate flow cycle information; correcting the initial seasonal label based on regional level information where the real estate is located to obtain a corrected seasonal label; obtaining a real estate cycle conversion probability of the region based on the corrected seasonal label and the confidence level; and obtaining a cycle prediction result of the real estate based on the real estate cycle conversion probability.

[0005] In some embodiments, the feature data includes transaction volume and price; performing seasonal state determination on the feature data to obtain an initial seasonal label and a confidence level of the real estate includes: inputting the transaction volume and the price into a seasonal state machine; determining the transaction volume based on a first preset threshold by the seasonal state machine; determining the price based on a second preset threshold by the seasonal state machine; and outputting the initial seasonal label and the confidence level of the real estate.

[0006] In some embodiments, the initial seasonal label includes spring, summer, autumn, winter and transition state, wherein the spring reflects a state of stable price increase of the real estate, the summer reflects a state of stable price increase of the real estate, the autumn reflects a state of stable price decrease of the real estate, the winter reflects a state of stable price decrease of the real estate, and the confidence level includes spring confidence, summer confidence, autumn confidence, winter confidence and transition confidence; the transaction volume relative change is determined by the seasonal state machine based on a first preset threshold, the price relative change is determined by the seasonal state machine based on a second preset threshold, and the initial seasonal label and the confidence level of the real estate are output, including: if the transaction volume relative change is greater than or equal to the first preset threshold, and the absolute value of the price relative change is less than or equal to the second preset threshold, it is determined that the initial seasonal label is spring, and the spring confidence corresponding to the spring is output; if the transaction volume relative change is greater than or equal to the first preset threshold, and the price relative change is greater than the second preset threshold, it is determined that the initial seasonal label is summer, and the summer confidence corresponding to the summer is output; if the transaction volume relative change is less than the first preset threshold, and the absolute value of the price relative change is less than or equal to the second preset threshold, it is determined that the initial seasonal label is autumn, and the autumn confidence corresponding to the autumn is output; if the transaction volume relative change is less than the first preset threshold, and the price relative change is less than the third preset threshold, it is determined that the initial seasonal label is winter, and the winter confidence corresponding to the winter is output; if the transaction volume relative change is less than the first preset threshold, and the price relative change is greater than the second preset threshold, it is determined that the initial seasonal label is transition state, and the transition confidence corresponding to the transition state is output.

[0007] In some embodiments, based on the regional level information of the real estate, the initial seasonal label is phase corrected to obtain a corrected seasonal label, including: obtaining historical regional phase difference data of the region; classifying the region based on the regional level information to obtain regional level classification data; establishing a lag mapping relationship between the regional level and the season based on the historical regional phase difference data and the regional level classification data; calculating a seasonal offset based on the seasonal transition rule, the lag mapping relationship and the initial seasonal label; correcting the initial seasonal label based on the seasonal offset to obtain the corrected seasonal label.

[0008] In some embodiments, based on the corrected seasonal label and the confidence level, a real estate cycle transition probability of the region is obtained, including: obtaining a regional adjustment factor and a historical seasonal transition influence matrix that affect the region where the real estate is located; determining an initial seasonal probability vector of the real estate based on the corrected seasonal label and the confidence level; performing weighted calculation on the initial seasonal probability vector and the historical seasonal transition influence matrix based on the regional adjustment factor to obtain a target seasonal transition influence matrix; performing probability normalization processing on the target seasonal transition influence matrix based on the initial seasonal probability vector to obtain the real estate cycle transition probability of the region.

[0009] In some embodiments, based on the regional adjustment factor, the initial seasonal probability vector and the historical seasonal transition influence matrix are weighted to obtain a target seasonal transition influence matrix, including: based on the Bayesian algorithm, the historical seasonal transition influence matrix is weighted based on the regional adjustment factor to obtain a regional adjustment weight; the historical seasonal transition influence matrix is weighted and adjusted based on the regional adjustment weight to obtain the target seasonal transition influence matrix.

[0010] In some embodiments, the method further comprises: performing probability normalization processing on the target seasonal transition influence matrix to obtain a regional seasonal heat map and a strategy recommendation matrix of the region.

[0011] The scheme provided by the present application realizes deep analysis and forward-looking prediction of the real estate market cycle fluctuation rule by constructing a four-season mathematical definition system based on the quantity-price synergistic relationship. Through the seasonal state degradation mechanism of integrating micro-behavior signals, the abstract market cycle is converted into quantifiable and operable decision information. The system not only can capture the instant changes of market internal indicators such as quantity-price synergy, but also can output the real estate cycle prediction results with spatio-temporal specificity and dynamic adaptability based on the integer seasonal correction algorithm of regional phase difference and the Bayesian probability updating framework of regional adjustment factor, providing a prediction model connecting market reality and real estate cycle rule decision-making, and providing a decision-making scheme for real estate cycle prediction with both macro-trend grasping and micro-decision support.

[0012] The embodiment of the present application provides a real estate cycle prediction system, the system comprising: an acquisition module for acquiring feature data affecting real estate cycle rotation; a determination module for determining the seasonal state of the feature data to obtain an initial seasonal label and a confidence degree of the real estate cycle, wherein the initial seasonal label represents the flow cycle information of the real estate; a correction module for correcting the initial seasonal label based on regional level information of the real estate to obtain a corrected seasonal label; a calculation module for obtaining a real estate cycle transition probability of the region based on the corrected seasonal label and the confidence degree; and a prediction module for obtaining a cycle prediction result of the real estate based on the real estate cycle transition probability.

[0013] The embodiment of the present application provides an electronic device, which comprises a memory and one or more processors in communication connection with the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method of any one of the above-mentioned embodiments.

[0014] The embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method of any one of the above-mentioned embodiments.

[0015] Embodiments of the present application provide a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method according to any of the above embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A schematic diagram of a causal defect chain of traditional real estate cycle analysis provided for an embodiment of the present application is shown. Figure 2 A flowchart of a real estate cycle prediction method provided for the present application is shown. Figure 3 A schematic diagram of a real estate cycle prediction system architecture provided for an embodiment of the present application is shown. Figure 4 A schematic diagram of a real estate cycle prediction system provided for the present application is shown. Figure 5 A block diagram of an electronic device provided for an embodiment of the present application is shown. DETAILED DESCRIPTION

[0017] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0018] Figure 1 A schematic diagram of a causal defect chain of traditional real estate cycle analysis provided for an embodiment of the present application is shown.

[0019] As shown in Figure 1 , for example, traditional real estate cycle analysis has a clear causal defect chain, mainly in three aspects: relying on a single price index can easily confuse the market short-term promotion behavior of "price for quantity" with the essential difference of real recovery; ignoring structural differences between different cities can lead to a loss of precision in city rotation strategies; and using static and unchanging thresholds to divide market stages, especially during periods of frequent policy disturbance, the misjudgment rate can exceed 60%.

[0020] For example, based on the historical data backtesting results of fifty key cities from 2018 to 2024, the following table shows the results:

[0021] As shown in the above table, the real estate cycle prediction method provided by the present application has an accuracy of 92% in season transition identification, which is 58.6% higher than the traditional method of 58%. The city rotation strategy yield is greatly improved from 11.3% to 28.7%, with an increase of ↑ 154%. The regional adjustment window period misjudgment rate is reduced from 47% to 9%, with an improvement of 81%. The most breakthrough is that the market inflection point prediction lead time is extended from the median of 0.8 months to 2.3 months, nearly three times, which provides a valuable time window for decision makers.

[0022] Typical examples of two regions A and B are provided. In the A region market in March 2024, the traditional model incorrectly determines "summer" based on the surface feature of rising prices. However, the scheme provided by the present application accurately predicts the "autumn" adjustment signal by capturing the micro-behavioral changes of "real estate listing volume surge superimposed with viewing conversion rate decline", and the subsequent two months' transaction volume drops by 32% to verify this judgment.

[0023] In the early stage of the introduction of a regional adjustment factor in region B, the regional adjustment factor includes real estate policy information. The traditional model infers that the market is in the "spring" recovery stage due to the relatively loose and simple regional adjustment. However, the scheme provided by the present application identifies the strong internal energy through the quantity-price coordination coefficient of 0.93 (close to the strong positive correlation threshold), and confidently determines that it should be the "summer" prosperity stage. The subsequent three months' real estate price rise of 12% confirms the prediction.

[0024] These quantitative results collectively reflect that the real estate cycle prediction method provided by the present application can not only significantly improve the accuracy of real estate cycle prediction, but also provide key and effective decision-making schemes by identifying the inflection point in advance.

[0025] The limitations of these methods reflect the inadequate adaptability of traditional analysis methods in dynamic market environments, and there is an urgent need for a more scientific and flexible prediction and evaluation system to improve the accuracy of judgment.

[0026] Therefore, the present application provides a real estate cycle prediction method, which can improve the accuracy of real estate cycle rotation prediction.

[0027] Figure 2 A flowchart of a real estate cycle prediction method provided by the present application.

[0028] As Figure 2 shown, the real estate cycle prediction method 200 provided by the present application includes steps S210-S250.

[0029] Step S210: Obtain feature data affecting real estate cycle rotation.

[0030] Exemplarily, the characteristic data affecting the real estate cycle rotation include, for example, real estate price fluctuation data, real estate inventory data, developer behavior data, consumer behavior data, and the like.

[0031] Step S220: Seasonal state determination is performed on the characteristic data to obtain an initial seasonal label and a confidence degree of the real estate, wherein the initial seasonal label represents the flow cycle information of the real estate.

[0032] Exemplarily, through comprehensive analysis of the multi-dimensional characteristic data, the system can quantitatively interpret the flow of the real estate market cycle. The initial seasonal label can directly represent the periodic stage in which the current real estate market is located, and the confidence degree measures the reliability of the label from the probability angle.

[0033] Step S230: Based on the regional level information of the real estate, the initial seasonal label is corrected to obtain a corrected seasonal label.

[0034] Exemplarily, in the real estate market cycle analysis, the regional level information plays a key calibration role. In the scheme provided in the present application, the Bayesian correction algorithm can be used to correct the initial seasonal label based on the regional level information. The regional level information is, for example, the city energy level dimension. The system introduces the city energy level dimension to finely adjust the spatial dimension of the initial seasonal label, so as to output a corrected seasonal label that is more consistent with the real situation of the regional market.

[0035] Step S240: Based on the corrected seasonal label and the confidence degree, a real estate cycle conversion probability of the region is obtained.

[0036] Exemplarily, based on the corrected seasonal label and the confidence degree, the system can generate a real estate cycle conversion probability reflecting the evolution trend of the real estate market cycle of the region. The probability output is essentially a quantitative prediction of the possible future path of the market.

[0037] Step S250: Based on the real estate cycle conversion probability, a cycle prediction result of the real estate is obtained.

[0038] The scheme provided in the present application realizes deep analysis and forward-looking prediction of the real estate market cycle fluctuation rule by constructing a four-season mathematical definition system based on the quantity-price coordination relationship. Through the seasonal state degradation mechanism of integrating micro-behavior signals, the abstract market cycle is converted into quantifiable and operable decision information. The system can not only capture the instant changes of market internal indicators such as quantity-price coordination, but also output a cycle prediction result of the real estate with spatio-temporal specificity and dynamic adaptability based on the integerized seasonal correction algorithm of regional phase difference and the Bayesian probability update framework of regional adjustment factor, thereby providing a prediction model connecting market reality and real estate cycle rule decision, and providing a decision scheme for real estate cycle prediction with both macro-trend grasping and micro-decision support.

[0039] Figure 3 A schematic diagram of a real estate cycle prediction system architecture provided by an embodiment of the present application.

[0040] In another embodiment, the feature data includes transaction volume month-on-month and price month-on-month; the seasonal state determination on the feature data obtains an initial seasonal label and confidence of the real estate, including: inputting the transaction volume month-on-month and the price month-on-month into a seasonal state machine, determining the transaction volume month-on-month based on a first preset threshold by the seasonal state machine, determining the price month-on-month based on a second preset threshold by the seasonal state machine, and outputting the initial seasonal label and the confidence of the real estate.

[0041] As shown in Figure 3 , for example, the feature data can include city volume and price data, micro-behavior data, and macro-policy factors in addition to transaction volume month-on-month and price month-on-month; the city volume and price data can include real estate transaction volume data, real estate price data, and real estate inventory constraint conditions in addition to transaction volume month-on-month and price month-on-month; the micro-behavior data includes, for example, real estate consumer data and real estate developer data; and the macro-policy factors include, for example, real estate financial policy factors and administrative control factors.

[0042] As shown in Figure 3 , for example, the real estate cycle prediction system architecture proposed in the embodiment includes an input layer, a processing layer, and an output layer. Based on the feature engine module in the input layer, the feature data is preliminarily processed, and then the preliminarily processed feature data is input into the four-season determination engine of the processing layer for in-depth analysis.

[0043] Specifically, in the four-season determination link, the system first processes 30-day data and preliminarily determines through transaction volume month-on-month changes. When the transaction volume month-on-month growth exceeds a first preset threshold, for example, the first preset threshold is 5%, the change in the absolute value of the price month-on-month is further investigated: if the absolute value of the price month-on-month fluctuation does not exceed a second preset threshold, for example, the second preset threshold is 1%, it is determined as "spring" state; if the absolute value of the price month-on-month increases by more than 1%, it is determined as "summer" state. When the transaction volume month-on-month is less than 5%, the absolute value of the price month-on-month fluctuation does not exceed 1%, it is "autumn" state, and the price month-on-month drop is lower than -1%, it is determined as "winter" state.

[0044] In the embodiment of the present application, the feature data including city volume and price data, micro-behavior data, and macro-policy factors is preliminarily extracted, and then the above preliminarily processed data is analyzed by the four-season determination engine for volume and price coordination, and then an accurate seasonal state is output, which can greatly improve the accuracy of real estate cycle prediction.

[0045] In another embodiment, the initial seasonal label includes spring, summer, autumn, winter and transition, wherein the spring reflects a state of stable price and rising volume of real estate, the summer reflects a state of rising price and volume of real estate, the autumn reflects a state of stable price and falling volume of real estate, the winter reflects a state of falling price and volume of real estate, and the confidence level includes spring confidence, summer confidence, autumn confidence, winter confidence and transition confidence; the season state machine determines the volume ring ratio based on the first preset threshold, determines the price ring ratio based on the second preset threshold, and outputs the initial seasonal label and the confidence level of the real estate.

[0046] For example, the first preset threshold is 5%, the second preset threshold is 1%, and the third preset threshold is -1%.

[0047] In the embodiments provided in the present application, the market fluctuations are converted into intuitive natural law metaphors by using seasons to represent the real estate cycle. Based on the four seasons, the real estate market life cycle is expressed in a lower dimension, and the four seasons correspond to the recovery period of stable price and rising volume, the prosperity period of rising price and volume, the stagnant adjustment period of stable price and falling volume, and the market freezing period of falling price and volume.

[0048] The mathematical definition system of the volume-price relationship four seasons provided in the embodiment specifically includes: If the volume ring ratio is greater than or equal to the first preset threshold, and the absolute value of the price ring ratio is less than or equal to the second preset threshold, the initial seasonal label is determined to be spring, and the spring confidence corresponding to the spring is output. For example, if the volume ring ratio growth is greater than or equal to 5%, and the absolute value of the price ring ratio fluctuation does not exceed 1%, the initial seasonal label of the current can be determined to be spring, reflecting a state of stable price and rising volume of real estate, and the spring confidence of this state is, for example, 0.8, and the range of the volume-price coordination coefficient is 0.6-0.8.

[0049] For example, the initial seasonal label is obtained by the first labeling processing of the season state machine in the real estate cycle prediction system architecture as shown in Figure 3 The essence of the initial seasonal label is the original state machine output without phase correction, which is completely based on the first threshold matching result of the input data of the volume-price coordination analysis module.

[0050] If the volume ring ratio is greater than or equal to the first preset threshold, and the price ring ratio is greater than the second preset threshold, the initial seasonal label is determined to be summer, and the summer confidence corresponding to the summer is output.

[0051] Exemplarily, if the trading volume growth is greater than or equal to 5% and the absolute value of the price growth is greater than 1%, the initial season label is determined to be spring, reflecting a state of quantity and price rising together of the real estate, and the summer confidence corresponding to the state is 0.7, for example, obtaining a range of the quantity-price coordination coefficient of 0.8-1.0. The dynamic adjustment rule of the summer confidence is that the confidence is increased by 0.2 (upper limit 1.0) for each 1% increase in the price growth.

[0052] If the trading volume growth is less than the first preset threshold and the absolute value of the price growth is less than or equal to the second preset threshold, the initial season label is determined to be autumn, and the autumn confidence corresponding to the autumn is output.

[0053] Exemplarily, if the trading volume growth is less than 5% and the absolute value of the price growth is not more than 1%, the initial season label is determined to be autumn, reflecting a state of quantity falling and price stable of the real estate, and the autumn confidence corresponding to the state is 0.7, for example, obtaining a range of the quantity-price coordination coefficient of 0.5-0.7.

[0054] If the trading volume growth is less than the first preset threshold and the price growth is less than the third preset threshold, the initial season label is determined to be winter, and the winter confidence corresponding to the winter is output.

[0055] Exemplarily, if the trading volume growth is less than 5% and the absolute value of the price growth is not more than -1%, the initial season label is determined to be winter, reflecting a state of quantity and price falling together of the real estate, and the winter confidence corresponding to the state is 0.9, for example, obtaining a range of the quantity-price coordination coefficient of 0.9-1.0. The dynamic adjustment rule of the winter confidence is that the confidence is decreased by 0.1 (lower limit 0.5) for each 1% increase in the price fall.

[0056] If the trading volume growth is less than the first preset threshold and the price growth is greater than the second preset threshold, the initial season label is determined to be a transition state, and the transition confidence corresponding to the transition state is output.

[0057] Exemplarily, if the trading volume growth is less than 5% and the price growth is greater than 1%, the initial season label is determined to be a transition state, and the transition confidence corresponding to the transition state is 0.5.

[0058] In the embodiments of the present application, by analyzing and processing multi-dimensional data of the real estate market, relying on the dual processing of the feature engine and the four-season judgment engine, revealing the internal linkage characteristics of the market through quantity-price coordinated analysis, achieving accurate discrimination of the cycle stage through the seasonal state machine, forming a layer-by-layer progressive intelligent analysis chain, ensuring that the decision basis provided has both scientificity and accuracy, and guaranteeing the analysis depth while greatly improving the accuracy of the results. The three-dimensional input system constructed in the embodiments of the present application can ensure that the output results cover multiple elements such as behavior subjects and regional factors, so that the real estate cycle prediction results effectively reflect the multi-level content of micro and macro.

[0059] In another embodiment, based on the regional level information of the real estate, the initial season label is corrected to obtain a corrected season label, including: obtaining historical regional phase difference data of the region; classifying the region based on the regional level information to obtain regional level classification data; establishing a lag mapping relationship between the regional level and the season based on the historical regional phase difference data and the regional level classification data; calculating a season offset based on the season transition rule, the lag mapping relationship and the initial season label; correcting the initial season label based on the season offset to obtain the corrected season label.

[0060] Exemplarily, the regional level information of the real estate includes, for example, the administrative level, economic level, population size, radiation capacity and other dimensional information of the city where the real estate is located, and the regional level classification data includes, for example, first-tier cities, new first-tier cities, second-tier cities and third-fourth-tier cities. The historical regional phase difference data is, for example, the difference in response speed of different regional levels to the cycle change of the real estate market.

[0061] Exemplarily, the region can be classified according to the regional level information first, for example, first-tier cities are classified into the first category, new first-tier cities are classified into the second category, second-tier cities are classified into the third category, and third-fourth-tier cities are classified into the fourth category. The historical regional phase difference data is, for example, that first-tier cities lag by 0 months, new first-tier cities lag by 1 month, second-tier cities lag by 3 months, and third-fourth-tier cities lag by 6 months respectively. Then, according to the seasonal rotation rule, the seasonal cycle is defined as a closed loop sequence of "spring→summer→autumn→winter", and then according to the input regional level information, the corresponding lag months are matched to establish the lag mapping relationship between the regional level and the season. Finally, the season offset is calculated by dividing the lag months by the season transition rule (3 months), and the initial season label is corrected based on the season offset to obtain the corrected season label. For example, when the third-fourth-tier city (lagging by 6 months) inputs the current season as "summer", the algorithm will backtrack two complete seasons to return the "spring" label, thereby reflecting the season offset of the third-fourth-tier city relative to the first-tier city.

[0062] In the embodiments of the present application, by introducing the regional level information where the real estate is located, the initial season label is phase-corrected, effectively solving the cycle misjudgment problem caused by regional development differences in traditional analysis. Through the integer season correction algorithm of regional phase difference, the initial season label is automatically corrected, making the corrected season label more adaptive to the regional level information where the real estate is located, and making the cycle prediction result more consistent with the real development stage of the region.

[0063] In another embodiment, based on the corrected season label and the confidence, the real estate cycle conversion probability of the region is obtained, including: obtaining a regional adjustment factor affecting the region where the real estate is located, a historical season conversion influence matrix; based on the corrected season label and the confidence, determining an initial season probability vector of the real estate; based on the regional adjustment factor, performing weighted calculation on the initial season probability vector and the historical season conversion influence matrix to obtain a target season conversion influence matrix; based on the initial season probability vector, performing probability normalization processing on the target season conversion influence matrix to obtain the real estate cycle conversion probability of the region.

[0064] In another embodiment, based on the regional adjustment factor, the weighted calculation is performed on the initial season probability vector and the historical season conversion influence matrix to obtain the target season conversion influence matrix, including: based on the Bayesian algorithm, the weighted calculation is performed on the historical season conversion influence matrix according to the regional adjustment factor to obtain a regional adjustment weight; based on the regional adjustment weight, the weighted adjustment is performed on the historical season conversion influence matrix to obtain the target season conversion influence matrix.

[0065] Exemplarily, the embodiments provided by the present application also propose to construct the target season conversion influence matrix based on the Bayesian method. The target season conversion influence matrix is obtained by adjusting the historical season conversion influence matrix, and expresses the transition probability between seasons in a mathematical form, covering all possible state transitions from spring to winter. The matrix is used to construct a season state transition probability model under the influence of the regional adjustment factor, which makes it possible to dynamically adjust the transition probability between season states based on the regional adjustment factor, and provides a theoretical basis and calculation tool for the cycle prediction result of the real estate under regional adjustment.

[0066] Specifically, the historical season conversion influence matrix, for example, describes the evolution law between the four stages of spring, summer, autumn and winter in the natural state of the market, and each element in the matrix represents the objective possibility of transition from the current season to the next season, for example: When the market is in spring, there is a 60% probability that it will remain in spring (i.e. the quantity rises and the price is stable), a 30% probability that it will enter summer (i.e. the quantity and price rise simultaneously), a 10% probability that it will jump to autumn (i.e. the quantity falls and the price is stable), and a 0% probability that it will directly fall into winter (i.e. the quantity and price fall simultaneously).

[0067] When the market is in summer, there is a 50% probability of maintaining summer, a 40% probability of turning into autumn, a 10% probability of accidentally falling back to spring, and a 0% probability of directly jumping into winter.

[0068] When the market is in autumn, there is a 50% probability of maintaining autumn, a 30% probability of turning into winter, a 20% probability of turning back to summer, and a 0% probability of directly jumping back to spring.

[0069] When the market is in winter, there is a 40% probability of maintaining winter, a 40% probability of warming up to spring, a 20% probability of turning into autumn, and a 0% probability of directly jumping into summer.

[0070] When a new regional adjustment factor is input, the system can generate a dynamic weight coefficient of 0 to 2 times according to the influence strength of the regional adjustment factor, which is valued in the range of -5 to +5 of the regional adjustment factor, and then adjust the historical seasonal transition influence matrix through exponential weighting to obtain the target seasonal transition influence matrix. The following formula can be used for weighted calculation:

[0071]

[0072] wherein, is a regional adjustment factor, is a weighting coefficient, is a historical seasonal transition influence matrix, is a target seasonal transition influence matrix.

[0073] If the regional adjustment factor is strong (such as a significant reduction in the down payment ratio), the algorithm will amplify the transition probability of summer; if the regional adjustment factor is tight (such as additional purchase restrictions), the transition weight from autumn to winter will be enhanced. Finally, the initial seasonal probability vector is combined with the target seasonal transition influence matrix through matrix multiplication to output the real estate cycle transition probability calibrated by the regional adjustment factor.

[0074] Through the embodiments of the present application, the basic law framework of the market cycle can be retained, and the regional adjustment factor can be flexibly adjusted, so as to dynamically reflect the disturbance effect of macro-control on the real estate cycle rotation, realize the phase difference correction of the real estate cycle, provide data support for the prediction result, and improve the accuracy of the prediction.

[0075] In another embodiment, the target seasonal transition influence matrix is also subjected to probability normalization processing, and a regional seasonal heat map and a strategy recommendation matrix of the region can also be obtained.

[0076] Exemplarily, the application provides a micro-macro dual-track early warning mechanism. At the micro level, dynamic early warning is achieved by monitoring changes in specific business indicators. For example, when the viewing conversion rate decay coefficient is greater than 0.15 and the negotiation space is greater than 8%, the system forces the "summer" to be downgraded to "spring". The viewing conversion rate decay coefficient is calculated based on the transaction volume and the viewing volume of real estate, for example. At the macro level, the system makes judgments based on wider regional adjustments. For example, when the stimulating intensity of the regional adjustment factor is greater than 3.0, the seasonal state reevaluation is triggered. The dual-track mechanism takes into account both the details and the overall trend, effectively improving the accuracy and adaptability of early warning.

[0077] As shown in Figure 3 Exemplarily, the regional seasonal heat map is, for example, a city seasonal heat map of different cities. The content output by the output layer of the system can include a city seasonal heat map, a cycle conversion probability, and a strategy recommendation matrix. The city seasonal heat map can present the spatial distribution of the cycle state of each city, for example. The strategy recommendation matrix can output a differentiated action plan for macro-control factors, sellers, and consumers, for example.

[0078] Figure 4 A schematic diagram of a real estate cycle prediction system provided by the application.

[0079] As shown in Figure 4 The real estate cycle prediction system 400 provided by the application includes: The acquisition module 410 is configured to acquire feature data that affects the real estate cycle rotation.

[0080] The determination module 420 is configured to determine the seasonal state of the feature data to obtain an initial seasonal label and a confidence level of the real estate cycle, wherein the initial seasonal label represents the flow cycle information of the real estate.

[0081] The correction module 430 is configured to correct the initial seasonal label based on regional level information of the real estate to obtain a corrected seasonal label.

[0082] The calculation module 440 is configured to obtain a real estate cycle conversion probability of the region based on the corrected seasonal label and the confidence level.

[0083] The prediction module 450 is configured to obtain a cycle prediction result of the real estate based on the real estate cycle conversion probability.

[0084] In another embodiment, the determination module 420 is further configured to: input the transaction volume and the price into a seasonal state machine, determine the transaction volume based on a first preset threshold by the seasonal state machine, determine the price based on a second preset threshold by the seasonal state machine, and output the initial seasonal label and the confidence level of the real estate.

[0085] In another embodiment, the determining module 420 is further configured to: if the trading volume relative change is greater than or equal to a first preset threshold value, and the absolute value of the price relative change is less than or equal to a second preset threshold value, determine that the initial seasonal label is spring, and output a spring confidence corresponding to the spring; if the trading volume relative change is greater than or equal to the first preset threshold value, and the price relative change is greater than the second preset threshold value, determine that the initial seasonal label is summer, and output a summer confidence corresponding to the summer; if the trading volume relative change is less than the first preset threshold value, and the absolute value of the price relative change is less than or equal to the second preset threshold value, determine that the initial seasonal label is autumn, and output an autumn confidence corresponding to the autumn; if the trading volume relative change is less than the first preset threshold value, and the price relative change is less than a third preset threshold value, determine that the initial seasonal label is winter, and output a winter confidence corresponding to the winter; if the trading volume relative change is less than the first preset threshold value, and the price relative change is greater than the second preset threshold value, determine that the initial seasonal label is transition, and output a transition confidence corresponding to the transition.

[0086] In another embodiment, the correcting module 430 is further configured to: obtain historical regional phase difference data of the region; classify the region based on the regional level information to obtain regional level classification data; establish a lag mapping relationship between the regional level and the season based on the historical regional phase difference data and the regional level classification data; calculate a seasonal offset based on the seasonal transition rule, the lag mapping relationship, and the initial seasonal label; and correct the initial seasonal label based on the seasonal offset to obtain a corrected seasonal label.

[0087] In another embodiment, the obtaining module 410 is further configured to: obtain a regional adjustment factor and a historical seasonal transition influence matrix that affect a region where the real estate is located; and the calculating module 440 is further configured to: determine an initial seasonal probability vector of the real estate based on the corrected seasonal label and the confidence; perform weighted calculation on the initial seasonal probability vector and the historical seasonal transition influence matrix based on the regional adjustment factor to obtain a target seasonal transition influence matrix; and perform probability normalization processing on the target seasonal transition influence matrix based on the initial seasonal probability vector to obtain a real estate cycle transition probability of the region.

[0088] In another embodiment, the calculating module 440 is further configured to: perform weighted calculation on the historical seasonal transition influence matrix based on the Bayesian algorithm and the regional adjustment factor to obtain a regional adjustment weight; and perform weighted adjustment on the historical seasonal transition influence matrix based on the regional adjustment weight to obtain the target seasonal transition influence matrix.

[0089] In another embodiment, the calculating module 440 is further configured to: perform probability normalization processing on the target seasonal transition influence matrix to obtain a regional seasonal heat map and a strategy recommendation matrix of the region.

[0090] It can be understood that the specific description of the real estate cycle prediction system 400 can refer to the description of the real estate cycle prediction method 200 in the foregoing.

[0091] The embodiment of the present application provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps of the method in any of the foregoing embodiments.

[0092] The embodiment of the present application provides a computer program product, which includes instructions. The instructions are executed by a processor of a computer device to enable the computer device to perform the steps of the method in any of the foregoing embodiments.

[0093] Figure 5 The embodiment of the present application provides a block diagram of an electronic device.

[0094] The embodiment of the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program. The processor executes the computer program to implement the method in any of the foregoing embodiments.

[0095] As shown in Figure 5 To facilitate understanding, the embodiment of the present application shows a specific electronic device 500.

[0096] The electronic device 500 is intended to represent various forms including digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components, their connections, and their functions, as described and illustrated herein, are by way of example only, and are not intended to limit the implementations of the disclosure described and / or claimed in this document.

[0097] As shown in Figure 5 The device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded into a random access memory (RAM) 503 from a storage unit 508. In the RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0098] A plurality of components in the electronic device 500, including an input unit 506, such as a keyboard, a mouse, etc., an output unit 507, such as various types of displays, a speaker, etc., a storage unit 508, such as a magnetic disk, an optical disk, etc., and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc., are connected to the I / O interface 505. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0099] The computing unit 501 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 501 performs various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program, which is tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded onto the RAM 503 and executed by the computing unit 501, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform any one or more of the methods described above by other any appropriate means, such as by means of firmware.

[0100] It should be noted that the logical and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a list of ordered steps to implement logic functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- based system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or in conjunction with which the instructions can be executed. In the context of this application, a "computer-readable medium" can be any means that can contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electronic connection (an electronic device having one or more wires), a portable computer diskette (a magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example, via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a suitable format.

[0101] It should be understood that portions of the application can be implemented in hardware, software, firmware, or combinations thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies known in the art, or combinations thereof, can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0102] In the description of the present application, references to terms such as "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. The illustrative appearances of the above-mentioned terms in various places in the specification are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0103] In the description of the present application, it needs to be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the purpose of facilitating the description of the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0104] In addition, the terms "first", "second", and the like used in the embodiments of the present application are only for the purpose of description, and cannot be understood as indicating or implying relative importance, or implicitly indicating the number of technical features referred to in the embodiments. Therefore, the features defined with the terms "first", "second" and the like in the embodiments of the present application can be explicitly or implicitly indicated to include at least one of the features. In the description of the present application, the meaning of the word "plurality" is at least two or two or more, such as two, three, four, etc., unless otherwise specifically limited in the embodiments.

[0105] In the present application, unless otherwise specifically defined or limited in the embodiments, the terms "mounting", "connecting", "connecting" and "fixing" and the like appearing in the embodiments should be understood broadly, for example, the connection can be fixed connection, or detachable connection, or integral, which can be understood, or mechanical connection, electrical connection, etc. Of course, it can also be directly connected, or indirectly connected through an intermediate medium, or the internal communication of two elements, or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific implementation situation.

[0106] In the present application, unless otherwise specifically defined or limited, the first feature is "on" or "under" the second feature, which can be direct contact between the first and second features, or indirect contact between the first and second features through an intermediate medium. Moreover, the first feature "above", "above" and "above" the second feature can be directly above or obliquely above the first feature, or only indicate that the horizontal height of the first feature is higher than that of the second feature. The first feature "below", "below" and "below" the second feature can be directly below or obliquely below the first feature, or only indicate that the horizontal height of the first feature is less than that of the second feature.

Claims

1. A real estate cycle prediction method characterized by, The method comprises: obtaining characteristic data affecting real estate cycle rotation; seasonal state determination is performed on the characteristic data to obtain an initial seasonal label and a confidence level of the real estate, wherein the initial seasonal label represents flow cycle information of the real estate; based on regional level information where the real estate is located, the initial seasonal label is corrected to obtain a corrected seasonal label; based on the corrected seasonal label and the confidence level, a real estate cycle conversion probability of the region is obtained; based on the real estate cycle conversion probability, a cycle prediction result of the real estate is obtained.

2. The method of claim 1, wherein, The characteristic data includes transaction volume and price; the seasonal state determination on the characteristic data to obtain the initial seasonal label and the confidence level of the real estate comprises: inputting the transaction volume and the price into a seasonal state machine, determining the transaction volume based on a first preset threshold by the seasonal state machine, determining the price based on a second preset threshold by the seasonal state machine, and outputting the initial seasonal label and the confidence level of the real estate.

3. The method of claim 2, wherein, The initial seasonal label includes spring, summer, autumn, winter and transition state, wherein the spring reflects the state of quantity rising and price stable of the real estate, the summer reflects the state of quantity and price rising of the real estate, the autumn reflects the state of quantity falling and price stable of the real estate, the winter reflects the state of quantity and price falling of the real estate, and the confidence level includes spring confidence, summer confidence, autumn confidence, winter confidence and transition confidence; the determination of the initial seasonal label and the confidence level of the real estate by the seasonal state machine based on the first preset threshold and the second preset threshold comprises: if the transaction volume is greater than or equal to the first preset threshold, and the absolute value of the price is less than or equal to the second preset threshold, it is determined that the initial seasonal label is the spring, and the spring confidence corresponding to the spring is outputted; if the transaction volume is greater than or equal to the first preset threshold, and the price is greater than the second preset threshold, it is determined that the initial seasonal label is the summer, and the summer confidence corresponding to the summer is outputted; if the transaction volume is less than the first preset threshold, and the absolute value of the price is less than or equal to the second preset threshold, it is determined that the initial seasonal label is the autumn, and the autumn confidence corresponding to the autumn is outputted; if the transaction volume is less than the first preset threshold, and the price is less than the third preset threshold, it is determined that the initial seasonal label is the winter, and the winter confidence corresponding to the winter is outputted; if the transaction volume is less than the first preset threshold, and the price is greater than the second preset threshold, it is determined that the initial seasonal label is the transition state, and the transition confidence corresponding to the transition state is outputted.

4. The method of claim 1, wherein, The correction of the initial seasonal label based on the regional level information where the real estate is located to obtain the corrected seasonal label comprises: obtain historical regional phase difference data of the region; classify the region based on the regional level information to obtain regional level classification data; establish a lag mapping relationship between the regional level and the season based on the historical regional phase difference data and the regional level classification data; calculate a season offset based on a season transition rule, the lag mapping relationship, and the initial season label; correct the initial season label based on the season offset to obtain a corrected season label.

5. The method of claim 1, wherein, The notional cycle transition probability of the region is obtained based on the corrected season label and the confidence, including: obtain regional adjustment factors and historical season transition influence matrices that affect the region where the notional property is located; determine an initial season probability vector of the notional property based on the corrected season label and the confidence; weight the initial season probability vector and the historical season transition influence matrix based on the regional adjustment factors to obtain a target season transition influence matrix; perform probability normalization processing on the target season transition influence matrix based on the initial season probability vector to obtain the notional cycle transition probability of the region.

6. The method of claim 5, wherein, The target season transition influence matrix is obtained by weighting the initial season probability vector and the historical season transition influence matrix based on the regional adjustment factors, including: weight the historical season transition influence matrix based on the regional adjustment factors to obtain a regional adjustment weight based on a Bayesian algorithm; weight and adjust the historical season transition influence matrix based on the regional adjustment weight to obtain the target season transition influence matrix.

7. The method of claim 5, wherein, The method further includes: perform probability normalization processing on the target season transition influence matrix to obtain a regional season heat map and a strategy recommendation matrix of the region.

8. A real estate cycle prediction system characterized by, The system includes: an acquisition module configured to acquire feature data affecting notional cycle rotation; a determination module configured to determine a season state of the feature data to obtain an initial season label and a confidence of the notional cycle, wherein the initial season label represents flow cycle information of the notional property; a correction module configured to correct the initial season label based on regional level information of a region where the notional property is located to obtain a corrected season label; a calculation module configured to obtain a notional cycle transition probability of the region based on the corrected season label and the confidence; a prediction module configured to obtain a cycle prediction result of the notional property based on the notional cycle transition probability. 9.An electronic device comprising a memory and a processor, the memory storing a computer program, wherein, The processor executes the computer program to implement the steps of the method of any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1-7.

Images