Intelligent comprehensive rating method for buildings

Abstract

The invention provides an intelligent comprehensive rating method for buildings, and belongs to the technical field of intelligent rating, and the method comprises the steps: constructing a historical score matrix, splitting the historical score matrix of the corresponding historical building according to the development general situation of the position of the building of each historical building, obtaining the development factors of the corresponding historical building, and obtaining the development factors of the corresponding historical building; determining a first planning value of the target building based on each physical factor and a second planning value of the target building based on each location factor according to the constructed information of the target building, and obtaining an initial rating of the target building in combination with a realization capability coefficient of the target building, and mining to-be-planned information of the target building from a network platform, and performing matching analysis with the development general situation of each historical building, determining the development probability of the target building based on different factor types, optimizing the initial rating according to all the development probabilities, obtaining the comprehensive rating of the target building, and outputting and displaying the comprehensive rating. Accurate and dynamic evaluation of the building value is realized, and the investment risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent rating, and in particular to an intelligent comprehensive rating method for real estate projects. Background Art

[0002] With the continuous development and changes of cities, the development status of the location of real estate has an increasingly significant impact on its value. However, the traditional real estate rating method is often relatively simple, and most of them are based on the current static information of the real estate, such as the quality of the house and the status of the surrounding supporting facilities. It is difficult to fully and dynamically reflect the real value changes of the real estate in the future. For example, if there are major plans around the real estate, such as new transportation hubs and commercial centers, the value of surrounding real estate will increase; on the contrary, if negative factors appear in the area, such as intensified environmental pollution problems and industrial decline, it may cause the value of the real estate to fall. However, the existing rating system is difficult to effectively capture the impact of these dynamic factors on the value of real estate. If rating is based solely on information that has already been built, there will often be large deviations, and there is a possibility of increased investment risk due to inaccurate ratings.

[0003] Therefore, the present invention proposes an intelligent comprehensive rating method for real estate projects. Summary of the invention

[0004] The present invention provides an intelligent comprehensive rating method for real estate projects, which is used to achieve accurate and dynamic evaluation of the value of real estate projects and reduce investment risks by considering the historical data of the real estate projects, location and physical factors, liquidity coefficients, development probabilities of different factor types brought about by information to be planned, etc.

[0005] The present invention provides an intelligent comprehensive rating method for real estate projects, comprising: Step 1: Collect historical rating sets of different historical buildings and construct a historical rating matrix. According to the development profile of the location of each historical building, split the historical rating matrix of the corresponding historical building to obtain the development factor of the corresponding historical building, wherein the development factor includes: stability factor, negative factor and positive factor; Step 2: Determine the first planning value based on each physical factor and the second planning value based on each location factor of the target building according to the built information of the target building; Step 3: Obtaining an initial rating of the target building based on the first planning value, the second planning value, and the liquidity coefficient of the target building; Step 4: mining the planned information of the target building from the network platform, and determining the development probability of the target building based on different factor types by matching and analyzing the development profile of each historical building; Step 5: Optimize the initial rating according to all development probabilities, obtain the comprehensive rating of the target property and output it for display.

[0006] Preferably, the historical score set includes: the historical score of each physical factor at different historical time points, the historical score of each location factor and the liquidity coefficient, wherein the physical factors include: the scale of the community, the newness rate of the building, the reputation of property management, the volume ratio, the greening rate and the reputation of the main apartment types; The location factors include: transportation convenience, environmental landscape maturity, educational facilities convenience and external facilities completeness; The liquidity coefficient is related to the market trend coefficient, the judicial auction discount ratio coefficient and the special factor coefficient.

[0007] Preferably, according to the development overview of the location of each historical building, the historical scoring matrix of the corresponding historical building is split, including: Assign a rating time label to each row in the historical rating matrix of the corresponding historical property; The development overview is compared, matched and disassembled according to the scoring time tags to obtain sub-development information between a first historical time point at which each scoring time tag is located and a second historical time point at which a previous scoring time tag is located; Inputting the sub-development information into a development analysis model to obtain a first improvement coefficient based on each physical factor and a second improvement coefficient based on each location factor; According to the first improvement coefficient and the second improvement coefficient, the first scoring vector and the second scoring vector are compared with the scoring difference under the same factor in the historical scoring vectors corresponding to the first historical time point and the second historical time point except for the monetization capability coefficient, and a label of whether to split is assigned to the corresponding first historical time point; The historical scoring matrix is ​​split according to the splitting labels assigned in the historical scoring matrix, wherein the historical scoring matrix is ​​constructed by placing each historical scoring vector in sequence according to the chronological order of the scoring time labels.

[0008] Preferably, assigning a label indicating whether to split to the corresponding first historical time point includes: Analyze the feasibility of development and improvement under the same factors; ;in, It indicates the validity of the development improvement under the jth factor; represents the second score corresponding to the second historical time point under the jth factor; represents the first score corresponding to the first historical time point under the jth factor; represents the improvement coefficient from the second historical time point to the first historical time point under the jth factor; represents the first score of the jth factor in the first evaluation after the completion of the corresponding historical building, where i is 2, 3, ..., n1, and n1 represents the total number of historical scores of the corresponding historical building; Improve the feasibility according to the development of each factor , and combined with the set weight of each factor, the qualification coefficient corresponding to the first scoring vector is obtained; ;in, represents the qualification coefficient corresponding to the first scoring vector; Indicates n1 Satisfied the number of Indicates n1 Satisfied The sum of the set weights of the factors; represents the set weight of the jth factor; Indicates the total number of factors involved; represents the set improvement threshold of the jth factor; If the qualified coefficient is greater than a preset threshold, a split label is assigned to the corresponding first historical time point; Otherwise, a non-split label is assigned to the corresponding first historical time point.

[0009] Preferably, the development factors of the corresponding historical buildings are obtained, including: Calculate the average value of each column vector in each split sub-matrix in sequence according to the time split order, and plot and fit all the average values ​​under the same dimension in time sequence to obtain the fitting coefficient; If the fitting coefficient is within the corresponding first set range, it is determined to be a positive dimension; If the fitting coefficient is within the corresponding second setting range, it is determined to be a stable dimension; If the fitting coefficient is within the corresponding third setting range, it is determined to be a negative dimension; According to the judgment results of each dimension, the development factor of the corresponding historical real estate is obtained.

[0010] Preferably, according to the determination result of each dimension, the development factor of the corresponding historical real estate is obtained, including: Counting respectively a first number M1 of positive dimensions, a second number M2 of stable dimensions, and a third number M3 of negative dimensions among the remaining results excluding the determination result of the dimension corresponding to the monetization capability coefficient; according to Filter out the development type corresponding to the maximum value, among which, represents the sum of the set weights of the factors involved in the first quantity M1; represents the sum of the set weights of the factors involved in the second number M2; represents the sum of the set weights of the factors involved in the third quantity M3; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the first quantity M1; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the second quantity M2; represents the influence coefficient of the determination result of the dimension corresponding to the liquidity conversion coefficient on the factors involved in the third quantity M3; max represents the maximum value symbol; The development type corresponding to the maximum value is regarded as the development factor.

[0011] Preferably, by matching and analyzing the development profile of each historical real estate project, the development probability of the target real estate project based on different factor types is determined, including: Analyze the similarity coefficients of the current overview of the target real estate and the historical overviews of each historical real estate at different historical split time points according to the similarity function, and construct a similarity set; When the variance of the similarity set is less than the variance threshold, the development factor of the corresponding historical building is used as the reference factor of the target building; Otherwise, the development factor of the corresponding historical building is not used as a reference factor for the target building; All reference factors of the target real estate are counted, the number of factors based on each factor type is determined respectively, and the development probability based on different factor types is obtained.

[0012] Preferably, the initial rating is optimized according to all development probabilities to obtain a comprehensive rating of the target property, including: Determine the type weight of each factor type based on the development probability of different factor types; The initial rating is optimized according to the type weight to obtain a comprehensive rating.

[0013] Compared with the prior art, the present invention has the following beneficial effects: By considering the historical data of the property, location and physical factors, liquidity coefficient, and the development probability of different factor types brought about by the information to be planned, we can achieve an accurate and dynamic assessment of the property value and reduce investment risks.

[0014] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0015] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of an intelligent comprehensive rating method for real estate in an embodiment of the present invention; Figure 2 A process diagram of primary rating in an embodiment of the present invention; Figure 3 This is a primary rating result mapping diagram in an embodiment of the present invention; Figure 4 This is an example diagram of primary rating in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0018] The present invention provides an intelligent comprehensive rating method for real estate projects. Figure 1 As shown, including: Step 1: Collect historical rating sets of different historical buildings and construct a historical rating matrix. According to the development profile of the location of each historical building, split the historical rating matrix of the corresponding historical building to obtain the development factor of the corresponding historical building, wherein the development factor includes: stability factor, negative factor and positive factor; Step 2: Determine the first planning value based on each physical factor and the second planning value based on each location factor of the target building according to the built information of the target building; Step 3: Obtaining an initial rating of the target building based on the first planning value, the second planning value, and the liquidity coefficient of the target building; Step 4: mining the planned information of the target building from the network platform, and determining the development probability of the target building based on different factor types by matching and analyzing the development profile of each historical building; Step 5: Optimize the initial rating according to all development probabilities, obtain the comprehensive rating of the target property and output it for display.

[0019] Preferably, the historical score set includes: the historical score of each physical factor at different historical time points, the historical score of each location factor and the liquidity coefficient, wherein the physical factors include: the scale of the community, the newness rate of the building, the reputation of property management, the volume ratio, the greening rate and the reputation of the main apartment types; The location factors include: transportation convenience, environmental landscape maturity, educational facilities convenience and external facilities completeness; The liquidity coefficient is related to the market trend coefficient, the judicial auction discount ratio coefficient and the special factor coefficient.

[0020] In this embodiment, the historical score set is a series of score results based on different factors obtained after the real estate appraisal experts or home buyers have conducted value appraisals on the real estate at different historical time points. For example, a historical real estate has been rated on house quality, property services, surrounding facilities, etc. every year in the past 5 years. The annual scores are summarized to form the historical score set of the real estate, where the score range under each dimension is 0 to 100, and the primary rating involved is implemented based on an evaluation system that integrates the two dimensions of real estate quality and real estate liquidity. The real estate quality measures the comfort of living and the maturity of surrounding living facilities from the perspectives of the real estate's physical factors and location factors, and the real estate liquidity coefficient reflects the degree of value reduction and ease of liquidity of the real estate from the real estate market trend and judicial auction transactions.

[0021] In this embodiment, each historical property corresponds to a historical rating matrix, and each historical rating matrix contains a row vector consisting of the rating results of different rating dimensions at different historical time points, specifically: .

[0022] In this embodiment, the development overview refers to the surrounding development conditions involved in the corresponding historical real estate from the beginning of its establishment to the present time, including but not limited to the development of economy, transportation, environment, policy, etc. In this embodiment, the stability factor, negative factor, and positive factor are analyzed based on the average value of each sub-matrix in each dimension after the matrix is ​​split.

[0023] In this embodiment, the constructed information is input into the planning analysis model to obtain the planning value based on each factor. It should be noted that the planning analysis model is based on the real estate construction information of different historical real estate projects and the scoring results of experts or home buyers on the constructed information under different factors as samples for training the neural network model (CNN model), and the training samples are greater than 1,000 cases. Therefore, the scores under different factors can be directly obtained, which are the planning values.

[0024] In this embodiment, the target building is the building that needs to be rated, and the constructed information refers to the actual information based on the completed building, including: physical aspects such as building structure, apartment design, community greening, as well as location information such as the building's geographical location and surrounding existing supporting facilities.

[0025] In this embodiment, the value range of the monetization capability coefficient is 0 to 1.

[0026] In this embodiment, future planning information of the real estate project is mined from network platforms (government planning websites, real estate information platforms, etc.), including but not limited to traffic planning, commercial planning, public facilities planning, etc. For example, it is necessary to build a key primary school at the location of the target real estate project, which is the information to be planned.

[0027] In this embodiment, the value range of the development probability is 0 to 1.

[0028] In this embodiment, the comprehensive rating is obtained by optimizing the initial rating based on all development probabilities.

[0029] In this embodiment, the specific rating process for the initial rating is: Building quality score = building location factor score × regional weight + building physical factor score × physical weight; Liquidity coefficient = (market trend coefficient × market weight + judicial auction discount ratio coefficient × judicial auction weight) × special factor coefficient; Comprehensive property score = property quality score × liquidity coefficient.

[0030] Rating mapping is performed based on the comprehensive rating of the real estate to obtain the initial rating, such as Figures 2 to 4 shown.

[0031] In this embodiment, the matching analysis is achieved by performing matching analysis on the information based on a similarity function, and the value range is 0 to 1.

[0032] The beneficial effect of the above technical solution is: by considering the historical data of the real estate, location and physical factors, liquidity coefficient, development probability of different factor types brought by the information to be planned, etc., it is possible to achieve accurate and dynamic evaluation of the value of the real estate and reduce investment risks.

[0033] The present invention provides an intelligent comprehensive rating method for real estate projects, which splits the historical rating matrix of the corresponding historical real estate projects according to the development overview of the location of each historical real estate project, including: Assign a rating time label to each row in the historical rating matrix of the corresponding historical property; The development overview is compared, matched and disassembled according to the scoring time tags to obtain sub-development information between a first historical time point at which each scoring time tag is located and a second historical time point at which a previous scoring time tag is located; Inputting the sub-development information into a development analysis model to obtain a first improvement coefficient based on each physical factor and a second improvement coefficient based on each location factor; According to the first improvement coefficient and the second improvement coefficient, the first scoring vector and the second scoring vector are compared with the scoring difference under the same factor in the historical scoring vectors corresponding to the first historical time point and the second historical time point except for the monetization capability coefficient, and a label of whether to split is assigned to the corresponding first historical time point; The historical scoring matrix is ​​split according to the splitting labels assigned in the historical scoring matrix, wherein the historical scoring matrix is ​​constructed by placing each historical scoring vector in sequence according to the chronological order of the scoring time labels.

[0034] In this embodiment, the scoring time label is used to mark the time of the corresponding row vector, so as to facilitate the subsequent analysis of the sub-development information at two adjacent time points.

[0035] In this embodiment, the development analysis model is obtained by training the CNN model with the improvement coefficients for each physical factor and location factor brought about by different development information as samples. Therefore, the improvement coefficient for each factor under the sub-development information can be directly obtained, and the value range of the improvement coefficient is 0 to 0.5.

[0036] In this embodiment, the comparison and matching decomposition is obtained by directly intercepting the information of the development overview according to the first historical time point and the second historical time point as the boundary time point.

[0037] In this embodiment, the first scoring vector={the remaining scoring result of the historical scoring vector at the first historical time point except the monetization capability coefficient}; The second scoring vector = {the remaining scoring result of the historical scoring vector at the second historical time point except the monetization ability coefficient}; And the second historical time point is before the first historical time point.

[0038] In this embodiment, labels are assigned in order to reasonably split the matrix. For example, historical time point 3 is assigned a split label, and historical time points 1 and 2 are assigned non-split labels. At this time, the row vectors at historical time point 1 and historical time point 2 are grouped together as a submatrix.

[0039] The beneficial effect of the above technical solution is: by assigning a scoring time label to each row vector in the matrix, it is convenient to subsequently match and obtain the sub-development information between adjacent time points from the development profile, and analyze the sub-development information based on the development analysis model to obtain the improvement coefficient based on each factor, and then through the comparative analysis of the difference between the improvement coefficients and the difference between the scoring vectors, it is convenient to accurately determine whether to set the split label, to ensure the local analysis brought about by the split, and to provide a reliable basis for the subsequent determination of the development factors.

[0040] The present invention provides an intelligent comprehensive rating method for real estate, which assigns a label of whether to split to the corresponding first historical time point, including: Analyze the feasibility of development and improvement under the same factors; ;in, It indicates the validity of the development improvement under the jth factor; represents the second score corresponding to the second historical time point under the jth factor; represents the first score corresponding to the first historical time point under the jth factor; represents the improvement coefficient from the second historical time point to the first historical time point under the jth factor; represents the first score of the jth factor in the first evaluation after the completion of the corresponding historical building, where i is 2, 3, ..., n1, and n1 represents the total number of historical scores of the corresponding historical building; Improve the feasibility according to the development of each factor , and combined with the set weight of each factor, the qualification coefficient corresponding to the first scoring vector is obtained; ;in, represents the qualification coefficient corresponding to the first scoring vector; Indicates n1 Satisfied the number of Indicates n1 Satisfied The sum of the set weights of the factors; represents the set weight of the jth factor; Indicates the total number of factors involved; represents the set improvement threshold of the jth factor; If the qualified coefficient is greater than a preset threshold, a split label is assigned to the corresponding first historical time point; Otherwise, a non-split label is assigned to the corresponding first historical time point.

[0041] In this embodiment, the set weight of each factor in the process of calculating the qualified coefficient is pre-set, and the factors at this time only include location factors and physical factors, and the sum of the set weights of all factors is 1. Specifically: the set weights of community scale, newness rate of real estate, property management reputation, floor area ratio, greening rate, and main apartment type reputation are: 0.08, 0.08, 0.08, 0.06, 0.07, and 0.08 respectively.

[0042] The set weights for transportation convenience, environmental landscape maturity, educational facilities convenience, and external facilities completeness are 0.15, 0.1, 0.15, and 0.15 respectively.

[0043] In this embodiment, the preset threshold is 0.3.

[0044] In this embodiment, the set improvement thresholds under different factors are different. The set improvement thresholds for community scale, building newness rate, property management popularity, floor area ratio, greening rate, and main apartment type popularity are 0.01, 0.03, 0.1, 0.02, 0.05, and 0.01, respectively.

[0045] The set improvement thresholds for transportation convenience, environmental landscape maturity, educational facilities convenience, and external facilities completeness are 0.08, 0.08, 0.08, and 0.08 respectively.

[0046] The beneficial effect of the above technical solution is: by comparing and analyzing the scores of different factors set at two adjacent historical times, the feasibility of development improvements under different circumstances is calculated, and then combined with the set weights, the qualified coefficient is calculated to compare with the preset threshold to achieve the reasonable assignment of labels.

[0047] The present invention provides an intelligent comprehensive rating method for real estate to obtain the development factor of the corresponding historical real estate, including: Calculate the average value of each column vector in each split sub-matrix in sequence according to the time split order, and plot and fit all the average values ​​under the same dimension in time sequence to obtain the fitting coefficient; If the fitting coefficient is within the corresponding first set range, it is determined to be a positive dimension; If the fitting coefficient is within the corresponding second setting range, it is determined to be a stable dimension; If the fitting coefficient is within the corresponding third setting range, it is determined to be a negative dimension; According to the judgment results of each dimension, the development factor of the corresponding historical real estate is obtained.

[0048] In this embodiment, the split sub-matrix is ​​obtained by splitting the historical rating matrix based on the assigned split labels, which includes the monetization ability coefficient of the real estate.

[0049] In this embodiment, the average value is obtained by averaging each column vector in the sub-matrix.

[0050] In this embodiment, the same dimension refers to the elements of the column vector under the corresponding dimension in the split sub-matrix, which is convenient for curve drawing and fitting of all average values ​​involved in the same dimension in each sub-matrix after the historical rating matrix is ​​split. The fitting is implemented based on excl, mainly to obtain the linear fitting coefficient, and the linear fitting coefficient result can be directly obtained.

[0051] In this embodiment, for example, the average values ​​obtained for traffic convenience are 80, 82, 81, and 83, respectively. At this time, the obtained linear fitting coefficient is 0.8.

[0052] In this embodiment, the first setting range is: greater than or equal to 0.5; The second setting range is: greater than or equal to -0.2 and less than 0.5; The third setting range is: less than -0.2.

[0053] In this embodiment, the type of the corresponding dimension can be obtained by comparing the fitting coefficient with the range.

[0054] The beneficial effect of the above technical solution is: by calculating the average value of each dimension in each sub-matrix, curve drawing and fitting are performed to obtain the fitting coefficient, and further through range comparison, the judgment type is directly locked, providing a basis for determining the development factor.

[0055] The present invention provides an intelligent comprehensive rating method for real estate projects, and obtains the development factor of the corresponding historical real estate projects according to the determination results of each dimension, including: Counting respectively a first number M1 of positive dimensions, a second number M2 of stable dimensions, and a third number M3 of negative dimensions among the remaining results excluding the determination result of the dimension corresponding to the monetization capability coefficient; according to Filter out the development type corresponding to the maximum value, among which, represents the sum of the set weights of the factors involved in the first quantity M1; represents the sum of the set weights of the factors involved in the second number M2; represents the sum of the set weights of the factors involved in the third quantity M3; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the first quantity M1; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the second quantity M2; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the third quantity M3; max represents the maximum value symbol; The development type corresponding to the maximum value is regarded as the development factor.

[0056] In this embodiment, the determination results under the factor combination-liquidity coefficient-impact comparison table are respectively obtained for The comparison table contains the influence coefficient of different factor combinations and the judgment results of the monetization ability coefficient on different factor combinations, and the value range is 0 to 1. For example, the first quantity M1 corresponds to factor 1, factor 4, and factor 5, and the judgment result of the monetization ability coefficient is a negative type. At this time, the influence coefficient of the combination of factors 1, factor 4, and factor 5 is 0.1.

[0057] The beneficial effect of the above technical solution is: by quantitatively counting the remaining judgment results except the monetization ability coefficient, and combining the weight and the influence coefficient of the judgment result of the monetization ability coefficient on the factor combination, the type corresponding to the maximum value is comprehensively screened as a development factor to ensure the reliability of the rating and make it more realistic.

[0058] The present invention provides an intelligent comprehensive rating method for real estate projects, which determines the development probability of a target real estate project based on different factor types by matching and analyzing the development profile of each historical real estate project, including: Analyze the similarity coefficients of the current overview of the target real estate and the historical overviews of each historical real estate at different historical split time points according to the similarity function, and construct a similarity set; When the variance of the similarity set is less than the variance threshold, the development factor of the corresponding historical building is used as the reference factor of the target building; Otherwise, the development factor of the corresponding historical building is not used as a reference factor for the target building; All reference factors of the target real estate are counted, the number of factors based on each factor type is determined respectively, and the development probability based on different factor types is obtained.

[0059] In this embodiment, the similarity coefficient = sim (current profile, historical profile at the corresponding historical split time point). It should be noted that the current profile and the historical profile at the corresponding historical split time point both obtain all information related to the relevant real estate at the corresponding time point, that is, each profile obtains all the information about the real estate.

[0060] In this embodiment, the value range of the similarity coefficient is 0 to 1.

[0061] In this embodiment, similarity set={similarity coefficient between the current profile and the historical profile at each historical split time point}.

[0062] In this embodiment, the variance is calculated based on the variance formula, which belongs to the prior art.

[0063] In this embodiment, the variance threshold is 0.05.

[0064] In this embodiment, the reference factor is a development factor that satisfies the variance of the similarity set to be less than the variance threshold.

[0065] In this embodiment, the factor type is the development type mentioned above, for example, factor 1 is a negative type, factor 2 is a positive type, and factor 3 is a positive type. At this time, the development probability under the positive type = 2 / 3.

[0066] The probability of development under the negative type = 1 / 3.

[0067] The beneficial effect of the above technical solution is: through the similarity function, the similarity comparison analysis of the current profile and the historical profile at different historical i split time points is carried out to construct a similarity set, and by comparing with the variance threshold, it is ensured that the reference factor has acquisition value, providing a basis for determining the development probability.

[0068] The present invention provides an intelligent comprehensive rating method for real estate projects, which optimizes the initial rating according to all development probabilities to obtain a comprehensive rating of the target real estate project, including: Determine the type weight of each factor type based on the development probability of different factor types; The initial rating is optimized according to the type weight to obtain a comprehensive rating.

[0069] In this embodiment, type weight=development probability of corresponding factor type / sum of development probabilities of all factor types.

[0070] If the positive type weight > stable type weight + negative type weight, then the initial rating of the initial rating × (1 + positive type weight) will be remapped to the interval ranking to obtain the corresponding comprehensive rating result, that is, Figure 3 The grade results mentioned in .

[0071] In this embodiment, if the negative type weight > stable type weight + positive type weight, then the initial rating of the initial rating × (1-negative type weight) is used to remap the obtained result to the interval ranking to obtain the corresponding comprehensive rating result, that is, Figure 3 The grade results mentioned in .

[0072] Otherwise, keep treating the initial rating as the composite rating.

[0073] The beneficial effect of the above technical solution is: by developing probabilities to determine the type weights of different factor types, the initial rating is then optimized to ensure the accuracy and reliability of the rating, which is conducive to reducing investment risks.

[0074] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent comprehensive rating method for real estate, characterized in that: include: Step 1: Collect historical rating sets of different historical buildings and construct a historical rating matrix. According to the development profile of the location of each historical building, split the historical rating matrix of the corresponding historical building to obtain the development factor of the corresponding historical building, wherein the development factor includes: stability factor, negative factor and positive factor; Step 2: Determine the first planning value based on each physical factor and the second planning value based on each location factor of the target building according to the built information of the target building; Step 3: Obtaining an initial rating of the target building based on the first planning value, the second planning value, and the liquidity coefficient of the target building; Step 4: mining the planned information of the target building from the network platform, and determining the development probability of the target building based on different factor types by matching and analyzing the development profile of each historical building; Step 5: Optimize the initial rating according to all development probabilities, obtain the comprehensive rating of the target property and output it for display.

2. The intelligent comprehensive rating method for real estate according to claim 1, characterized in that: The historical score set includes: the historical score of each physical factor at different historical time points, the historical score of each location factor and the liquidity coefficient, wherein the physical factors include: community scale, building newness rate, property management reputation, volume ratio, greening rate and main apartment type reputation; The location factors include: transportation convenience, environmental landscape maturity, educational facilities convenience and external facilities completeness; The liquidity coefficient is related to the market trend coefficient, the judicial auction discount ratio coefficient and the special factor coefficient.

3. The intelligent comprehensive rating method for real estate according to claim 1 is characterized in that: According to the development overview of the location of each historical building, the historical scoring matrix of the corresponding historical building is split, including: Assign a rating time label to each row in the historical rating matrix of the corresponding historical property; The development overview is compared, matched and disassembled according to the scoring time tags to obtain sub-development information between a first historical time point at which each scoring time tag is located and a second historical time point at which a previous scoring time tag is located; Inputting the sub-development information into a development analysis model to obtain a first improvement coefficient based on each physical factor and a second improvement coefficient based on each location factor; According to the first improvement coefficient and the second improvement coefficient, the first scoring vector and the second scoring vector are compared with the scoring difference under the same factor in the historical scoring vectors corresponding to the first historical time point and the second historical time point except for the monetization capability coefficient, and a label of whether to split is assigned to the corresponding first historical time point; The historical scoring matrix is ​​split according to the splitting labels assigned in the historical scoring matrix, wherein the historical scoring matrix is ​​constructed by placing each historical scoring vector in sequence according to the chronological order of the scoring time labels.

4. The intelligent comprehensive rating method for real estate according to claim 3 is characterized in that: Assigning a label indicating whether to split the corresponding first historical time point includes: Analyze the feasibility of development and improvement under the same factors; ;in, It indicates the validity of the development improvement under the jth factor; represents the second score corresponding to the second historical time point under the jth factor; represents the first score corresponding to the first historical time point under the jth factor; represents the improvement coefficient from the second historical time point to the first historical time point under the jth factor; represents the first score of the jth factor in the first evaluation after the completion of the historical building, where i is 2, 3, ..., n1, and n1 represents the total number of historical scores of the corresponding historical building; improve the validity according to the development of each factor , and combined with the set weight of each factor, the qualification coefficient corresponding to the first scoring vector is obtained; ;in, represents the qualification coefficient corresponding to the first scoring vector; Indicates n1 Satisfied the number of Indicates n1 Satisfied The sum of the set weights of the factors; represents the set weight of the jth factor; Indicates the total number of factors involved; represents the set improvement threshold of the jth factor; If the qualified coefficient is greater than a preset threshold, a split label is assigned to the corresponding first historical time point; Otherwise, a non-split label is assigned to the corresponding first historical time point.

5. The intelligent comprehensive rating method for real estate according to claim 1 is characterized in that: Get the development factors of the corresponding historical buildings, including: Calculate the average value of each column vector in each split sub-matrix in sequence according to the time split order, and plot and fit all the average values ​​under the same dimension in time sequence to obtain the fitting coefficient; If the fitting coefficient is within the corresponding first set range, it is determined to be a positive dimension; If the fitting coefficient is within the corresponding second setting range, it is determined to be a stable dimension; If the fitting coefficient is within the corresponding third setting range, it is determined to be a negative dimension; According to the judgment results of each dimension, the development factor of the corresponding historical real estate is obtained.

6. The intelligent comprehensive rating method for real estate according to claim 5 is characterized in that: According to the judgment results of each dimension, the development factors of the corresponding historical real estate are obtained, including: Counting respectively a first number M1 of positive dimensions, a second number M2 of stable dimensions, and a third number M3 of negative dimensions among the remaining results excluding the determination result of the dimension corresponding to the monetization capability coefficient; according to Filter out the development type corresponding to the maximum value, among which, represents the sum of the set weights of the factors involved in the first quantity M1; represents the sum of the set weights of the factors involved in the second number M2; represents the sum of the set weights of the factors involved in the third quantity M3; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the first quantity M1; represents the influence coefficient of the determination result of the dimension corresponding to the monetization capability coefficient on the factors involved in the second quantity M2; represents the influence coefficient of the determination result of the dimension corresponding to the liquidity conversion coefficient on the factors involved in the third quantity M3; max represents the maximum value symbol; The development type corresponding to the maximum value is regarded as the development factor.

7. The intelligent comprehensive rating method for real estate according to claim 1, characterized in that: By matching and analyzing the development profile of each historical property, the development probability of the target property based on different factor types is determined, including: Analyze the similarity coefficients of the current overview of the target real estate and the historical overviews of each historical real estate at different historical split time points according to the similarity function, and construct a similarity set; When the variance of the similarity set is less than the variance threshold, the development factor of the corresponding historical building is used as the reference factor of the target building; Otherwise, the development factor of the corresponding historical building is not used as a reference factor for the target building; All reference factors of the target real estate are counted, the number of factors based on each factor type is determined respectively, and the development probability based on different factor types is obtained.

8. The intelligent comprehensive rating method for real estate according to claim 1 is characterized in that: The initial rating is optimized according to all development probabilities to obtain a comprehensive rating of the target property, including: Determine the type weight of each factor type based on the development probability of different factor types; The initial rating is optimized according to the type weight to obtain a comprehensive rating.

Images