Third-party stock house value evaluation method and system

Abstract

The invention relates to the technical field of intelligent evaluation, and particularly discloses a third-party stock house value evaluation method and system, and the method comprises the steps: obtaining the position of a transacted and authorized stock house in a preset time range, querying the facility information in a preset space range based on the stock house, and carrying out the evaluation of the value of the third-party stock house. Determining a location index of the stock house according to the facility information; obtaining a house type image, decoration information and check-in information of the stock house, and determining individual indexes according to the house type image, the decoration information and the check-in information; and querying the transaction price of the stock house, and training an evaluation function of each index by taking the transaction price as a true value to obtain a value evaluation model. According to the method, the location index and the individual index of the stock house are obtained, the index matrix is obtained, the price of the stock house is inquired, the index matrix is used as a feature, the price is used as a label to train the value evaluation model, and during actual transaction, the predicted price is provided by means of the value evaluation model, reference is provided for both parties, and the transaction process is promoted.

Description

Technical Field

[0001] The present invention relates to the field of intelligent evaluation technology, and in particular to a third-party existing housing value evaluation method and system. Background Art

[0002] Second-hand housing transactions involve the sale and purchase of properties that have been purchased or used by others. It is a common type of real estate transaction. The pricing of second-hand housing is mainly set by the owners themselves. Since owners have feelings for their own houses, the pricing is generally higher than the actual price, or even much higher. In the eyes of buyers, second-hand houses are similar to regular second-hand items and their prices should be lower. This conflict in concepts actually hinders the second-hand housing transaction process. If a third-party value assessment model can be provided to provide both parties with a more accurate reference based on big data, then the second-hand housing transaction process can be effectively promoted. Summary of the invention

[0003] The purpose of the present invention is to provide a third-party existing housing value assessment method and system to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A third-party existing housing value assessment method, the method comprising:

[0006] Obtain the location of existing houses that have been traded and authorized within a preset time range, query facility information within a preset space range based on the existing houses, and determine the location index of the existing houses according to the facility information;

[0007] Obtaining floor plans, decoration information, and occupancy information of existing houses, and determining individual indicators based on the floor plans, decoration information, and occupancy information;

[0008] Query the transaction price of existing houses, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; in the process of determining the location indicator and individual indicator, simultaneously determine the correlation with the price;

[0009] An information acquisition template is generated based on the determined location indicators and individual indicators, various indicators of the house to be traded are obtained based on the information acquisition template, the trained value assessment model is input, and the predicted price is output as the assessment result.

[0010] As a further solution of the present invention: the step of obtaining the locations of the existing houses that have been traded and authorized within a preset time range, querying the facility information within a preset space range based on the existing houses, and determining the location index of the existing houses according to the facility information includes:

[0011] Receive the time period selected by the user, take the current date as the right endpoint, and determine the time range based on the time period;

[0012] Obtain the location of existing houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating;

[0013] Determine the indicator type based on the facility type, and determine the indicator value based on the facility location and facility rating;

[0014] Take the indicator type as the column and the number of the existing house as the row to count all the indicator values ​​and get the location indicator of the existing house;

[0015] The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include:

[0016] Z ij =αw j L j / d ij ; In the formula, Z ij is the index value corresponding to the jth index type of the ith stock house, α is the preset correction coefficient, and w j is the weight determined according to the facility type corresponding to the jth indicator type. The weight has a sign. When the sign is positive, it indicates a positive correlation with the transaction price. When the sign is negative, it indicates a negative correlation with the transaction price. L j is the facility rating of the facility type corresponding to the jth indicator type, d ij is the distance between the i-th stock housing and the facility type corresponding to the j-th indicator type.

[0017] As a further solution of the present invention: the step of obtaining the floor plan, decoration information and occupancy information of the existing housing, and determining the individual indicators according to the floor plan, decoration information and occupancy information comprises:

[0018] Obtain the floor plans of existing houses, input the floor plans into the trained floor plan evaluation model, and obtain the evaluation score;

[0019] Obtaining a visual image of the existing house containing relative positions, identifying the visual image, and obtaining the degree of decoration completeness with relative positions as labels;

[0020] Obtaining the occupancy information of the existing houses, and correcting all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay;

[0021] The evaluation points and the degree of decoration completeness at each relative position are statistically analyzed to obtain individual indicators.

[0022] As a further solution of the present invention: the step of querying the transaction price of the existing housing, taking the transaction price as the true value, training the evaluation function of each indicator, and obtaining the value evaluation model includes:

[0023] Counting location indicators and individual indicators, and reading the correlation between each indicator and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation;

[0024] Calculating the weight of each indicator, and selecting an evaluation function for each indicator according to the weight and the correlation;

[0025] Query the transaction price of existing users as the true value, train the evaluation function corresponding to each indicator, and calculate the error;

[0026] When the error is less than a preset error threshold, an evaluation function is output to obtain a value evaluation model.

[0027] As a further solution of the present invention: the step of calculating the weight of each indicator and selecting an evaluation function for each indicator according to the weight and the correlation comprises:

[0028] Read the variance of all indicator values ​​for each indicator, and determine the weight according to the inverse ratio of the variance;

[0029] A power function is selected as a reference function, the sign of the coefficient is determined according to the correlation, and the exponent is determined according to the weight to obtain an evaluation function;

[0030] Among them, the coefficients serve as target parameters in the training process.

[0031] As a further solution of the present invention: the steps of generating an information acquisition template based on the determined location index and individual index, acquiring various indexes of the house to be traded based on the information acquisition template, inputting a trained value assessment model, and outputting a predicted price as the assessment result include:

[0032] Generate information acquisition template based on determined location indicators and individual indicators;

[0033] When receiving an evaluation request uploaded by a user, a permission acquisition request is sent to the user;

[0034] When receiving the permission granted by the user, sending the information acquisition template to the user, and acquiring various indicators of the house to be traded based on the information acquisition template;

[0035] The obtained indicators are input into the trained value assessment model, and the predicted price is output as the assessment result.

[0036] The technical solution of the present invention also provides a third-party existing housing value assessment system, the system comprising:

[0037] A location index determination module is used to obtain the location of the existing houses that have been traded and authorized within a preset time range, query the facility information within a preset space range based on the existing houses, and determine the location index of the existing houses according to the facility information;

[0038] An individual indicator determination module is used to obtain the floor plan, decoration information and occupancy information of the existing housing, and determine the individual indicator according to the floor plan, decoration information and occupancy information;

[0039] The evaluation model training module is used to query the transaction price of existing houses, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; in the process of determining the location indicator and individual indicator, the correlation with the price is determined simultaneously;

[0040] The evaluation model application module is used to generate an information acquisition template based on the determined location indicators and individual indicators, obtain various indicators of the house to be traded based on the information acquisition template, input the trained value evaluation model, and output the predicted price as the evaluation result.

[0041] As a further solution of the present invention: the location index determination module includes:

[0042] A time range determination unit, configured to receive a time period selected by a user, take the current date as the right endpoint, and determine a time range based on the time period;

[0043] A facility information query unit, used to obtain the location of the stock houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating;

[0044] An indicator value determination unit, used to determine the indicator type according to the facility type, and determine the indicator value according to the facility location and facility rating;

[0045] The indicator value statistical unit is used to count all indicator values ​​by taking the indicator type as a column and the number of the existing house as a row to obtain the location indicator of the existing house;

[0046] The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include:

[0047] Z ij =αw j L j / d ij ; In the formula, Z ij is the index value corresponding to the jth index type of the ith stock house, α is the preset correction coefficient, and w jis the weight determined according to the facility type corresponding to the jth indicator type. The weight has a sign. When the sign is positive, it indicates a positive correlation with the transaction price. When the sign is negative, it indicates a negative correlation with the transaction price. L j is the facility rating of the facility type corresponding to the jth indicator type, d ij is the distance between the i-th stock housing and the facility type corresponding to the j-th indicator type.

[0048] As a further solution of the present invention: the individual indicator determination module includes:

[0049] An evaluation score calculation unit is used to obtain the floor plan of the existing house, input the floor plan into the trained floor plan evaluation model, and obtain the evaluation score;

[0050] An image recognition unit is used to obtain a visual image containing relative positions of the existing houses, recognize the visual image, and obtain the degree of decoration completeness with relative positions as labels;

[0051] A completeness correction unit is used to obtain the occupancy information of the existing houses and correct all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay;

[0052] The statistical unit is used to calculate the evaluation points and the degree of decoration completeness at each relative position to obtain individual indicators.

[0053] As a further solution of the present invention: the evaluation model training module includes:

[0054] A correlation reading unit is used to count the location index and the individual index, and read the correlation between each index and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation;

[0055] A weight calculation unit, used to calculate the weight of each indicator, and select an evaluation function for each indicator according to the weight and the correlation;

[0056] The error calculation unit is used to query the transaction price of existing users, take it as the true value, train the evaluation function corresponding to each indicator, and calculate the error;

[0057] The model output unit is used to output an evaluation function to obtain a value evaluation model when the error is less than a preset error threshold.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention obtains the location index and individual index of the existing housing, obtains the index matrix, queries the price of the existing housing, uses the index matrix as a feature, and uses the price as a label to train the value assessment model. The more samples of existing housing obtained, the higher the accuracy of the value assessment model. In actual transactions, the value assessment model can be used to provide a price predicted by a third party, which provides a reference for both parties and promotes the transaction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0060] Figure 1 This is a flowchart of the third-party existing housing value assessment method.

[0061] Figure 2 This is the structural block diagram of the third-party existing housing value assessment system. DETAILED DESCRIPTION

[0062] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0063] Figure 1 The flowchart of the third-party existing housing value assessment method is as follows. In an embodiment of the present invention, a third-party existing housing value assessment method is provided, and the method includes:

[0064] Step S100: obtaining the locations of existing houses that have been traded and authorized within a preset time range, querying facility information within a preset space range based on the existing houses, and determining the location index of the existing houses according to the facility information;

[0065] The transaction process of existing housing is itself a buying and selling relationship. The buying and selling price and the subject of the transaction need to be granted permission by both parties before they can be obtained. In the technical solution of the present invention, it is assumed that the permission granted by the user has been obtained. On the basis of having the permission, the location of the existing housing that has been traded within a preset time range is obtained. With the help of the existing map App, the facility information within the preset space range can be queried, and the location index of the existing housing is determined according to the facility information; among which, the type of location index is pre-set by the staff, such as the distance to schools, hospitals and shopping malls, or the distance to highways, elevated roads and production workshops. These types are collectively referred to as location indicators.

[0066] Step S200: Obtaining the floor plan, decoration information and occupancy information of the existing housing, and determining individual indicators according to the floor plan, decoration information and occupancy information;

[0067] For existing housing, in addition to locational advantages, individual advantages are also included. The prices of existing housing with different house types, different decorations and different occupancy conditions are also different. The floor plans, decoration information and occupancy information of existing housing are obtained and counted as individual indicators.

[0068] After the location indicators and individual indicators are determined, the unique label of the existing house (address is sufficient) is used as the row and each indicator is used as the column to obtain an overall matrix to represent the characteristics of each existing house under different indicators.

[0069] Step S300: query the transaction price of the existing housing, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; wherein, in the process of determining the location indicator and the individual indicator, the correlation with the price is simultaneously determined;

[0070] For each indicator, a benchmark function with unknown parameters is determined, and then the sum is performed to obtain an evaluation function with unknown parameters that includes all indicators, which is called a value assessment model. Then, the transaction price of existing housing is obtained based on the granted authority, and the transaction price is used as the true value to determine the error of the predicted value of the value assessment model. The unknown parameters are then continuously adjusted, and the gradient descent method is used for training, so that the predicted value of the value assessment model becomes closer to the true value.

[0071] Step S400: Generate an information acquisition template based on the determined location index and individual index, obtain various indexes of the house to be traded based on the information acquisition template, input the trained value assessment model, and output the predicted price as the assessment result;

[0072] For the existing houses that need to be evaluated, an information acquisition template is prepared based on the determined location indicators and individual indicators, and various indicators of the houses to be traded are obtained based on the information acquisition template. After obtaining various indicators, an array is obtained, and the array is input into the trained value assessment model. The predicted value obtained is used as the predicted price as the evaluation result; wherein, the premise for executing step S400 is that the accuracy of the value assessment model is high enough.

[0073] Regarding step S100: the steps of obtaining the locations of existing houses that have been traded and authorized within a preset time range, querying facility information within a preset space range based on the existing houses, and determining the location index of the existing houses according to the facility information include:

[0074] Receive the time period selected by the user, take the current date as the right endpoint, and determine the time range based on the time period;

[0075] Obtain the location of existing houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating;

[0076] Determine the indicator type based on the facility type, and determine the indicator value based on the facility location and facility rating;

[0077] Take the indicator type as the column and the number of existing houses as the row to count all the indicator values ​​and get the location index of the existing houses.

[0078] In an example of the technical solution of the present invention, a time period input by a user is received, which is generally one quarter. A time range is determined with the current date as the right endpoint, i.e., one quarter ago. The locations of existing houses that have been traded and granted permissions within the preset time range are obtained, and with the help of existing map services, facility information near the location of the existing houses is queried. The concept of nearby is represented by a spatial range, which may be within a range of five kilometers. The facility information includes facility type, facility location, and facility rating. The indicator type is determined according to the facility type, and the indicator value is determined according to the facility location and the facility rating. Then, the indicator type is used as a column and the number of the existing house is used as a row to count all indicator values ​​to obtain the location indicator of the existing house.

[0079] The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include:

[0080] Z ij =αw j L j / d ij ; In the formula, Z ij is the index value corresponding to the jth index type of the ith stock house, α is the preset correction coefficient, and w j is the weight determined according to the facility type corresponding to the jth indicator type. The weight has a sign. When the sign is positive, it indicates a positive correlation with the transaction price. When the sign is negative, it indicates a negative correlation with the transaction price. L j is the facility rating of the facility type corresponding to the jth indicator type, d ij is the distance between the i-th stock housing and the facility type corresponding to the j-th indicator type.

[0081] The process of determining the indicator value is very simple. The distance between the existing housing and the facility is calculated based on the location of the facility. Then, the calculated indicator value is inversely proportional to the distance and directly proportional to the facility rating. The facility rating is determined by the scale parameter and conventional assessment parameters can be used. For example, for a hospital, Grade 3A is the highest rating. Multiply it by the corresponding benchmark level of the hospital to get a value as the facility rating.

[0082] Regarding step S200: the step of obtaining the floor plan, decoration information and occupancy information of the existing housing, and determining the individual indicators according to the floor plan, decoration information and occupancy information comprises:

[0083] Obtain the floor plans of existing houses, input the floor plans into the trained floor plan evaluation model, and obtain the evaluation score;

[0084] Obtaining a visual image of the existing house containing relative positions, identifying the visual image, and obtaining the degree of decoration completeness with relative positions as labels;

[0085] Obtaining the occupancy information of the existing houses, and correcting all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay;

[0086] The evaluation points and the degree of decoration completeness at each relative position are statistically analyzed to obtain individual indicators.

[0087] The above content specifically limits the process of determining individual indicators. The floor plan of the existing house is obtained, and the floor plan is input into a trained floor plan evaluation model to obtain a numerical value used to evaluate the quality of the floor plan, which is called an evaluation score. The floor plan evaluation model is software based on preset rules. For example, the evaluation score is determined based on the length-to-width ratio of each room and the position relationship of each room. The setting process is not complicated. For the technical solution of the present invention, it is regarded as prior art and will not be repeated here.

[0088] Then, a visual image containing relative positions of the existing houses is obtained. The relative positions indicate which position in the house the visual image corresponds to, including the ground, ceiling and wall. The relative positions refer to the relative positions in the floor plan. After the visual image is recognized, the degree of decoration completeness can be determined. The degree of decoration completeness can be determined by contour recognition technology to calculate the number of contours, contour shapes and contour color values. Generally, the number of contours and contour shapes can be used to determine the degree of decoration completeness. The fewer the number of contours and the closer the contour shape is to a rectangle, the higher the degree of decoration completeness.

[0089] Finally, obtain the occupancy information of the existing houses, which includes who has lived there and how long they have lived there. Adjust the decoration completeness based on the occupancy information by calculating the sum of the years of service of the occupants and amplifying the decoration completeness based on the sum of the years. The practical significance of this is that when the sum of the years is higher, the decoration completeness is more affected by human factors and less by its own quality. Therefore, the higher the years, the more obvious the amplification of the decoration completeness.

[0090] Regarding step S300: the step of querying the transaction price of the existing housing, taking the transaction price as the true value, training the evaluation function of each indicator, and obtaining the value evaluation model includes:

[0091] Counting location indicators and individual indicators, and reading the correlation between each indicator and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation;

[0092] Calculating the weight of each indicator, and selecting an evaluation function for each indicator according to the weight and the correlation;

[0093] Query the transaction price of existing users as the true value, train the evaluation function corresponding to each indicator, and calculate the error;

[0094] When the error is less than a preset error threshold, an evaluation function is output to obtain a value evaluation model.

[0095] Positive correlation means that the larger the value of the indicator is, the higher the transaction price is; negative correlation means that the larger the value of the indicator is, the lower the transaction price is. Whether each indicator is positively correlated or negatively correlated with the transaction price can be determined in advance by the staff and stored in the correlation table, which can be directly read when actually used.

[0096] After knowing the correlation between each indicator and the transaction price, calculate the weight of each indicator, select an evaluation function for each indicator based on the weight and the correlation, sum the evaluation functions of each indicator to obtain the final evaluation function, and use the transaction price as the true value to verify the error of the final evaluation function. When the error is small enough, obtain the value evaluation model.

[0097] The step of calculating the weight of each indicator and selecting an evaluation function for each indicator according to the weight and the correlation comprises:

[0098] Read the variance of all indicator values ​​for each indicator, and determine the weight according to the inverse ratio of the variance;

[0099] A power function is selected as a reference function, the sign of the coefficient is determined according to the correlation, and the exponent is determined according to the weight to obtain an evaluation function;

[0100] Among them, the coefficients serve as target parameters in the training process.

[0101] The above content describes the process of determining weights in detail. For any indicator, the variance of all its values ​​is calculated as the weight. The larger the variance, the greater the fluctuation of the indicator. In this case, the weight is larger. Then, the power function y=Ax is selected. n As the benchmark function, A is the coefficient, n is the exponent, the sign of the coefficient is determined by the correlation, the exponent is determined according to the weight, and the evaluation function is obtained; wherein, the training target is A, and finally the coefficient of the evaluation function of each indicator is determined to obtain the value evaluation model; wherein, the exponent is proportional to the weight.

[0102] Regarding step S400: the steps of generating an information acquisition template based on the determined location index and individual index, acquiring various indexes of the house to be traded based on the information acquisition template, inputting the trained value assessment model, and outputting the predicted price as the assessment result include:

[0103] Generate information acquisition template based on determined location indicators and individual indicators;

[0104] When receiving an evaluation request uploaded by a user, a permission acquisition request is sent to the user;

[0105] When receiving the permission granted by the user, sending the information acquisition template to the user, and acquiring various indicators of the house to be traded based on the information acquisition template;

[0106] The obtained indicators are input into the trained value assessment model, and the predicted price is output as the assessment result.

[0107] The application process of the model is very simple. The key point is that when an evaluation request uploaded by a user is received, a permission acquisition request is sent to the user. Only when the permission granted by the user is received can the corresponding data be obtained; this corresponds to the permission in step S100, that is, the granted permission refers to the permission granted by the user when obtaining data.

[0108] Figure 2 1 is a structural block diagram of a third-party existing housing value assessment system. In an embodiment of the present invention, a third-party existing housing value assessment system 10 includes:

[0109] The location index determination module 11 is used to obtain the location of the existing houses that have been traded and authorized within a preset time range, query the facility information within a preset space range based on the existing houses, and determine the location index of the existing houses according to the facility information;

[0110] The individual index determination module 12 is used to obtain the floor plan, decoration information and occupancy information of the existing housing, and determine the individual index according to the floor plan, decoration information and occupancy information;

[0111] The evaluation model training module 13 is used to query the transaction price of the existing housing, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; wherein, in the process of determining the location indicator and the individual indicator, the correlation with the price is simultaneously determined;

[0112] The evaluation model application module 14 is used to generate an information acquisition template based on the determined location indicators and individual indicators, obtain various indicators of the house to be traded based on the information acquisition template, input the trained value evaluation model, and output the predicted price as the evaluation result.

[0113] Furthermore, the location index determination module 11 includes:

[0114] A time range determination unit, configured to receive a time period selected by a user, take the current date as the right endpoint, and determine a time range based on the time period;

[0115] A facility information query unit, used to obtain the location of the stock houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating;

[0116] An indicator value determination unit, used to determine the indicator type according to the facility type, and determine the indicator value according to the facility location and facility rating;

[0117] The indicator value statistical unit is used to count all indicator values ​​by taking the indicator type as a column and the number of the existing house as a row to obtain the location indicator of the existing house;

[0118] The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include:

[0119] Specifically, the individual indicator determination module 12 includes:

[0120] An evaluation score calculation unit is used to obtain the floor plan of the existing house, input the floor plan into the trained floor plan evaluation model, and obtain the evaluation score;

[0121] An image recognition unit is used to obtain a visual image containing relative positions of the existing houses, recognize the visual image, and obtain the degree of decoration completeness with relative positions as labels;

[0122] A completeness correction unit is used to obtain the occupancy information of the existing houses and correct all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay;

[0123] The statistical unit is used to calculate the evaluation points and the degree of decoration completeness at each relative position to obtain individual indicators.

[0124] In addition, the evaluation model training module 13 includes:

[0125] A correlation reading unit is used to count the location index and the individual index, and read the correlation between each index and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation;

[0126] A weight calculation unit, used to calculate the weight of each indicator, and select an evaluation function for each indicator according to the weight and the correlation;

[0127] The error calculation unit is used to query the transaction price of existing users, take it as the true value, train the evaluation function corresponding to each indicator, and calculate the error;

[0128] The model output unit is used to output an evaluation function to obtain a value evaluation model when the error is less than a preset error threshold.

[0129] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A third-party existing housing value assessment method, characterized in that: The method comprises: Obtain the location of existing houses that have been traded and authorized within a preset time range, query facility information within a preset space range based on the existing houses, and determine the location index of the existing houses according to the facility information; Obtaining floor plans, decoration information, and occupancy information of existing houses, and determining individual indicators based on the floor plans, decoration information, and occupancy information; Query the transaction price of existing houses, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; in the process of determining the location indicator and individual indicator, simultaneously determine the correlation with the price; An information acquisition template is generated based on the determined location indicators and individual indicators, various indicators of the house to be traded are obtained based on the information acquisition template, the trained value assessment model is input, and the predicted price is output as the assessment result.

2. The third-party existing housing value assessment method according to claim 1, characterized in that: The steps of obtaining the locations of the existing houses that have been traded and authorized within a preset time range, querying the facility information within a preset space range based on the existing houses, and determining the location index of the existing houses according to the facility information include: Receive the time period selected by the user, take the current date as the right endpoint, and determine the time range based on the time period; Obtain the location of existing houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating; Determine the indicator type based on the facility type, and determine the indicator value based on the facility location and facility rating; Take the indicator type as the column and the number of the existing house as the row to count all the indicator values ​​and get the location indicator of the existing house; The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include: Z ij =αw j L j / d ij ; In the formula, Z ij is the index value corresponding to the jth index type of the ith stock house, α is the preset correction coefficient, and w j is the weight determined according to the facility type corresponding to the jth indicator type. The weight has a sign. When the sign is positive, it indicates a positive correlation with the transaction price. When the sign is negative, it indicates a negative correlation with the transaction price. L j is the facility rating of the facility type corresponding to the jth indicator type, d ij is the distance between the i-th stock housing and the facility type corresponding to the j-th indicator type.

3. The third-party existing housing value assessment method according to claim 1, characterized in that: The steps of obtaining the floor plan, decoration information and occupancy information of the existing housing, and determining the individual indicators according to the floor plan, decoration information and occupancy information include: Obtain the floor plans of existing houses, input the floor plans into the trained floor plan evaluation model, and obtain the evaluation score; Obtaining a visual image of the existing house containing relative positions, identifying the visual image, and obtaining the degree of decoration completeness with relative positions as labels; Obtaining the occupancy information of the existing houses, and correcting all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay; The evaluation points and the degree of decoration completeness at each relative position are statistically analyzed to obtain individual indicators.

4. The third-party existing housing value assessment method according to claim 1, characterized in that: The steps of querying the transaction price of the existing housing, taking the transaction price as the true value, training the evaluation function of each indicator, and obtaining the value evaluation model include: Counting location indicators and individual indicators, and reading the correlation between each indicator and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation; Calculating the weight of each indicator, and selecting an evaluation function for each indicator according to the weight and the correlation; Query the transaction price of existing users as the true value, train the evaluation function corresponding to each indicator, and calculate the error; When the error is less than a preset error threshold, an evaluation function is output to obtain a value evaluation model.

5. The third-party existing housing value assessment method according to claim 4 is characterized in that: The step of calculating the weight of each indicator and selecting an evaluation function for each indicator according to the weight and the correlation comprises: Read the variance of all indicator values ​​for each indicator, and determine the weight according to the inverse ratio of the variance; A power function is selected as a reference function, the sign of the coefficient is determined according to the correlation, and the exponent is determined according to the weight to obtain an evaluation function; Among them, the coefficients serve as target parameters in the training process.

6. The third-party existing housing value assessment method according to claim 1, characterized in that: The steps of generating an information acquisition template based on the determined location index and individual index, acquiring various indexes of the house to be traded based on the information acquisition template, inputting the trained value assessment model, and outputting the predicted price as the assessment result include: Generate information acquisition template based on determined location indicators and individual indicators; When receiving an evaluation request uploaded by a user, a permission acquisition request is sent to the user; When receiving the permission granted by the user, sending the information acquisition template to the user, and acquiring various indicators of the house to be traded based on the information acquisition template; The obtained indicators are input into the trained value assessment model, and the predicted price is output as the assessment result.

7. A third-party existing housing value assessment system, characterized in that: The system comprises: A location index determination module is used to obtain the location of the existing houses that have been traded and authorized within a preset time range, query the facility information within a preset space range based on the existing houses, and determine the location index of the existing houses according to the facility information; An individual indicator determination module is used to obtain the floor plan, decoration information and occupancy information of the existing housing, and determine the individual indicator according to the floor plan, decoration information and occupancy information; The evaluation model training module is used to query the transaction price of existing houses, take the transaction price as the real value, train the evaluation function of each indicator, and obtain the value evaluation model; in the process of determining the location indicator and individual indicator, the correlation with the price is determined simultaneously; The evaluation model application module is used to generate an information acquisition template based on the determined location indicators and individual indicators, obtain various indicators of the house to be traded based on the information acquisition template, input the trained value evaluation model, and output the predicted price as the evaluation result.

8. The third-party existing housing value assessment system according to claim 7, characterized in that: The location index determination module includes: A time range determination unit, configured to receive a time period selected by a user, take the current date as the right endpoint, and determine a time range based on the time period; A facility information query unit, used to obtain the location of the stock houses that have been traded and authorized within a preset time range, and query the facility information within the preset spatial range in the map service based on the location; the facility information includes facility type, facility location and facility rating; An indicator value determination unit, used to determine the indicator type according to the facility type, and determine the indicator value according to the facility location and facility rating; The indicator value statistical unit is used to count all indicator values ​​by taking the indicator type as a column and the number of the existing house as a row to obtain the location indicator of the existing house; The steps of determining the indicator type according to the facility type and determining the indicator value according to the facility location and facility rating include: Z ij =αw j L j / d ij ; In the formula, Z ij is the index value corresponding to the jth index type of the ith stock house, α is the preset correction coefficient, and w j is the weight determined according to the facility type corresponding to the jth indicator type. The weight has a sign. When the sign is positive, it indicates a positive correlation with the transaction price. When the sign is negative, it indicates a negative correlation with the transaction price. L j is the facility rating of the facility type corresponding to the jth indicator type, d ij is the distance between the i-th stock housing and the facility type corresponding to the j-th indicator type.

9. The third-party existing housing value assessment system according to claim 7, characterized in that: The individual indicator determination module comprises: An evaluation score calculation unit is used to obtain the floor plan of the existing house, input the floor plan into the trained floor plan evaluation model, and obtain the evaluation score; An image recognition unit is used to obtain a visual image containing relative positions of the existing houses, recognize the visual image, and obtain the degree of decoration completeness with relative positions as labels; A completeness correction unit is used to obtain the occupancy information of the existing houses and correct all decoration completeness according to the occupancy information; the occupancy information includes the number of occupants and the years of stay; The statistical unit is used to calculate the evaluation points and the degree of decoration completeness at each relative position to obtain individual indicators.

10. The third-party existing housing value assessment system according to claim 7, characterized in that: The evaluation model training module includes: A correlation reading unit is used to count the location index and the individual index, and read the correlation between each index and the transaction price according to a preset correlation table; the correlation includes positive correlation and negative correlation; A weight calculation unit, used to calculate the weight of each indicator, and select an evaluation function for each indicator according to the weight and the correlation; The error calculation unit is used to query the transaction price of existing households, take it as the true value, train the evaluation function corresponding to each indicator, and calculate the error; The model output unit is used to output the evaluation function to obtain the value evaluation model when the error is less than a preset error threshold.

Images