WO2018205372A1 - 房产信息处理方法、装置、计算机设备及存储介质 - Google Patents

房产信息处理方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2018205372A1
WO2018205372A1 PCT/CN2017/090580 CN2017090580W WO2018205372A1 WO 2018205372 A1 WO2018205372 A1 WO 2018205372A1 CN 2017090580 W CN2017090580 W CN 2017090580W WO 2018205372 A1 WO2018205372 A1 WO 2018205372A1
Authority
WO
WIPO (PCT)
Prior art keywords
configuration information
geographic location
property
target geographic
price
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/090580
Other languages
English (en)
French (fr)
Inventor
王健宗
黄章成
吴天博
肖京
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Priority to US16/313,793 priority Critical patent/US20190259071A1/en
Priority to SG11201810669VA priority patent/SG11201810669VA/en
Publication of WO2018205372A1 publication Critical patent/WO2018205372A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0278—Product appraisal
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/16—Real estate

Definitions

  • the present invention relates to the field of computer processing, and in particular, to a method, device, computer device and storage medium for processing property information.
  • a property information processing method, apparatus, computer device, and storage medium are provided.
  • a method for processing property information including:
  • a house price evaluation model is used to determine a property price estimate corresponding to the target geographic location based on the standardized feature value.
  • a real estate information processing device comprising:
  • a geographic location obtaining module configured to obtain a target geographic location corresponding to the real estate information to be evaluated
  • a configuration information acquiring module configured to acquire all configuration information in a preset range, centering on the target geographic location
  • a point value determining module configured to determine a score value corresponding to each configuration information according to a preset scoring standard
  • a feature value determining module configured to project the determined score value according to a distance between each configuration information and the target geographic location to obtain a standardized feature value
  • a property price determining module configured to perform, according to the standardized
  • the feature value uses a house price assessment model to determine an estimate of the property price corresponding to the target location.
  • a computer device comprising a memory and a processor, the memory storing computer readable instructions, the computer readable instructions being executed by the processor, causing the processor to perform the step of: obtaining a property to be evaluated The target geographic location corresponding to the information;
  • a house price evaluation model is used to determine a property price estimate corresponding to the target geographic location based on the standardized feature value.
  • One or more non-transitory readable storage mediums storing computer readable instructions, when executed by one or more processors, cause the one or more processors to execute Next steps:
  • a house price evaluation model is used to determine a property price estimate corresponding to the target geographic location based on the standardized feature value.
  • FIG. 1 is a block diagram showing the internal structure of a terminal in an embodiment
  • FIG. 2 is a block diagram showing the internal structure of a server in an embodiment
  • FIG. 3 is a flow chart of a method for processing property information in an embodiment
  • FIG. 4 is a flow chart of a method for establishing a house price evaluation model in an embodiment
  • FIG. 5 is a flow chart of a method for projecting a determined score value to obtain a standardized feature value in one embodiment
  • FIG. 6 is a flow chart of a method for determining a property price estimate corresponding to a target geographic location by using a house price evaluation model according to a standardized feature value in an embodiment
  • FIG. 7 is a flowchart of a method for determining a score value corresponding to each configuration information according to a preset scoring standard in an embodiment
  • Figure 8 is a block diagram showing the structure of a property information processing apparatus in an embodiment
  • FIG. 9 is a structural block diagram of a property information processing apparatus in another embodiment.
  • Figure 10 is a block diagram showing the structure of a property price determination module in one embodiment.
  • the internal structure of the terminal 102 is as shown in FIG. 1, including a processor connected through a system bus, an internal memory, a non-volatile storage medium, a network interface, a display screen, and an input device. .
  • the non-volatile storage medium of the terminal 102 stores an operating system and computer readable instructions executable by the processor to implement a property information processing method suitable for the terminal 102.
  • the processor is used to provide computing and control capabilities to support the operation of the entire terminal.
  • the internal memory in the terminal provides an environment for operating the operating system and computer readable instructions in the non-volatile storage medium.
  • the network interface is used to connect to the network for communication.
  • the display screen of the terminal 102 may be a liquid crystal display or an electronic ink display screen.
  • the input device may be a touch layer covered on the display screen, or may be a button, a trackball or a touchpad provided on the outer casing of the electronic device, or may be An external keyboard, trackpad, or mouse.
  • the terminal can be a tablet, a laptop, a desktop computer, or the like.
  • FIG. 1 A person skilled in the art can understand that the structure shown in FIG. 1 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied.
  • the specific terminal may include a ratio. More or fewer components are shown in the figures, or some components are combined, or have different component arrangements.
  • the internal structure of server 104 includes a processor coupled through a system bus, a non-volatile storage medium, an internal memory, and a network interface.
  • the non-volatile storage medium includes an operating system and computer readable instructions executable by the processor to implement a property information processing method suitable for the server 104, the processor of the server for providing calculations and Control capabilities that support the operation of the entire server.
  • the internal memory in the server provides for the operation of the operating system and computer readable instructions in the non-volatile storage medium surroundings.
  • the server's network interface is used to communicate with external servers or terminals over a network connection. It will be understood by those skilled in the art that the structure shown in FIG.
  • FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the server to which the solution of the present application is applied.
  • the specific server may include a ratio. More or fewer components are shown in the figures, or some components are combined, or have different component arrangements.
  • a method for processing property information is proposed.
  • the method can be applied to a terminal or a server, and specifically includes the following steps:
  • Step 302 Obtain a target geographic location corresponding to the real estate information to be evaluated.
  • the target geographical location refers to the location of the property information to be evaluated, and the target geographic location is represented by the latitude and longitude values. Since the average transaction price of some real estates (that is, the transaction price per unit area) is not directly available, it is necessary to estimate the average transaction price of these real estates that cannot be directly obtained, and the average transaction price is close to the geographical location. Related, wherein the average transaction price refers to the transaction price per unit area, that is, the transaction price per square meter.
  • Step 304 Locating all the configuration information in the preset range centering on the target geographic location.
  • factors affecting the price of the real estate include, in addition to the geographical location, configuration facilities around the real estate, such as hospitals and schools. Therefore, in addition to obtaining the target geographical location of the property to be evaluated, it is also necessary to obtain all the configuration information around the real estate centered on the target geographic location, wherein the configuration information includes schools, hospitals, shopping malls, transportation, attractions, hotels, etc.
  • the life factor of house price information The influence of the configuration information around the property on the price of the property is related to the distance. If the distance is too far, the influence is almost negligible. Therefore, it is only necessary to obtain configuration information within a preset range (for example, within 2000 meters).
  • Step 306 Determine a score value corresponding to each configuration information according to a preset scoring standard.
  • the score value corresponding to the configuration information In this embodiment, in order to be able to quantify the influence degree of each configuration information on the property price, after obtaining all the configuration information within the preset range of the real estate, it is necessary to determine each according to the preset scoring standard.
  • the score value corresponding to the configuration information Different types of configuration information have different scoring standards. For example, for schools, they can be scored according to the primary, secondary, and local rankings. For hospitals, they can be scored according to the hospital level, for example, according to the top three hospitals, the dimethyl hospital, etc. Score. Specifically, first, attribute information corresponding to each configuration information is obtained, and the attribute information refers to a category to which the configuration information belongs, for example, belonging to a school, a hospital, a transportation facility, or the like.
  • the configuration information is classified according to the attribute information in advance, and then the scoring standards corresponding to different configuration information are set, that is, the rating standards corresponding to different attribute information are different. Therefore, after obtaining the attribute information corresponding to each configuration information, the rating standard corresponding to the configuration information may be obtained according to the attribute information corresponding to the configuration information, and then the score value corresponding to the configuration information is determined according to the rating standard. For example, if the configuration information is a hospital and the rating criteria for the hospital are determined according to the hospital's rating, then it is a top three hospital. Then the corresponding score value can be set to 3. If it is a hospital, the corresponding score value can be set to 2.
  • Step 308 projecting the determined score value according to the distance between each configuration information and the target geographic location to obtain a standardized feature value.
  • the influence degree of the configuration information on the real estate is not only related to the configuration information itself, but also related to the distance between the configuration information and the real estate, that is, the influence of the same configuration information at different distances on the real estate is different. Therefore, after obtaining the score value corresponding to each configuration information, it is also necessary to project the calculated score value according to the distance from the geographic location of the real estate, and obtain a standardized feature value. The standardized feature value is subsequently used as a basis for evaluating the price of the property to conduct a corresponding house price evaluation.
  • the attribute values of the configuration information are collectively projected to a value between 0 and 1 according to the distance of the configuration information, and then the score value obtained above is multiplied by the corresponding attribute value to obtain a normalized feature value. That is to say, a coefficient factor related to the distance is set for each configuration information, and the obtained score values are uniformly standardized by multiplying the coefficient factors, so as to more accurately determine the weight of each configuration information on the house price.
  • the coefficient factors corresponding to different distances may be set, for example, within 0-100 m, the coefficient factor is set to 1, and in the range of 100 m-200 m, the coefficient factor is set to 0.9, and the range is set in the range of 200-500 m.
  • the coefficient factor is 0.8, etc., the further the distance, the smaller the corresponding coefficient factor.
  • the corresponding coefficient factor can be based on the actual situation Live setting. For example, suppose a distance from a real estate has a score of 3 in a top three hospital within 200m, and a score of 2 in a dimethyl hospital within 500m, then the value of the corresponding score is standardized to be 3 *0.9 and 2*0.8.
  • the sigmoid function can be used as a coefficient factor for the distance:
  • the value of x represents the distance between the configuration information and the target geographic location
  • the fractional value corresponding to the configuration information is projected by the formula to obtain a standardized feature value.
  • Step 310 Determine a property price estimate corresponding to the target geographic location by using a house price evaluation model according to the standardized feature value.
  • the pre-established house price evaluation model is used according to the standardized feature value to calculate the property price estimate corresponding to the target geographic location. value.
  • the property evaluation model is obtained by training in advance according to the obtained known property price and the standardized feature value corresponding to the configuration information around the property. Therefore, the price of the property can be estimated based on the standardized feature values corresponding to the configuration information around the property.
  • all the configuration information in the preset range is obtained by acquiring the target geographic location corresponding to the real estate information to be evaluated, and the configuration information corresponding to each configuration information is determined according to the preset scoring standard.
  • the score value is obtained by projecting the determined score value according to the distance between each configuration information and the target geographical position to obtain a standardized feature value, according to the standardized
  • the value of the property is determined using a house price assessment model to determine the price of the property corresponding to the target location.
  • the method for processing the above property information can automatically evaluate the house price according to the geographical location of the property and the configuration information around the geographic location by using the established house price evaluation model, compared with the method of traditional professional evaluation. Not only saves time and effort, but also because the establishment of the housing price assessment model is based on big data, it can greatly reduce the bias caused by manual evaluation, and is conducive to improving the accuracy of assessment.
  • the method before the step of obtaining the target geographic location corresponding to the real estate information to be evaluated, the method further includes: establishing a house price evaluation model.
  • establishing a house price evaluation model specifically includes the following steps:
  • step 312 an initialized house price evaluation model is established.
  • the price information is related to the geographical location of the real estate, it is also closely related to the surrounding configuration facilities.
  • it is first necessary to establish an initial price assessment model. Since the configuration information around the property is to improve people's quality of life, it can be assumed that the various configuration information factors obtained are positively correlated with the property price.
  • a simple linear regression model can be used as the initial price assessment model.
  • X 1 , X 2 , X 3 , ..., X n represents the characteristic value taken by the corresponding configuration information
  • Y represents the corresponding estimated price of the house price.
  • Step 314 Train the initialized house price evaluation model according to the collected property price and the configuration information around the property to determine corresponding model parameters.
  • the initialized house price evaluation model After the initialized house price evaluation model is established, it is necessary to construct a training set for training the initial price evaluation model.
  • the collected property information of known house prices and the property information of the surrounding information of the property are used as training sets, and according to preset rating standards, respectively Determining the score value corresponding to each configuration information, and normalizing the determined score value according to the distance to obtain a corresponding feature value, and adopting the corresponding house price evaluation model according to the feature value corresponding to the configuration information and the corresponding known house price information
  • the machine learning algorithm performs training learning to obtain corresponding coefficient values, wherein the machine learning algorithm may adopt algorithms such as least squares method and gradient descent.
  • cross-validation is considered during model training.
  • Cross-validation is a practical way to statistically cut data samples into smaller subsets, which can be analyzed on a subset first, while other subsets are used for subsequent validation and verification of this analysis.
  • the first subset is called the training set, while the other subset is called the validation set or test set.
  • the goal of cross-validation is to define a data set to the test model during the training phase in order to reduce problems like overfitting.
  • the value corresponding to each feature is related to the distance between the feature and the real estate, in addition to the feature itself.
  • the established house price evaluation model is trained by using the data in the collected data space to obtain corresponding coefficient values, thereby determining the final house price evaluation model.
  • the goals of the model may be set as follows:
  • the formula represents a function When taking the minimum value, the corresponding value of w is the calculated model parameter.
  • y represents the true value
  • X represents the matrix of the various feature factors collected above
  • w is the final learned parameter, representing the parameters of the model.
  • w is a one-dimensional vector
  • y is also a one-dimensional vector (each value represents the average price of a real estate)
  • subscript 2 represents the vector norm
  • the vector norm is the square of each element in the vector. And re-open the root number.
  • the initialization evaluation model is trained using a machine learning algorithm to make the function
  • Step 316 obtaining a house price evaluation model according to the determined model parameters.
  • the house price evaluation model is obtained according to the determined model parameters.
  • the obtained model needs to be verified and evaluated.
  • the evaluation criteria include the mean and variance of the absolute deviation, the median of the absolute deviation, the R2 score, etc., and only the qualified ones are qualified.
  • the model is allowed to be used to make the corresponding house price forecast, otherwise the corresponding model parameters need to be adjusted repeatedly until the model can meet the corresponding standards. specifically,
  • n samples represent the number of all features
  • the mean of the absolute deviation reflects the degree of the overall offset, but there will be deviations in the evaluation of extremely high prices and very low prices (such as villa area, economically applicable area), the variance of the absolute deviation If the mean is equal, the lower variance can indicate that the evaluation error is smaller; the median of the absolute deviation takes into account the deviation in most cases, but the overall deviation may be large, and the R2 score is also called the goodness of fit. It reflects the difference between the variance of the predicted value and the variance of the true value. The closer to 1 is, the overall distribution from which the categorical data comes from is consistent with the predicted distribution.
  • the step 308 of projecting the determined score value according to the distance between each configuration information and the target geographic location to obtain a standardized feature value comprises:
  • Step 308A Calculate the distance between each configuration information and the target geographic location.
  • Step 308B According to the distance between each configuration information and the target geographical location, the determined score value is projected by using a sigmoid function to obtain a standardized feature value.
  • the score value after determining the score value corresponding to each configuration information, the score value needs to be standardized according to a preset rule to facilitate subsequent calculation. Specifically, first, the distance between each configuration information and the target geographic location is calculated, and then the calculated distance value is projected by the sigmoid function according to the calculated distance to obtain a standardized feature value. This eigenvalue facilitates subsequent evaluation of the property price based on a pre-established house price assessment model. Specifically, the sigmoid function is used as a coefficient factor of the distance:
  • the value of x represents the distance between the configuration information and the target geographic location
  • the fractional value corresponding to the configuration information is projected by the formula to obtain a normalized feature value, that is, the fractional value corresponding to the configuration information is multiplied by the coefficient factor to obtain a normalized feature value corresponding to the configuration information.
  • the step 310 of determining a property price estimate corresponding to the target geographic location using the house price assessment model based on the normalized feature values includes:
  • Step 310A classify the configuration information according to the attribute of the configuration information, and determine a standardized feature value corresponding to each type of configuration information.
  • the configuration information may be classified according to the attributes of the configuration information, and then each type of configuration information is used as a feature to determine a standardized feature value corresponding to each type of configuration information.
  • the configuration information can be divided into education, medical, transportation, tourism, business, and life services according to the attributes of the configuration information. For example, primary school, middle school, etc. are classified as education, and the top three hospitals and the dimethyl hospital are unified into medical treatment, and commercial center factors and commercial food factors are unified into business. For example, if there are two hospitals around the property, one is the top three hospitals, the characteristic value is 3, and the other is the dimethyl hospital. The characteristic value is 2 points, then the medical characteristics corresponding to the property are corresponding to the characteristic values. Sum.
  • Step 310B Substituting the determined standardized feature values corresponding to each type of configuration information into the real estate evaluation model to determine an estimated price of the real estate corresponding to the target geographic location.
  • the determined feature value corresponding to each type of configuration information is substituted into the real estate evaluation model, and the real estate evaluation model evaluates the corresponding real estate price according to the input characteristic values. Get an estimate of the price of the property.
  • the property evaluation model also trains the configuration information after training and learning.
  • the step 306 of determining the score value corresponding to each configuration information according to a preset scoring criterion includes:
  • Step 306A Obtain attribute information corresponding to each configuration information.
  • the attribute information refers to the type of the surrounding configuration facility itself.
  • the attribute information corresponding to the hospital is medical
  • the attribute information corresponding to the school is education
  • the attribute information corresponding to the mall is commercial
  • the supermarket corresponds.
  • the attribute information is for living services.
  • the attribute information corresponding to each configuration information is obtained in order to obtain a scoring standard corresponding to the configuration information. Because different kinds of configuration information scoring standards are definitely different, for example, schools are ranked according to rankings, and hospitals can use ratings to score, while shopping malls and supermarkets can be scored according to size.
  • Step 306B Acquire a scoring standard corresponding to the attribute information.
  • the correspondence between the attribute information and the scoring standard is stored in advance. Therefore, after the attribute information corresponding to the configuration information is obtained, the corresponding scoring standard can be searched according to the obtained attribute information.
  • the scoring criteria are set to quantify the impact of configuration information on house prices. The subsequent price assessment can then be carried out based on the determined score values.
  • Step 306C Determine a score value corresponding to the configuration information according to the scoring standard.
  • the score value corresponding to the configuration information may be calculated according to the scoring standard, and specifically, the feature of the configuration information is acquired to determine a corresponding score value, such as If the configuration information is a hospital, then the corresponding rating and word of mouth of the hospital are obtained, and the corresponding score value is determined according to the rating and word of mouth. For example, if the score of the top three hospital is 3, if the word of mouth is excellent, then 1 is added. Points, if it is normal, no points will be added. If it is bad, it will be reduced by 1 point.
  • a property information processing apparatus comprising:
  • the geographic location obtaining module 802 is configured to obtain a target geographic location corresponding to the real estate information to be evaluated.
  • the configuration information obtaining module 804 is configured to obtain all configuration information in the preset range centering on the target geographic location.
  • the score value determining module 806 is configured to determine a score value corresponding to each configuration information according to a preset scoring standard.
  • the feature value determining module 808 is configured to project the determined score value according to the distance between each configuration information and the target geographic location to obtain a standardized feature value.
  • the property price determining module 810 is configured to determine, according to the standardized feature value, a house price estimation model corresponding to the target geographic location by using a house price evaluation model.
  • the foregoing property information processing apparatus further includes:
  • a module 812 is created for establishing an initialized house price assessment model.
  • the model parameter determining module 814 is configured to train the initialized house price evaluation model according to the collected property price and configuration information around the property to determine corresponding model parameters.
  • the model determination module 816 is configured to obtain a house price evaluation model according to the determined model parameters.
  • the feature value determining module 808 is further configured to calculate a distance between each configuration information and the target geographic location, according to a distance between each configuration information and the target geographic location, using a sigmoid function The determined score values are projected to obtain standardized feature values.
  • the property price determination module 810 includes:
  • the classification module 810A is configured to classify the configuration information according to the attribute of the configuration information, and determine that each type of configuration information corresponds to the standardized feature value.
  • the estimated value determining module 810B is configured to substitute the determined standardized feature value corresponding to each type of configuration information into the property evaluation model to determine a property price estimate corresponding to the target geographic location.
  • the score value determining module 806 is further configured to acquire attribute information corresponding to each configuration information, obtain a scoring standard corresponding to the attribute information, and determine, according to the scoring standard, a corresponding to the configuration information. Score value.
  • the various modules in the above described property information processing apparatus may be implemented in whole or in part by software, hardware, and combinations thereof.
  • the network interface may be an Ethernet card or a wireless network card.
  • the above modules may be embedded in the hardware in the processor or in the memory in the server, or may be stored in the memory in the server, so that the processor calls the corresponding operations of the above modules.
  • the processor can be a central processing unit (CPU), a microprocessor, a microcontroller, or the like.
  • a computer apparatus comprising a memory and a processor, the memory storing computer readable instructions that, when executed by the processor, cause the processor to execute The following steps: obtaining the target geographic location corresponding to the real estate information to be evaluated; taking all the configuration information in the preset range as the center of the target geographical location; determining the score value corresponding to each configuration information according to the preset scoring standard; And calculating the score value according to the distance between each configuration information and the target geographic location to obtain a standardized feature value; A house price evaluation model is used to determine a property price estimate corresponding to the target geographic location based on the standardized feature value.
  • the processor executing the computer program is further used to implement the following steps: establishing an initialized house price evaluation model; The property price and the configuration information around the property are trained to determine the corresponding model parameters; and the house price evaluation model is obtained according to the determined model parameters.
  • the processor performs a projection according to the distance between each configuration information and the target geographic location to determine the score value to obtain a standardized feature value, including: calculating each configuration information. a distance from the target geographic location; according to the distance between each configuration information and the target geographic location, the determined score value is projected using a sigmoid function to obtain a standardized feature value.
  • the determining, by the processor, the property price estimation value corresponding to the target geographic location by using the house price evaluation model according to the standardized feature value comprising: configuring information according to an attribute of the configuration information. Performing classification, determining feature values corresponding to each type of configuration information; and substituting the determined feature values corresponding to each type of configuration information into the property evaluation model to determine an estimated price of the property corresponding to the target geographic location.
  • the step of determining, by the processor, the score value corresponding to each configuration information according to a preset scoring criterion comprises: acquiring attribute information corresponding to each configuration information; acquiring the attribute information Corresponding scoring criteria; determining a score value corresponding to the configuration information according to the scoring standard.
  • one or more non-transitory computer readable storage media storing computer readable instructions, when executed by one or more processors, causing the one or more The processor performs the following steps: obtaining a target geographic location corresponding to the real estate information to be evaluated; taking all the configuration information in the preset range as the center of the target geographic location; determining each configuration information according to a preset scoring standard a fractional value; projecting the determined score value according to the distance between each configuration information and the target geographic location to obtain a standardized feature a value; a house price evaluation model is used to determine an estimate of the property price corresponding to the target geographic location based on the standardized feature value.
  • the computer readable instructions are executed by one or more processors prior to the step of obtaining a target geographic location corresponding to the property information to be evaluated, such that the one or more processors further Performing the following steps: establishing an initialized house price evaluation model; training the initialized house price evaluation model according to the collected property price and configuration information around the property to determine corresponding model parameters; and obtaining house price evaluation according to the determined model parameters model.
  • the one or more processors perform projecting the determined score values according to the distance between each configuration information and the target geographic location to obtain standardized feature values, including: The distance between each configuration information and the target geographic location; according to the distance between each configuration information and the target geographic location, the determined score value is projected by a sigmoid function to obtain a standardized feature value.
  • the determining, by the one or more processors, the property price estimation value corresponding to the target geographic location by using a house price evaluation model according to the standardized feature value including: according to configuration information The attribute classifies the configuration information, determines the feature value corresponding to each type of configuration information, and substitutes the determined feature value corresponding to each type of configuration information into the property evaluation model to determine an estimated price of the property corresponding to the target geographic location.
  • the step of determining, by the one or more processors, the score value corresponding to each configuration information according to a preset scoring criterion comprises: acquiring attribute information corresponding to each configuration information; acquiring and a rating criterion corresponding to the attribute information; determining a score value corresponding to the configuration information according to the rating criterion.
  • the storage medium may be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Accounting & Taxation (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种房产信息处理方法，包括：获取待评估的房产信息对应的目标地理位置，以所述目标地理位置为中心，获取预设范围内的所有配置信息，根据预设的评分标准确定每一个配置信息对应的分数值，根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值，根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。

Description

房产信息处理方法、装置、计算机设备及存储介质
本申请要求于2017年5月10日提交中国专利局、申请号为2017103264787、发明名称为“房产信息处理方法、装置、计算机设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及计算机处理领域，特别是涉及一种房产信息处理方法、装置、计算机设备及存储介质。
背景技术
随着经济的发展，出现了房地产金融业的繁荣，正确评估房地产价值对房产金融的开发和投资有十分重要的意义。对房地产进行估价，通常需要由持有《房地产估价人员岗位合格证书》或《房地产估价师注册证》的专业人员，根据估价目的，遵循估价原则，按照估价程序，运用估价方法，在综合分析影响房地产价格因素的基础上，结合估价经验及对影响房地产价格因素的分析，对房地产最可能实现的合理价格所作出的估计、推测与判断。但是传统的基于从业人员的估价不仅耗时耗力，而且往往会由于个人认知的偏差导致评估不够准确。
发明内容
根据本申请的各种实施例，提供一种房产信息处理方法、装置、计算机设备及存储介质。
一种房产信息处理方法，包括：
获取待评估的房产信息对应的目标地理位置；
以所述目标地理位置为中心，获取预设范围内的所有配置信息；
根据预设的评分标准确定每一个配置信息对应的分数值；
根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
一种房产信息处理装置，包括：
地理位置获取模块，用于获取待评估的房产信息对应的目标地理位置；
配置信息获取模块，用于以所述目标地理位置为中心，获取预设范围内的所有配置信息；
分数值确定模块，用于根据预设的评分标准确定每一个配置信息对应的分数值；
特征值确定模块，用于根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及房产价格确定模块，用于根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步骤：获取待评估的房产信息对应的目标地理位置；
以所述目标地理位置为中心，获取预设范围内的所有配置信息；
根据预设的评分标准确定每一个配置信息对应的分数值；
根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
一个或多个存储有计算机可读指令的非易失性可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以 下步骤：
获取待评估的房产信息对应的目标地理位置；
以所述目标地理位置为中心，获取预设范围内的所有配置信息；
根据预设的评分标准确定每一个配置信息对应的分数值；
根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
本发明的一个或多个实施例的细节在下面的附图和描述中提出。本发明的其它特征、目的和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为一个实施例中终端的内部结构框图；
图2为一个实施例中服务器的内部结构框图；
图3为一个实施例中房产信息处理方法的流程图；
图4为一个实施例中建立房价评估模型的方法流程图；
图5为一个实施例中将确定的分数值进行投影得到标准化的特征值的方法流程图；
图6为一个实施例中根据标准化的特征值采用房价评估模型确定与目标地理位置对应的房产价格估计值的方法流程图；
图7为一个实施例中根据预设的评分标准确定每一个配置信息对应的分数值的方法流程图；
图8为一个实施例中房产信息处理装置的结构框图；
图9为另一个实施例中房产信息处理装置的结构框图；
图10为一个实施例中房产价格确定模块的结构框图。
具体实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
如图1所示，在一个实施例中，终端102的内部结构如图1所示，包括通过系统总线连接的处理器、内存储器、非易失性存储介质、网络接口、显示屏和输入装置。其中，终端102的非易失性存储介质存储有操作系统和计算机可读指令，该计算机可读指令可被处理器执行以实现适用于终端102的一种房产信息处理方法。该处理器用于提供计算和控制能力，支撑整个终端的运行。终端中的内存储器为非易失性存储介质中的操作系统和计算机可读指令运行提供环境。网络接口用于连接到网络进行通信。终端102的显示屏可以是液晶显示屏或者电子墨水显示屏等，输入装置可以是显示屏上覆盖的触摸层，也可以是电子设备外壳上设置的按键、轨迹球或触控板，也可以是外接的键盘、触控板或鼠标等。该终端可以是平板电脑、笔记本电脑、台式计算机等。本领域技术人员可以理解，图1中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的终端的限定，具体的终端可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
如图2所示，在一个实施例中，服务器104的内部结构如图2所示，包括通过系统总线连接的处理器、非易失性存储介质、内存储器和网络接口。其中，该非易失存储介质包括操作系统和计算机可读指令，该计算机可读指令可被处理器执行以实现适用于服务器104的一种房产信息处理方法，该服务器的处理器用于提供计算和控制能力，支撑整个服务器的运行。服务器中的内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供 环境。该服务器的网络接口用于与外部的服务器或终端通过网络连接通信。本领域技术人员可以理解，图2中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的服务器的限定，具体的服务器可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
如图3所示，在一个实施例中，提出了一种房产信息处理方法，该方法可以应用于终端或服务器中，具体包括以下步骤：
步骤302，获取待评估的房产信息对应的目标地理位置。
在本实施例中，目标地理位置是指待评估的房产信息所在的位置，该目标地理位置采用经纬度数值来表示。由于有些楼盘小区的成交均价(即单位面积的成交价格)没有办法直接获取，所以需要对这些不能直接获取到的房产的成交均价进行估计，而房产成交均价是与所在的地理位置紧密相关的，其中，成交均价是指单位面积的成交价格，即每平方米的成交价格。所以为了能够评估这些未知的房价信息，首先需要获取待评估的房产信息对应的目标地理位置，房产的目标地理位置是固定的，采用传统的定位技术即可获取该房产所在的经纬度信息。
步骤304，以目标地理位置为中心，获取预设范围内的所有配置信息。
在本实施例中，影响楼盘价格的因素除了地理位置外，还包括该楼盘周围的配置设施，比如，医院、学校。所以除了获取待评估房产的目标地理位置外，还需要以该目标地理位置为中心，获取该楼盘周围的所有配置信息，其中，配置信息包括学校、医院、商场、交通、景点、酒店等等影响房价信息的生活因素。楼盘周围的配置信息对该楼盘价格的影响是与距离是有关的，如果距离太远，那么影响度几乎可以忽略。所以只需要获取预设范围内的(比如，2000米以内)的配置信息。
步骤306，根据预设的评分标准确定每一个配置信息对应的分数值。
在本实施例中，为了能够量化每一个配置信息对房产价格的影响度，获取到房产预设范围内的所有配置信息后，需要根据预设的评分标准确定每一 个配置信息对应的分数值。不同种类的配置信息对应的评分标准不同，比如，对于学校，可以根据小学、中学以及在当地的排名进行打分；对于医院，可以根据医院等级进行评分，比如，根据三甲医院、二甲医院等进行打分。具体地，首先，获取每一个配置信息对应的属性信息，属性信息是指该配置信息所属的种类，比如，属于学校、医院还是交通设施或是其他。预先对配置信息按照属性信息进行分类，然后设置不同配置信息对应的评分标准，即不同的属性信息对应的评分标准不同。所以在获取到每一个配置信息对应的属性信息后，根据该配置信息对应的属性信息就可以获取到与该配置信息对应的评分标准，然后根据该评分标准确定与该配置信息对应的分数值。举个例子，如果配置信息为医院，且针对于医院的评分标准是根据医院的等级来确定的，那么如果是三甲医院。则对应的分数值可以设为3，如果是二甲医院，那么对应的分数值可以设为2。
步骤308，根据每一个配置信息与目标地理位置之间的距离将确定的分数值进行投影得到标准化的特征值。
在本实施例中，配置信息对房产的影响度不仅与配置信息本身有关，还与该配置信息与该房产的距离有关，即相同配置信息处于不同距离位置对该房产的影响也是不一样的，所以在获取到每个配置信息对应的分数值后，还需要根据与房产的地理位置之间的距离对计算得到的分数值进行投影，得到标准化的特征值。后续将该标准化的特征值作为评估房产价格的依据进行相应的房价评估。具体地，根据配置信息的距离将配置信息的属性值共同投影到一个0到1之间的值，然后将上述得到的分数值乘以相应的属性值得到标准化的特征值。也就是说，为每一个配置信息设置一个与距离有关的系数因子，通过乘以该系数因子将得到的分数值进行统一标准化，以便后续更加准确的确定每个配置信息对房价的影响权重。在一个实施例中，可以设置不同距离对应的系数因子，比如，在0-100m以内，设置系数因子为1，在100m-200m范围内，设置系数因子为0.9，在200-500m范围内，设置系数因子为0.8等等，距离越远，相应的系数因子越小。相应的系数因子可以根据实际情况灵 活的设置。举个例子，假设距离一个楼盘，在200m内有一个三甲医院的分数值为3，在500m内有一个二甲医院的分数值为2，那么将相应的分数值进行标准化后的值分为3*0.9和2*0.8。
在另一个实施例中，可以使用sigmoid函数作为距离的一个系数因子：
Figure PCTCN2017090580-appb-000001
其中，x的值表示配置信息与目标地理位置之间的距离，d的值为预先设置的距离范围，比如设置d＝1000m。τ的值决定了因子关于距离的衰减程度，取值越大，衰减越慢。比如，取值τ＝20。通过该公式将配置信息对应的分数值进行投影得到标准化的特征值。举一个例子，假如距离一个楼盘，在1000m处有一个三甲医院的评分为3，800m处有一个二甲医院评分为2，其中，d的值设为1500m，τ＝20，那么将相应的分数值进行标准化后分别为
Figure PCTCN2017090580-appb-000002
和
Figure PCTCN2017090580-appb-000003
步骤310，根据标准化的特征值采用房价评估模型确定与目标地理位置对应的房产价格估计值。
在本实施例中，在根据配置信息与目标地理位置之间的距离得到标准化的特征值后，根据该标准化的特征值采用预先建立好的房价评估模型来计算得到目标地理位置对应的房产价格估计值。其中，房产评估模型是预先根据获取到的已知房产价格与该房产周围的配置信息所对应的标准化的特征值进行训练得到的。所以根据房产周围的配置信息所对应的标准化的特征值就可以估计出该房产价格。
在本实施例中，通过获取待评估的房产信息对应的目标地理位置，以该目标地理位置为中心，获取预设范围内的所有配置信息，根据预设的评分标准确定每一个配置信息对应的分数值，根据每一个配置信息与目标地理位置之间的距离将确定的分数值进行投影得到标准化的特征值，根据标准化的特 征值采用房价评估模型确定与目标地理位置对应的房产价格估计值。上述房产信息处理的方法，通过采用建立的房价评估模型根据房产的地理位置以及该地理位置周围的配置信息就可以自动对该房价进行评估，与传统的专业人员进行评估的方式相比，该方法不仅省时省力，而且由于房价评估模型的建立是基于大数据进行建立的，能大幅度降低人工评估带来的偏差，有利于提高评估的准确性。
在一个实施例中，在获取待评估的房产信息对应的目标地理位置的步骤之前还包括：建立房价评估模型。
如图4所示，建立房价评估模型具体包括以下步骤：
步骤312，建立初始化的房价评估模型。
在本实施例中，考虑到房价信息与房产所在的地理位置有关外，还与其周围的配置设施有密切关系。为了能够根据房产的地理位置以及周围配置信息自动对房产的价格信息进行估计，首先需要建立一个初始化的房价评估模型。由于房产周围的配置信息都是为了提高人的生活品质，所以可以假设获取到的各个配置信息因子与房产价格都是呈现正相关的。在一个实施例中，可以采用一个简单的线性回归模型作为初始化的房价评估模型，设
Y＝β0+β1X1+β2X2+…+βnXn
其中，X1，X2，X3，......，Xn代表的是相应的配置信息所取的特征值，Y代表的是相应的房价估计值。初始化的房价评估模型建立后，接下来就是需要对线性回归模型进行训练学习以便得到各个配置信息所对应的系数值，即β0，β1，β2，...，βn的值。
步骤314，根据采集到的房产价格和该房产周围的配置信息对初始化的房价评估模型进行训练确定相应的模型参数。
在本实施例中，在建立了初始化的房价评估模型后，需要构建对该初始化的房价评估模型进行训练学习的训练集。将采集到的已知房价信息与该房产周围的配置信息的房产数据作为训练集，并且按照预设的评分标准，分别 确定每一个配置信息对应的分数值，并对该确定的分数值根据距离进行标准化得到相应的特征值，根据配置信息所对应的特征值以及相应的已知的房价信息对初始化的房价评估模型采用机器学习算法进行训练学习，以便获取到相应的系数值，其中，机器学习算法可以采用最小二乘法和梯度下降等算法。
如果使用采集到的全部的数据进行学习，那么很容易导致过拟合现象。所以，在一个实施例中，在模型训练的时候考虑交叉验证。交叉验证是一种统计学上将数据样本切割成较小子集的实用方法，可以先在一个子集上做分析，而其它子集则用来做后续对此分析的确认以及验证。一开始的子集被称为训练集，而其它的子集则被称为验证集或测试集。交叉验证的目标是定义一个数据集到测试的模型，在训练阶段以便减少像过拟合的问题。
在一个实施例中，首先定义采集到的所有的数据空间为p＝(χ，y)，其中χ＝(X1，X2，X3……)代表了采集到的数据的特征空间，其中的每一个Xi对应着相应的数据点yi，yi代表了每一个楼盘地产的真实价值。需要注意的是，Xi＝{xi1，xi2，xi3……}，其中的每一个值xij代表了一个特征。而每一个特征所对应的值除了与该特征本身有关外，还与该特征与楼盘之间的距离相关。通过使用采集到的数据空间中的数据对建立的初始化的房价评估模型进行训练以便得到相应的系数值，进而确定最终的房价评估模型。具体地，不妨设定模型的目标如下：
Figure PCTCN2017090580-appb-000004
该公式表示函数
Figure PCTCN2017090580-appb-000005
取最小值时，对应的w的值就是计算得到的模型参数。其中，y代表的是真实价值，X代表的是上述搜集到的各类特征因子的矩阵，而w是最终学习出来的参数，代表了模型的参数。需要说明的是：w是一个一维向量，y也是一个一维向量(每一个值代表一个楼盘的均价)，下标2代表向量范数，向量的2范数是向量中各个元素平方之和再开根号。采用机器学习的算法对该初始化的评估模型进行训练，使得函数
Figure PCTCN2017090580-appb-000006
取最小值时，确定对应的模型参数w，其中，w＝(β0，β1，β2，...，βn)。即确定模型参数w的值就是分别确定β0，β1，β2，...，βn的值。
步骤316，根据确定的模型参数得到房价评估模型。
在本实施例中，在根据上述方式计算得到模型参数后，根据确定的模型参数得到房价评估模型。
在一个实施例中，在确定了各个模型参数后，需要对该得到的模型进行验证评估，评估标准包括绝对偏差的均值和方差，绝对偏差的中位数，R2分值等，只有验证合格的模型才允许用来进行相应的房价预测，不然需要反复调整相应的模型参数，直到模型能够符合相应的标准。具体地，
绝对偏差的均值：
Figure PCTCN2017090580-appb-000007
绝对偏差的方差：
Figure PCTCN2017090580-appb-000008
绝对偏差的中位数：
Figure PCTCN2017090580-appb-000009
R2分值：
Figure PCTCN2017090580-appb-000010
其中
Figure PCTCN2017090580-appb-000011
其中，nsamples代表所有特征的数量，yi和
Figure PCTCN2017090580-appb-000012
分别代表每一个地产的真实价值和用上述房价评估模型计算得到的估计值。
以上不同评估标准分别从不同角度对模型进行了验证评估，描述了我们选择的特征以及模型的置信程度。以上几种评估标准可以选择使用，但是作用不同。其中，绝对偏差的均值反映了总体偏移程度的高低，但是会对极高价和极低价的评估出现偏差(例如别墅区、经济适用区)，绝对偏差的方差主 要是在均值相当的情况下，方差更低可以表明评估误差更小；绝对偏差的中位数考量了大多数情况下的偏差，但是总体偏差或许会很大，R2分值也叫拟合优度，反映的是预测值的方差和真实值方差的差别，越接近于1表示分类数据所来自的总体分布与预测分布相一致。
如图5所示，在一个实施例中，根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值的步骤308包括：
步骤308A，计算每一个配置信息与目标地理位置之间的距离。
步骤308B，根据每一个配置信息与目标地理位置之间的距离，采用sigmoid函数将确定的分数值进行投影得到标准化的特征值。
在本实施例中，当确定了每一个配置信息对应的分数值后，需要对该分数值按照预设的规则进行标准化，便于后续的计算。具体地，首先，计算每一个配置信息与目标地理位置之间的距离，然后根据计算得到距离采用sigmoid函数将确定的分数值进行投影得到标准化的特征值。该特征值便于后续根据预先建立的房价评估模型进行房产价格的评估。具体地，将sigmoid函数作为距离的一个系数因子：
Figure PCTCN2017090580-appb-000013
其中，x的值表示配置信息与目标地理位置之间的距离，d的值为预先设置的距离范围，比如设置d＝1000m。τ的值决定了因子关于距离的衰减程度，取值越大，衰减越慢。比如，取值τ＝20。通过该公式将配置信息对应的分数值进行投影得到标准化的特征值，即将配置信息对应的分数值乘以该系数因子就得到了该配置信息对应的标准化的特征值。
如图6所示，在一个实施例中，根据标准化的特征值采用房价评估模型确定与目标地理位置对应的房产价格估计值的步骤310包括：
步骤310A，根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的标准化的特征值。
在本实施例中，由于一个房产周围可能有很多配置信息，如果将每一个配置信息单独作为一个特征，不仅显得混乱，而且在建立房价评估模型时也会因为特征太多而难以拟合，所以可以根据配置信息的属性将配置信息进行分类，然后将每一类配置信息作为一个特征，确定每一类配置信息对应的标准化的特征值。比如，可以按照配置信息的属性将配置信息分为教育、医疗、交通、旅游、商业、生活服务等多种类型。比如，将小学、中学等统一归为教育，将三甲医院、二甲医院等统一归为医疗，将商业中心因子和商业美食因子等统一归为商业。举个例子，如果在房产周围总共有两个医院，一个是三甲医院，特征值为3，一个是二甲医院，特征值为2分，那么该房产周围的医疗特征对应的特征值为两者之和。
步骤310B，将确定的每一类配置信息对应的标准化的特征值代入房产评估模型确定与目标地理位置对应的房产价格估计值。
在本实施例中，将配置信息按照属性进行分类后，将确定的每一类配置信息对应的特征值代入房产评估模型，该房产评估模型根据输入的各个特征值对相应的房产价格进行估值，得到该房产的价格估计值。与之相应的，房产评估模型在训练学习时也是同样地将配置信息进行分类后进行训练的。
如图7所示，在一个实施例中，根据预设的评分标准确定每一个配置信息对应的分数值的步骤306包括：
步骤306A，获取每一个配置信息对应的属性信息。
在本实施例中，属性信息是指获取到的周边的配置设施本身所属的种类，比如，医院对应的属性信息为医疗，学校对应的属性信息为教育，商城对应的属性信息为商业，超市对应的属性信息为生活服务。获取每一个配置信息对应的属性信息是为了获取与该配置信息对应的评分标准。因为不同种类的配置信息评分标准肯定不同，比如，学校是按照排名来评分的，而医院可以采用评级来进行评分，而商场、超市则可以根据规模来评分。
步骤306B，获取与属性信息对应的评分标准。
在本实施例中，预先存储了属性信息与评分标准之间的对应关系，所以在获取到配置信息对应的属性信息后，就可以根据获取到的属性信息查找对应的评分标准。评分标准的设置是为了量化配置信息对房价的影响度。从而可根据确定的分数值来进行后续的房价评估。
步骤306C，根据评分标准确定与配置信息对应的分数值。
在本实施例中，在获取到与配置信息对应的评分标准后，就可以根据该评分标准来计算与配置信息对应的分数值，具体地，获取配置信息的特征来确定对应的分数值，比如，如果配置信息是医院，那么就获取该医院对应的评级及口碑，根据该评级和口碑来确定对应的分数值，比如，如果是三甲医院分数值为3，口碑若为优秀，则再加1分，如果一般，则不加分，如果差评，则相应的减1分。
如图8所示，在一个实施例中，提出了一种房产信息处理装置，该装置包括：
地理位置获取模块802，用于获取待评估的房产信息对应的目标地理位置。
配置信息获取模块804，用于以所述目标地理位置为中心，获取预设范围内的所有配置信息。
分数值确定模块806，用于根据预设的评分标准确定每一个配置信息对应的分数值。
特征值确定模块808，用于根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值。
房产价格确定模块810，用于根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
如图9所示，在一个实施例中，上述房产信息处理装置还包括：
建立模块812，用于建立初始化的房价评估模型。
模型参数确定模块814，用于根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数。
模型确定模块816，用于根据确定的所述模型参数得到房价评估模型。
在一个实施例中，特征值确定模块808还用于计算每一个配置信息与所述目标地理位置之间的距离，根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
如图10所示，在一个实施例中，房产价格确定模块810包括：
分类模块810A，用于根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应标准化的特征值。
估计值确定模块810B，用于将确定的每一类配置信息对应的标准化的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
在一个实施例中，所述分数值确定模块806还用于获取每一个配置信息对应的属性信息，获取与所述属性信息对应的评分标准，根据所述评分标准确定与所述配置信息对应的分数值。
上述房产信息处理装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。其中，网络接口可以是以太网卡或无线网卡等。上述各模块可以硬件形式内嵌于或独立于服务器中的处理器中，也可以以软件形式存储于服务器中的存储器中，以便于处理器调用执行以上各个模块对应的操作。该处理器可以为中央处理单元(CPU)、微处理器、单片机等。
在一个实施例中，提出了一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步骤：获取待评估的房产信息对应的目标地理位置；以所述目标地理位置为中心，获取预设范围内的所有配置信息；根据预设的评分标准确定每一个配置信息对应的分数值；根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值； 根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
在一个实施例中，在所述获取待评估的房产信息对应的目标地理位置的步骤之前，所述处理器执行所述计算机程序是还用于实现以下步骤：建立初始化的房价评估模型；根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；根据确定的所述模型参数得到房价评估模型。
在一个实施例中，所述处理器所执行的根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值，包括：计算每一个配置信息与所述目标地理位置之间的距离；根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
在一个实施例中，所述处理器所执行的所述根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值，包括：根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的特征值；将确定的每一类配置信息对应的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
在一个实施例中，所述处理器所执行的所述根据预设的评分标准确定每一个配置信息对应的分数值的步骤包括：获取每一个配置信息对应的属性信息；获取与所述属性信息对应的评分标准；根据所述评分标准确定与所述配置信息对应的分数值。
在一个实施例中，一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：获取待评估的房产信息对应的目标地理位置；以所述目标地理位置为中心，获取预设范围内的所有配置信息；根据预设的评分标准确定每一个配置信息对应的分数值；根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征 值；根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
在一个实施例中，在所述获取待评估的房产信息对应的目标地理位置的步骤之前，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器还执行以下步骤：建立初始化的房价评估模型；根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；根据确定的所述模型参数得到房价评估模型。
在一个实施例中，所述一个或多个处理器所执行的根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值，包括：计算每一个配置信息与所述目标地理位置之间的距离；根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
在一个实施例中，所述一个或多个处理器所执行的所述根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值，包括：根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的特征值；将确定的每一类配置信息对应的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
在一个实施例中，所述一个或多个处理器所执行的所述根据预设的评分标准确定每一个配置信息对应的分数值的步骤包括：获取每一个配置信息对应的属性信息；获取与所述属性信息对应的评分标准；根据所述评分标准确定与所述配置信息对应的分数值。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，该计算机程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，前述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)等非易失性存储介质，或随机存储记忆体(Random Access Memory，RAM)等。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本发明的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本发明构思的前提下，还可以做出若干变形和改进，这些都属于本发明的保护范围。因此，本发明专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种房产信息处理方法，包括：
    获取待评估的房产信息对应的目标地理位置；
    以所述目标地理位置为中心，获取预设范围内的所有配置信息；
    根据预设的评分标准确定每一个配置信息对应的分数值；
    根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
    根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
  2. 根据权利要求1所述的方法，其特征在于，在所述获取待评估的房产信息对应的目标地理位置之前，所述方法还包括：
    建立初始化的房价评估模型；
    根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；
    根据确定的所述模型参数得到房价评估模型。
  3. 根据权利要求1所述的方法，其特征在于，根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值包括：
    计算每一个配置信息与所述目标地理位置之间的距离；
    根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
  4. 根据权利要求1所述的方法，其特征在于，所述根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值包括：
    根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的标准化的特征值；
    将确定的每一类配置信息对应的标准化的特征值代入所述房产评估模型 确定与所述目标地理位置对应的房产价格估计值。
  5. 根据权利要求1所述的方法，其特征在于，所述根据预设的评分标准确定每一个配置信息对应的分数值包括：
    获取每一个配置信息对应的属性信息；
    获取与所述属性信息对应的评分标准；
    根据所述评分标准确定与所述配置信息对应的分数值。
  6. 一种房产信息处理装置，包括：
    地理位置获取模块，用于获取待评估的房产信息对应的目标地理位置；
    配置信息获取模块，用于以所述目标地理位置为中心，获取预设范围内的所有配置信息；
    分数值确定模块，用于根据预设的评分标准确定每一个配置信息对应的分数值；
    特征值确定模块，用于根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
    房产价格确定模块，用于根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
  7. 根据权利要求6所述的装置，其特征在于，还包括：
    建立模块，用于建立初始化的房价评估模型；
    模型参数确定模块，用于根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；
    模型确定模块，用于根据确定的所述模型参数得到房价评估模型。
  8. 根据权利要求6所述的装置，其特征在于，所述特征值确定模块还用于计算每一个配置信息与所述目标地理位置之间的距离，根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
  9. 根据权利要求6所述的装置，其特征在于，所述房产价格确定模块包括：
    分类模块，用于根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的标准化的特征值；
    估计值确定模块，用于将确定的每一类配置信息对应的标准化的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
  10. 根据权利要求6所述的装置，其特征在于，所述分数值确定模块还用于获取每一个配置信息对应的属性信息，获取与所述属性信息对应的评分标准，根据所述评分标准确定与所述配置信息对应的分数值。
  11. 一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行以下步骤：
    获取待评估的房产信息对应的目标地理位置；
    以所述目标地理位置为中心，获取预设范围内的所有配置信息；
    根据预设的评分标准确定每一个配置信息对应的分数值；
    根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
    根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
  12. 根据权利要求11所述的计算机设备，其特征在于，在所述获取待评估的房产信息对应的目标地理位置之前，所述处理器还用于执行以下步骤：
    建立初始化的房价评估模型；
    根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；
    根据确定的所述模型参数得到房价评估模型。
  13. 根据权利要求11所述的计算机设备，其特征在于，所述根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值包括：
    计算每一个配置信息与所述目标地理位置之间的距离；
    根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
  14. 根据权利要求11所述的计算机设备，其特征在于，所述根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值包括：
    根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的标准化的特征值；
    将确定的每一类配置信息对应的标准化的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
  15. 根据权利要求11所述的计算机设备，其特征在于，所述根据预设的评分标准确定每一个配置信息对应的分数值包括：
    获取每一个配置信息对应的属性信息；
    获取与所述属性信息对应的评分标准；
    根据所述评分标准确定与所述配置信息对应的分数值。
  16. 一个或多个存储有计算机可读指令的非易失性可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取待评估的房产信息对应的目标地理位置；
    以所述目标地理位置为中心，获取预设范围内的所有配置信息；
    根据预设的评分标准确定每一个配置信息对应的分数值；
    根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值；及
    根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值。
  17. 根据权利要求16所述的非易失性可读存储介质，其特征在于，在所述获取待评估的房产信息对应的目标地理位置之前，所述处理器还用于执行以下步骤：
    建立初始化的房价评估模型；
    根据采集到的房产价格和该房产周围的配置信息对所述初始化的房价评估模型进行训练确定相应的模型参数；
    根据确定的所述模型参数得到房价评估模型。
  18. 根据权利要求16所述的非易失性可读存储介质，其特征在于，所述处理器所执行的所述根据每一个配置信息与所述目标地理位置之间的距离将确定的所述分数值进行投影得到标准化的特征值的步骤包括：
    计算每一个配置信息与所述目标地理位置之间的距离；
    根据每一个配置信息与所述目标地理位置之间的距离，采用sigmoid函数将确定的所述分数值进行投影得到标准化的特征值。
  19. 根据权利要求16所述的非易失性可读存储介质，其特征在于，所述处理器所执行的所述根据所述标准化的特征值采用房价评估模型确定与所述目标地理位置对应的房产价格估计值的步骤包括：
    根据配置信息的属性将配置信息进行分类，确定每一类配置信息对应的标准化的特征值；
    将确定的每一类配置信息对应的标准化的特征值代入所述房产评估模型确定与所述目标地理位置对应的房产价格估计值。
  20. 根据权利要求16所述的非易失性可读存储介质，其特征在于，所述处理器所执行的所述根据预设的评分标准确定每一个配置信息对应的分数值的步骤包括：
    获取每一个配置信息对应的属性信息；
    获取与所述属性信息对应的评分标准；
    根据所述评分标准确定与所述配置信息对应的分数值。
PCT/CN2017/090580 2017-05-10 2017-06-28 房产信息处理方法、装置、计算机设备及存储介质 Ceased WO2018205372A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/313,793 US20190259071A1 (en) 2017-05-10 2017-06-28 Property information processing method and apparatus, computer device and storage medium
SG11201810669VA SG11201810669VA (en) 2017-05-10 2017-06-28 Property information processing method and apparatus, computer device and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710326478.7A CN107239967A (zh) 2017-05-10 2017-05-10 房产信息处理方法、装置、计算机设备及存储介质
CN201710326478.7 2017-05-10

Publications (1)

Publication Number Publication Date
WO2018205372A1 true WO2018205372A1 (zh) 2018-11-15

Family

ID=59985494

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/090580 Ceased WO2018205372A1 (zh) 2017-05-10 2017-06-28 房产信息处理方法、装置、计算机设备及存储介质

Country Status (4)

Country Link
US (1) US20190259071A1 (zh)
CN (1) CN107239967A (zh)
SG (1) SG11201810669VA (zh)
WO (1) WO2018205372A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109816454A (zh) * 2019-01-31 2019-05-28 拉扎斯网络科技（上海）有限公司 确定用于商户的策略的方法、装置、电子设备及存储介质

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108062729A (zh) * 2017-12-12 2018-05-22 链家网（北京）科技有限公司 一种房源限价方法
CN108304485B (zh) * 2017-12-29 2020-08-25 重庆金融资产交易所有限责任公司 政务数据图形化处理方法、装置、存储介质和计算机设备
CN110163705B (zh) * 2018-02-13 2024-09-24 北京京东尚科信息技术有限公司 用于推送信息的方法和装置
US11348170B2 (en) 2018-03-27 2022-05-31 Allstate Insurance Company Systems and methods for identifying and transferring digital assets
US11748817B2 (en) 2018-03-27 2023-09-05 Allstate Insurance Company Systems and methods for generating an assessment of safety parameters using sensors and sensor data
CN110737841A (zh) * 2018-07-03 2020-01-31 百度在线网络技术（北京）有限公司 用于生成信息的方法和装置
CN109376287B (zh) * 2018-09-21 2023-09-01 平安科技（深圳）有限公司 房产图谱构建方法、装置、计算机设备及存储介质
CN111380558B (zh) * 2018-12-29 2022-02-11 北京四维图新科技股份有限公司 兴趣点的排序方法、设备、服务器及存储介质
CN110322114A (zh) * 2019-05-23 2019-10-11 平安城市建设科技(深圳)有限公司 基于大数据的小区推荐方法、装置、设备及存储介质
US20200402116A1 (en) * 2019-06-19 2020-12-24 Reali Inc. System, method, computer program product or platform for efficient real estate value estimation and/or optimization
CN110399569A (zh) * 2019-07-19 2019-11-01 银联智策顾问（上海）有限公司 一种基于大数据评估土地价值的方法及评估装置
CN111523614B (zh) * 2020-05-09 2023-12-19 上海添玑网络服务有限公司 小区相似判断方法及装置
CN112085388A (zh) * 2020-09-10 2020-12-15 深圳禾略贝塔信息科技有限公司 一种土地价值评估方法、装置、终端以及可读存储介质
CN113344660B (zh) * 2021-05-28 2024-12-20 深圳市前海房极客网络科技有限公司 房源信息的处理方法、装置、电子设备及存储介质
CN114723338A (zh) * 2022-05-20 2022-07-08 中国工商银行股份有限公司 资源数据的评估方法、装置和计算机设备

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104751007A (zh) * 2015-04-16 2015-07-01 百度在线网络技术（北京）有限公司 基于建筑物价值评估的计算方法和装置
CN105321131A (zh) * 2015-11-06 2016-02-10 厦门均达网络科技有限公司 一种房地产价值评估系统及评估方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104751007A (zh) * 2015-04-16 2015-07-01 百度在线网络技术（北京）有限公司 基于建筑物价值评估的计算方法和装置
CN105321131A (zh) * 2015-11-06 2016-02-10 厦门均达网络科技有限公司 一种房地产价值评估系统及评估方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109816454A (zh) * 2019-01-31 2019-05-28 拉扎斯网络科技（上海）有限公司 确定用于商户的策略的方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
SG11201810669VA (en) 2018-12-28
CN107239967A (zh) 2017-10-10
US20190259071A1 (en) 2019-08-22

Similar Documents

Publication Publication Date Title
WO2018205372A1 (zh) 房产信息处理方法、装置、计算机设备及存储介质
US11068285B2 (en) Machine-learning models applied to interaction data for determining interaction goals and facilitating experience-based modifications to interface elements in online environments
CN109074368B (zh) 使用异构社交网络的图形框架
EP3965050A1 (en) Systems and methods for deriving rating for properties
Debarsy et al. Bayesian model averaging for spatial autoregressive models based on convex combinations of different types of connectivity matrices
Cabrerizo et al. A decision support system to develop a quality management in academic digital libraries
WO2021203728A1 (zh) 一种展业区域的选址方法、装置、计算机设备和介质
WO2020077824A1 (zh) 异常问题的定位方法、装置、设备及存储介质
Zhang et al. CRITID: enhancing CRITIC with advanced independence testing for robust multi-criteria decision-making
CN111311316B (zh) 商户画像的刻画方法、装置、电子设备、验证方法及系统
WO2021121296A1 (zh) 习题测试数据生成方法以及装置
CN115660292B (zh) 基于餐饮消费数据处理的碳排放监测方法及设备
CN110515929B (zh) 书籍展示方法、计算设备及存储介质
Jung et al. Two-step approach to partial factorial invariance: Selecting a reference variable and identifying the source of noninvariance
Lee et al. Controlling for localised spatio-temporal autocorrelation in long-term air pollution and health studies
CN110070380A (zh) 信息生成方法和装置
CN111626767A (zh) 资源数据的发放方法、装置及设备
Hu et al. A backfitting maximum likelihood estimator for hierarchical and geographically weighted regression modelling, with a case study of house prices in Beijing
Zandbergen Influence of street reference data on geocoding quality
Li et al. Lasso-type estimation for covariate-adjusted linear model
CN110909258A (zh) 一种信息推荐方法、装置、设备及存储介质
Ryu Multiple group analysis in multilevel structural equation model across level 1 groups
CN113391989B (zh) 程序评估方法、装置、设备、介质及程序产品
CN110263250A (zh) 一种推荐模型的生成方法及装置
CN106779181A (zh) 一种基于线性回归因子非负矩阵分解模型的医疗机构推荐方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17909621

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17909621

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 13.03.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17909621

Country of ref document: EP

Kind code of ref document: A1