WO2018120425A1 - 个人财产状态评估方法、装置、设备和存储介质 - Google Patents

个人财产状态评估方法、装置、设备和存储介质 Download PDF

Info

Publication number
WO2018120425A1
WO2018120425A1 PCT/CN2017/076463 CN2017076463W WO2018120425A1 WO 2018120425 A1 WO2018120425 A1 WO 2018120425A1 CN 2017076463 W CN2017076463 W CN 2017076463W WO 2018120425 A1 WO2018120425 A1 WO 2018120425A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
personal property
property
seed
data
Prior art date
Application number
PCT/CN2017/076463
Other languages
English (en)
French (fr)
Inventor
毕野
王建明
肖京
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2018120425A1 publication Critical patent/WO2018120425A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08—Insurance
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences

Definitions

  • the present invention relates to the field of information processing technologies, and in particular, to a method, device, device and storage medium for evaluating personal property status.
  • a financial institution When a financial institution provides customers with services such as household consumption management, financial planning management, asset allocation management, and investment management, it is necessary to comprehensively evaluate the customer's personal property status, and organize and analyze the personal property status to timely identify the customer's financial risks. To correct bad financial habits and improve their ability to resist financial risks.
  • the existing financial institutions mainly use the financial data of the customer's asset status and consumption flow to evaluate the personal property status, and the evaluation data source is single, resulting in a low accuracy rate of the personal property status assessment result.
  • a financial institution provides services to customers based on the results of personal property assessments with low accuracy, it provides services that do not match the status of their personal assets, which may lead to financial risks.
  • the invention provides a method, a device, a device and a storage medium for evaluating a personal property state, so as to solve the problem that the data source is single and the accuracy rate of the personal property state evaluation result is low when the personal property state evaluation in the prior art is evaluated.
  • the present invention provides a method for assessing a state of personal property, comprising:
  • geographical location fence information of the target area where the geographical location fence information includes house information and a corresponding target user;
  • All target users are divided into a seed data set and a candidate data set; the seed data set includes at least one seed user, the candidate data set including at least one candidate user;
  • the present invention provides a personal property status assessment apparatus, including:
  • a fence information obtaining module configured to acquire geographical location fence information of the target area, where the geographical location fence information includes the house information and the corresponding target user;
  • An image data obtaining module configured to acquire user portrait data associated with the target user
  • a data set dividing module configured to divide all target users into a seed data set and a candidate data set; the seed data set includes at least one seed user, and the candidate data set includes at least one candidate user;
  • An evaluation model training module configured to train the property evaluation model by using user image data of the seed user
  • the property state evaluation module is configured to evaluate the personal property status of the candidate user by using the property evaluation model according to the user image data of the candidate user, to output a personal property state evaluation result.
  • the present invention also provides a personal property status evaluation device comprising a processor and a memory, the memory storing computer executable instructions, the processor for executing the computer executable instructions to perform the following steps:
  • geographical location fence information of the target area where the geographical location fence information includes house information and a corresponding target user;
  • All target users are divided into a seed data set and a candidate data set; the seed data set includes at least one seed user, the candidate data set including at least one candidate user;
  • the present invention also provides a non-transitory computer readable storage medium for storing one or more computer executable instructions, the computer executable instructions being executed by one or more processors such that The one or more processors perform the personal property status assessment method of any of the ones.
  • the present invention has the following advantages: in the personal property state evaluation method, apparatus, device and storage medium provided by the present invention, the geographical location fence information (including the target user) of the target area is first acquired, and the target is acquired.
  • User image data associated with the user dividing the target user into seed users and candidate users; training the property evaluation model with the user image data of the seed user, and processing the user image data of the candidate user by using the trained property evaluation model, and outputting
  • the personal property assessment result of the candidate user is used to train the property evaluation model
  • the trained property evaluation model is used to evaluate the personal property of the candidate users in the target area, so that the personal property status evaluation result has high accuracy, objectivity and reliability. Sex.
  • FIG. 1 is a flow chart showing a method for evaluating a personal property state in a first embodiment of the present invention
  • Figure 2 is a schematic block diagram of a personal property state evaluation apparatus in a second embodiment of the present invention.
  • Figure 3 is a schematic diagram of a personal property status evaluation apparatus in a third embodiment of the present invention.
  • Fig. 1 is a flow chart showing a method of personal property status evaluation in the present embodiment.
  • the personal property status assessment method can be applied to a personal property status assessment device of a financial institution such as a bank or insurance to evaluate the personal property status of any user.
  • the personal property status evaluation method includes the following steps:
  • S10 Obtain geographical location fence information of the target area, where the geographical location fence information includes the housing information and the corresponding target user.
  • the target area may be any residential area.
  • the housing information may be information such as the location of the house, the house number, the size of the house, the average price of the house, and the average rent of the house in the target area (residential area).
  • the target user may be the owner of the house corresponding to the house information.
  • the geographic location fence information of the target area is obtained to obtain the housing information of each house in any residential cell and the corresponding target user. Since the target user lives in the same residential area, the personal property status has a certain value. Similarity, so that the target user determined based on the geographical location fence information performs personal property status assessment.
  • the residential area corresponding to the target area is preferably a residential area with a higher average selling price of the house, and the average selling price of the house is super high, and the corresponding house owner (ie, the target user) should have a corresponding personal property status.
  • step S10 specifically includes: using a web crawler to crawl a real estate intermediary platform and/or a real estate registration platform to obtain geographic location fence information of the target area.
  • the web crawler automatically captures the housing information in the real estate intermediary platform and/or the real estate registration platform according to a preset rule, and acquires the target user associated with the housing information, and uses the housing information and the target user as the geographical location information of the target area. Output.
  • the real estate intermediary platform and/or the real estate registration platform stores the housing information and the target user of any housing in the target area, and uses the web crawler to crawl the geographical location fence information of the target area from the real estate intermediary platform and/or the real estate registration platform. Crawling data content is clear and processing speed is faster.
  • the target user is a house owner corresponding to any house information in the target area
  • the personal property status of the target user in the same target area has a certain degree of similarity. Understandably, the sale of the home with the target area Other regions with the same or similar price can also serve as the same target region to expand the target user range of the target region.
  • the personal property assessment based on the target users in the target area can improve the accuracy and reliability of the personal property status assessment results to a certain extent.
  • S20 Acquire user portrait data associated with the target user.
  • the user portrait data (ie, Persona data) is a virtual representation of the real user, and is a target user model built on the Marketing Data/Usability Data.
  • the personal property status evaluation device of a financial institution such as a bank or insurance stores user image data of the target user, and the user image data includes, but is not limited to, a user name, an identification feature, a photo, a contact information, a home address, an office place, a job, and Income, etc.
  • the user corresponding to each user portrait data is associated with the housing information in the geographic information fence information and the target user, so that the user portrait data of each target user is obtained.
  • Each includes housing information of the target area for processing based on user portrait data of the target user related to the housing information,
  • the user portrait data includes location information based on location services, and the geographic location information includes POI information associated with time.
  • the geographical location information includes POI information of 0:00-24:00, and each POI information is used to indicate a point in the electronic map, including POI point name, longitude and latitude, and the like. information.
  • the user profile data can also include consumer features, investment features, or other characteristics that affect personal property assessment factors.
  • the target user frequently enters or exits a high-end consumer place, or has a large investment record in a financial institution, the target user's personal property is low, and the personal property evaluation result obtained is high to ensure personal property.
  • the location information based on the location service is the daily trajectory of the target user, and has objectivity.
  • the personal property status assessment based on the geographical location information can improve the objectivity and accuracy of the personal property status assessment result, and avoid the target user only.
  • the self-provided information is subjective to the personal property assessment, and the accuracy of the assessment results is low.
  • the location based service is to obtain the location information of the mobile terminal (ie, the target user) through the telecommunication mobile operator's radio communication network (such as GSM network, CDMA network) or external positioning mode (such as GPS). (Geographical coordinates, or geodetic coordinates), a value-added service that provides corresponding services to target users under the support of the Geographic Information System (GIS) platform.
  • GIS Geographic Information System
  • LBS is a combination of a mobile communication network and a computer network, and the two networks interact through a gateway.
  • the mobile terminal sends a request through the mobile communication network and transmits it to the LBS service platform through the gateway; the LBS service platform processes the target user request and the current location of the target user, and returns the result to the target user through the gateway.
  • POI Point Of Interest, impromptu Interest points or information points
  • the POI can be presented on an electronic map to indicate a certain landmark, attraction and other location information on the electronic map.
  • the location service-based mobile terminal is a smart phone
  • the location function of the smart phone is enabled by enabling the LBS service platform to obtain the geographical location information of the smart phone in real time, thereby understanding the geographical location of the target user carrying the smart phone. information.
  • the LSB service platform is connected to the personal property status evaluation device, so that the personal property status assessment device acquires the geographical location information of the target user and stores it as corresponding user portrait data.
  • the location information includes time and time in the POI information associated with the time, by which the POI information of the target user at any time can be known. It can be understood that the geographical location information is associated with the user ID of the target user, and the user ID is used to identify the uniquely identified user, which may be an ID number or a mobile phone number.
  • All target users are divided into a seed data set and a candidate data set; the seed data set includes at least one seed user, and the candidate data set includes at least one candidate user.
  • all target users in the target area are divided into seed data sets and candidate data sets according to whether the personal property assessment has been performed and the assessed personal property status is used as a division condition.
  • each seed user in the seed data set has the assessed personal property status.
  • Each candidate user in the candidate data set does not have an assessed personal property status.
  • S40 The property evaluation model is trained using the user image data of the seed user.
  • the user portrait data includes but is not limited to user name, identification feature, photo, contact information, family Address, office space, occupation, and income, etc., also includes location-based geographic location data that reflects the trajectory of the target user's daily life.
  • a commonality analysis is performed on the user portrait data of all seed users, and the relationship between the portrait data of the seed user and the state of the individual property is obtained to train the property evaluation model.
  • each seed user is the target user of the target area
  • the property evaluation model is trained by using the user image data of the seed user, which can improve the accuracy and reliability of the personal property status evaluation result to a certain extent.
  • the user portrait data includes geographic location data for embodying the trajectory of the daily life of the seed user, and has objectivity. The training of the property evaluation model based on the geographical location information may be beneficial to improving the objectivity and accuracy of the personal property status assessment result.
  • step S40 includes the following steps:
  • S41 The user image data of all seed users is classified by using a look-alike algorithm, and the common image data corresponding to each sub-cluster and each of the sub-clusters is obtained.
  • Look-alike which is similar population expansion, is a technology based on existing user/device ID, through a certain algorithm evaluation model, to find more similar groups with potential relevance.
  • the look-alike algorithm used in this embodiment The user image data of the seed user is used as a positive sample, and the classification model is trained to obtain the common image data, so that the user image data of the candidate user is used as a negative sample, and the classification model is used for screening.
  • a classification method based on PU-Learning Learning from Positive and UnLabled Example
  • the process is simple and convenient, which can effectively reduce the preparatory workload of manual classification and improve the classification accuracy.
  • PU-Learning Learning from Positive and UnLabled Example
  • the user portrait data of each seed user includes location service-based geographic location data reflecting the trajectory of the daily life of the target user
  • a common portrait in each sub-cluster obtained by classifying the user portrait data of all seed users by using the look-alike algorithm is used.
  • Data is associated with location-based geographic location data with objectivity and reliability.
  • S42 Acquire the personal property status of each seed user, and calculate the personal property average value of all seed users in each sub-cluster.
  • the seed user in each sub-cluster obtained by classifying the user portrait data of all seed users by using the look-alike algorithm also has the evaluated individual. Property status.
  • the average value of the personal property of all seed users in each sub-cluster is calculated, and the property evaluation model is constructed by using the average value of the individual assets.
  • S43 Perform logical regression processing on the common portrait data and the personal property mean value in each sub-cluster to obtain a property evaluation model.
  • the common portrait data of each sub-cluster and the average value of the personal assets of the sub-cluster are logically regression processed by a logistic regression algorithm to obtain a property evaluation model.
  • the common portrait data in the sub-cluster is mapped to the mean value of the individual assets.
  • the common image data is related to the geographical location data based on location service, which has objectivity and reliability, and the property evaluation model formed by it has objectivity and reliability.
  • Logistic Regression is a commonly used machine learning method in the industry to estimate the possibility of something.
  • ⁇ is the model parameter, that is, the regression coefficient
  • ⁇ is the sigmoid function.
  • This function is under The logarithmic probability of the face (that is, the logarithm of the ratio of the likelihood that x belongs to the positive class and the likelihood of the negative class) is transformed:
  • the property evaluation model is used to evaluate the personal property status of the candidate user to output the personal property status evaluation result.
  • the user portrait data is mapped to the personal property status.
  • the user image data of the candidate user can be input into the trained property evaluation model for personal property status evaluation, and the personal property status evaluation result can be output, and the evaluation process is simple and convenient.
  • the output of the personal property status assessment results is objective and accurate, and the evaluation process is simple and convenient.
  • step S50 includes the following steps:
  • S51 The similarity algorithm is used to calculate the similarity between the user image data of the candidate user and the common portrait data of each sub-cluster.
  • the text similarity algorithm is used to calculate the similarity between the user portrait data of the candidate user and the shared portrait data of each sub-cluster.
  • the use of the text similarity algorithm to calculate the similarity includes the following process: First, the user image data of the candidate user is subjected to pre-processing such as word segmentation and de-stopping. Text feature extraction and weighting are then performed based on TF-IDF or other weights. Finally, the vector space model VSM is used to calculate the cosine value to calculate the similarity between the user image data of the candidate user and the common image data of each sub-cluster.
  • TF Term frequency
  • IDF Inverse document frequency
  • the text similarity algorithm is used to calculate the similarity, which has the advantages of simple calculation process and fast calculation speed. It can be understood that a similarity algorithm such as a semantic similarity-based text similarity algorithm and a pinyin similarity-based Chinese fuzzy search algorithm can also be used for processing.
  • S52 Determine whether the similarity is greater than a preset similarity threshold.
  • the similarity threshold is preset, and is used to determine the value of the candidate user belonging to any sub-cluster, and can be set autonomously.
  • the similarity threshold is set to 70%. That is, when the similarity between the user image data of the candidate user and the shared portrait data of a sub-cluster is greater than the similarity threshold (70%), the candidate user is considered to be attributable to the sub-cluster.
  • the candidate user is considered to be attributable to the sub-cluster, and the average value of the personal property corresponding to the sub-cluster is used as the candidate user.
  • the personal property status evaluation result of any candidate user is associated with the user portrait data, and the user portrait data includes the location information based on the location service, and has an objective Sex and reliability.
  • the geographical location fence information (including the target user) of the target area is acquired first, and the user portrait data associated with the target user is acquired; the target user is divided into the seed user and the candidate user;
  • the user evaluation data of the seed user is used to train the property evaluation model, and the user image data of the candidate user is processed by the trained property evaluation model, and the personal property evaluation result of the candidate user is output.
  • the user image data of the seed user in the target area is used to train the property evaluation model, and the trained property evaluation model is used to evaluate the personal property of the candidate users in the target area.
  • the evaluation process is simple and convenient, and the output personal property status evaluation result has a high Accuracy, objectivity and reliability.
  • Fig. 2 is a flow chart showing the personal property state evaluation device in the embodiment.
  • the personal property status assessment device may be a personal property status assessment device applied in a financial institution such as a bank or insurance, and may be used to evaluate the personal property status of any user.
  • the personal property status evaluation device includes a fence information acquisition module 10, an image data acquisition module 20, a data set division module 30, an evaluation model training module 40, and a property status evaluation module 50.
  • the fence information obtaining module 10 is configured to acquire the geographical location fence information of the target area, where the geographical location fence information includes the house information and the corresponding target user.
  • the target area may be any residential area.
  • the housing information may be information such as the location of the house, the house number, the size of the house, the average price of the house, and the average rent of the house in the target area (residential area).
  • the target user may be the owner of the house corresponding to the house information.
  • the geographic location fence information of the target area is obtained to obtain the housing information of each house in any residential cell and the corresponding target user. Since the target user lives in the same residential area, the personal property status has a certain value. Similarity, so that the target user determined based on the geographical location fence information performs personal property status assessment.
  • the residential area corresponding to the target area is preferably a residential area with a higher average selling price of the house, and the average selling price of the house is super high, and the corresponding house owner (ie, the target user) should have a corresponding personal property status.
  • the fence information obtaining module 10 is configured to use a web crawler to crawl the real estate intermediary platform and/or the real estate registration platform to obtain the geographical location fence information of the target area.
  • the web crawler automatically captures the housing information in the real estate intermediary platform and/or the real estate registration platform according to a preset rule, and acquires the target user associated with the housing information, and uses the housing information and the target user as the geographical location information of the target area. Output.
  • the real estate intermediary platform and/or the real estate registration platform stores the housing information and the target user of any housing in the target area, and uses the web crawler to crawl the geographical location fence information of the target area from the real estate intermediary platform and/or the real estate registration platform. Crawling data content is clear and processing speed is faster.
  • the target user is the owner of the house corresponding to any house information in the target area, and the same target
  • the personal property status of the target users of the area has a certain degree of similarity. It can be understood that other areas that are the same or similar to the average sales price of the target area can also serve as the same target area to expand the target user range of the target area.
  • the personal property assessment based on the target users in the target area can improve the accuracy and reliability of the personal property status assessment results to a certain extent.
  • the image data obtaining module 20 is configured to acquire user portrait data associated with the target user.
  • the user portrait data (ie, Persona data) is a virtual representation of the real user, and is a target user model built on the Marketing Data/Usability Data.
  • the personal property status evaluation device of a financial institution such as a bank or insurance stores user image data of the target user.
  • the user portrait data includes, but is not limited to, user name, identification characteristics, photos, contact information, home address, office space, occupation, and income.
  • the user corresponding to each user portrait data is associated with the housing information in the geographic information fence information and the target user, so that the user portrait data of each target user is obtained.
  • Each includes housing information of the target area for processing based on user portrait data of the target user related to the housing information,
  • the user portrait data includes location information based on location services, and the geographic location information includes POI information associated with time.
  • the geographical location information includes POI information of 0:00-24:00, and each POI information is used to indicate a point in the electronic map, including POI point name, longitude and latitude, and the like. information.
  • the user profile data can also include consumer features, investment features, or other characteristics that affect personal property assessment factors.
  • the target user frequently enters or exits a high-end consumer place, or has a large investment record in a financial institution, the target user's personal property is low, and the personal property evaluation result obtained is high to ensure personal property.
  • the location information based on the location service is the daily trajectory of the target user, and has objectivity.
  • the personal property status assessment based on the geographical location information can improve the objectivity and accuracy of the personal property status assessment result, and avoid the target user only.
  • the self-provided information is subjective to the personal property assessment, and the accuracy of the assessment results is low.
  • the location based service is to obtain the location information of the mobile terminal (ie, the target user) through the telecommunication mobile operator's radio communication network (such as GSM network, CDMA network) or external positioning mode (such as GPS). (Geographical coordinates, or geodetic coordinates), a value-added service that provides corresponding services to target users under the support of the Geographic Information System (GIS) platform.
  • LBS is a combination of a mobile communication network and a computer network, and the two networks interact through a gateway.
  • the mobile terminal sends a request through the mobile communication network and transmits it to the LBS service platform through the gateway; the LBS service platform uses the target user request and target according to the target
  • the current location of the user is processed and the result is returned to the target user through the gateway.
  • POI Point Of Interest
  • the POI can be presented on the electronic map to indicate a certain landmark, attraction and other location information on the electronic map.
  • the location service-based mobile terminal is a smart phone
  • the location function of the smart phone is enabled by enabling the LBS service platform to obtain the geographical location information of the smart phone in real time, thereby understanding the geographical location of the target user carrying the smart phone. information.
  • the LSB service platform is connected to the personal property status evaluation device, so that the personal property status assessment device acquires the geographical location information of the target user and stores it as corresponding user portrait data.
  • the location information includes time and time in the POI information associated with the time, by which the POI information of the target user at any time can be known. It can be understood that the geographical location information is associated with the user ID of the target user, and the user ID is used to identify the uniquely identified user, which may be an ID number or a mobile phone number.
  • the data set dividing module 30 is configured to divide all target users into a seed data set and a candidate data set; the seed data set includes at least one seed user, and the candidate data set includes at least one candidate user.
  • all target users in the target area are divided into seed data sets and candidate data sets according to whether the personal property assessment has been performed and the assessed personal property status is used as a division condition.
  • each seed user in the seed data set has the assessed personal property status.
  • Each candidate user in the candidate data set does not have an assessed personal property status.
  • the evaluation model training module 40 is configured to train the property evaluation model using the user image data of the seed user.
  • the user portrait data includes but is not limited to user name, identification feature, photo, contact information, family Address, office space, occupation, and income, etc., also includes location-based geographic location data that reflects the trajectory of the target user's daily life.
  • a commonality analysis is performed on the user portrait data of all seed users, and the relationship between the portrait data of the seed user and the state of the individual property is obtained to train the property evaluation model.
  • each seed user is the target user of the target area
  • the property evaluation model is trained by using the user image data of the seed user, which can improve the accuracy and reliability of the personal property status evaluation result to a certain extent.
  • the user portrait data includes geographic location data for embodying the trajectory of the daily life of the seed user, and has objectivity. The training of the property evaluation model based on the geographical location information may be beneficial to improving the objectivity and accuracy of the personal property status assessment result.
  • the evaluation model training module 40 specifically includes an image data classification unit 41, a property average calculation unit 42, and an evaluation model processing unit 43.
  • the image data classification unit 41 is configured to classify user image data of all seed users by using a look-alike algorithm, and acquire common image data corresponding to each of the sub-clusters and each of the sub-clusters.
  • Look-alike which is similar population expansion, is a technology based on existing user/device ID, through a certain algorithm evaluation model, to find more similar groups with potential relevance.
  • the user image data of the seed user is used as a positive sample, and the classification model is trained to obtain the common image data, so that the user image data of the candidate user is used as a negative sample, and the classification model is used for screening.
  • the look-alike algorithm is used to classify the user portrait data of all seed users by using a classification method based on PU-Learning (Learning from Positive and UnLabled Example), and the classification process is performed. Simple and convenient, it can effectively reduce the preparatory workload of manual classification and improve classification accuracy. It can be understood that the user portrait data of all seed users is classified by the look-alike algorithm, and each sub-cluster obtained has the same common portrait data, which is an associated feature that can be used to evaluate the state of personal property.
  • PU-Learning Learning from Positive and UnLabled Example
  • the user portrait data of each seed user includes location service-based geographic location data reflecting the trajectory of the daily life of the target user
  • a common portrait in each sub-cluster obtained by classifying the user portrait data of all seed users by using the look-alike algorithm is used.
  • Data is associated with location-based geographic location data with objectivity and reliability.
  • the property mean value calculation unit 42 is configured to acquire the personal property status of each seed user, and calculate the personal property average value of all seed users in each sub-cluster.
  • the seed user in each sub-cluster obtained by classifying the user portrait data of all seed users by using the look-alike algorithm also has the evaluated individual. Property status.
  • the average value of the personal property of all seed users in each sub-cluster is calculated, and the property evaluation model is constructed by using the average value of the individual assets.
  • the evaluation model processing unit 43 is configured to perform logistic regression processing on the common image data and the personal property mean value in each sub-cluster to obtain a property evaluation model.
  • the common portrait data of each sub-cluster and the average value of the personal assets of the sub-cluster are logically regression processed by a logistic regression algorithm to obtain a property evaluation model.
  • the common portrait data in the sub-cluster is mapped to the mean value of the individual assets.
  • the common image data is related to the geographical location data based on location service, which has objectivity and reliability, and the property evaluation model formed by it has objectivity and reliability.
  • Logistic Regression is a commonly used machine learning method in the industry to estimate the possibility of something.
  • ⁇ is the model parameter, that is, the regression coefficient
  • ⁇ is the sigmoid function.
  • this function is transformed by the following logarithmic probability (that is, the logarithm of the ratio of the likelihood that x belongs to a positive class and the likelihood of a negative class):
  • the property status evaluation module 50 is configured to evaluate the personal property status of the candidate user by using the property evaluation model according to the user image data of the candidate user, to output the personal property status evaluation result.
  • the user portrait data is mapped to the personal property status.
  • the user image data of the candidate user can be input into the trained property evaluation model for personal property status evaluation, and the personal property status evaluation result can be output, and the evaluation process is simple and convenient.
  • the output of the personal property status assessment results is objective and accurate, and the evaluation process is simple and convenient.
  • the property state evaluation module 50 specifically includes a similarity calculation unit 51, a similarity comparison unit 52, and an evaluation result output unit 53.
  • the similarity calculation unit 51 is configured to calculate a similarity between the user portrait data of the candidate user and the common portrait data of each sub-cluster by using a similarity algorithm.
  • the text similarity algorithm is used to calculate the similarity between the user portrait data of the candidate user and the shared portrait data of each sub-cluster.
  • the use of the text similarity algorithm to calculate the similarity includes the following process: First, the user image data of the candidate user is subjected to pre-processing such as word segmentation and de-stopping. Text feature extraction and weighting are then performed based on TF-IDF or other weights. Finally, the vector space model VSM is used to calculate the cosine value to calculate the similarity between the user image data of the candidate user and the common image data of each sub-cluster.
  • TF Term frequency
  • IDF Inverse document frequency
  • the text similarity algorithm is used to calculate the similarity, which has the advantages of simple calculation process and fast calculation speed. It can be understood that a similarity algorithm such as a semantic similarity-based text similarity algorithm and a pinyin similarity-based Chinese fuzzy search algorithm can also be used for processing.
  • the similarity comparison unit 52 is configured to determine whether the similarity is greater than a preset similarity threshold.
  • the similarity threshold is preset, and is used to determine the value of the candidate user belonging to any sub-cluster, and can be set autonomously.
  • the similarity threshold is set to 70%. That is, when the similarity between the user image data of the candidate user and the shared portrait data of a sub-cluster is greater than the similarity threshold (70%), the candidate user is considered to be attributable to the sub-cluster.
  • the evaluation result output unit 53 is configured to, if yes, use the average value of the personal property corresponding to the sub-cluster as the personal property status Evaluation result output.
  • the candidate user is considered to be attributable to the sub-cluster, and the average value of the personal property corresponding to the sub-cluster is used as the candidate user.
  • the output of the personal property status assessment results is associated with the user portrait data, and the user portrait data includes location information based on the location service, which is objectivity and reliability.
  • the geographical location fence information (including the target user) of the target area is acquired first, and the user portrait data associated with the target user is acquired; the target user is divided into the seed user and the candidate user;
  • the user evaluation data of the seed user is used to train the property evaluation model, and the user image data of the candidate user is processed by the trained property evaluation model, and the personal property evaluation result of the candidate user is output.
  • the user image data of the seed user in the target area is used to train the property evaluation model, and the trained property evaluation model is used to evaluate the personal property of the candidate users in the target area.
  • the evaluation process is simple and convenient, and the output personal property status evaluation result has a high Accuracy, objectivity and reliability.
  • Fig. 3 is a block diagram showing the structure of a personal property state evaluation device 300 according to a third embodiment of the present invention.
  • the device 300 may be a mobile terminal having a certain data processing capability such as a mobile phone, a tablet computer, a personal digital assistant (PDA), or an on-board computer, or a terminal such as a desktop computer or a server.
  • the device 300 includes a radio frequency (RF) circuit 301, a memory 302, an input module 303, a display module 304, a processor 305, an audio circuit 306, a WiFi (Wireless Fidelity) module 307, and a power source 308.
  • RF radio frequency
  • the input module 303 and the display module 304 serve as user interaction means of the device 300 for implementing interaction between the user and the device 300, for example, receiving a property evaluation instruction input by the user and displaying a corresponding personal property status evaluation result.
  • the input module 303 is configured to receive a property evaluation instruction input by the user, and send the property evaluation instruction to the processor 305, where the property evaluation instruction includes user portrait data of the candidate user. It can be understood that the user portrait data of the candidate user refers to user portrait data that requires personal property evaluation.
  • the processor 305 is configured to receive the property evaluation instruction, acquire the personal property status assessment result based on the property evaluation instruction, and send the personal property status assessment result to the display module 304.
  • the display module 304 receives and displays the personal property status assessment result.
  • the input module 303 can be configured to receive numeric or character information input by a user, and to generate signal inputs related to user settings and function control of the device 300.
  • the input module 303 can include a touch panel 3031.
  • the touch panel 3031 also referred to as a touch screen, can collect touch operations on or near the user (such as the operation of the user using any suitable object or accessory such as a finger or a stylus on the touch panel 3031), and The corresponding connection device is driven according to a preset program.
  • the touch panel 3031 may include two parts of a touch detection device and a touch controller.
  • the touch detection device detects the touch orientation of the user, and detects a signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts the touch information into contact coordinates, and sends the touch information.
  • the processor 305 is provided and can receive commands from the processor 305 and execute them.
  • the touch panel 3031 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves.
  • the input module 303 may further include other input devices 3032.
  • the other input devices 3032 may include but are not limited to physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, joysticks, and the like. One or more of them.
  • display module 304 can be used to display information entered by a user or information provided to a user and various menu interfaces of device 300.
  • the display module 304 can include a display panel 3041.
  • the display panel 3041 can be configured in the form of an LCD or an Organic Light-Emitting Diode (OLED).
  • the touch panel 3031 can cover the display panel 3041 to form a touch display screen.
  • the touch display screen detects a touch operation on or near it, it is transmitted to the processor 305 to determine the type of the touch event, and then processed.
  • the 305 provides a corresponding visual output on the touch display based on the type of touch event.
  • the touch display includes an application interface display area and a common control display area.
  • the arrangement manner of the application interface display area and the display area of the common control is not limited, and the arrangement manner of the two display areas can be distinguished by up-and-down arrangement, left-right arrangement, and the like.
  • the application interface display area can be used to display the interface of the application. Each interface can contain interface elements such as at least one application's icon and/or widget desktop control.
  • the application interface display area can also be an empty interface that does not contain any content.
  • the common control display area is used to display controls with high usage, such as setting buttons, interface numbers, scroll bars, phone book icons, and the like.
  • the WiFi module 307 can be used as the network interface of the device 300 to implement data interaction between the device 300 and other devices.
  • the network interface can be connected to the remote storage device and the external display device through network communication.
  • the network interface is configured to receive the geographical location fence information and the user portrait data sent by the remote storage device, and send the geographical location fence information and the user portrait data to the processor 305; It is further configured to receive the personal property status assessment result sent by the processor 305, and send the personal property status assessment result to the external display device.
  • the external display device can receive and display the personal property status assessment result.
  • the remote storage device connected to the network interface through the WiFi network may be a cloud server or other database, where the remote location storage device stores the geographical location fence information and the user portrait data.
  • the geolocation fence information and the user portrait data may be sent to the WiFi module 307 through the WiFi network, and the WiFi module 307 sends the acquired geographical location fence information and the user portrait data to the processor. 305. Send the personal property status assessment result to the external display device.
  • the remote storage device may be a real estate intermediary platform and/or room that stores location fence information.
  • the production registration platform stores the geographical location fence information, and may also be a service platform of a financial institution such as a bank or insurance that stores user image data of the target user.
  • the memory 302 includes a first memory 3021 and a second memory 3022.
  • the first memory 3021 can be a non-transitory computer readable storage medium having an operating system, a database, and computer executable instructions stored thereon.
  • Computer executable instructions are executable by processor 305 for implementing the personal property status assessment method of the embodiment as shown in FIG.
  • the database on the memory 302 is used to store various types of data, for example, various data involved in the above-described personal property status evaluation method, such as geographical location information, user portrait data, and property evaluation models.
  • the second memory 3021 can be an internal memory of the device 300 that provides a cached operating environment for operating systems, databases, and computer executable instructions in a non-transitory computer readable storage medium.
  • processor 305 is the control center of device 300, which connects various portions of the entire handset using various interfaces and lines, by running or executing computer-executable collections and/or databases stored in first memory 3021. The data, performing various functions and processing data of the device 300, thereby performing overall monitoring of the device 300.
  • processor 305 can include one or more processing modules.
  • the processor 305 by executing the stored in the first executable memory 3021 and the data in the database, the processor 305 is configured to: acquire the geographical location fence information of the target area, where the geographical location fence information includes Housing information and corresponding target users; acquiring user portrait data associated with the target user; dividing all target users into seed data sets and candidate data sets; the seed data set including at least one seed user, the candidate data The set includes at least one candidate user; the property evaluation model is trained by using user image data of the seed user; and the personal property status of the candidate user is evaluated by using the property evaluation model according to the user image data of the candidate user To output the results of personal property status assessment.
  • the training the property evaluation model by using user image data of the seed user comprises:
  • the user portrait data of all seed users is classified by using a look-alike algorithm, and a plurality of the sub-cluster and the common portrait data corresponding to each of the sub-clusters are obtained;
  • the shared portrait data in each of the sub-clusters is logically regressed with the personal property mean to obtain the property evaluation model.
  • a personal property status assessment result including:
  • the individual property average corresponding to the sub-cluster is output as the personal property status evaluation result.
  • the user portrait data includes location information based on location services, the geographic location information including POI information associated with time.
  • the acquiring the geographical location fence information of the target area includes:
  • the web crawler is used to crawl the real estate intermediary platform and/or the real estate registration platform to obtain the geographical location fence information of the target area.
  • the processor 305 first acquires the geographical location fence information (including the target user) of the target area, and acquires the user portrait data associated with the target user; and divides the target user into the seed user and
  • the candidate user uses the user image data of the seed user to train the property evaluation model, and uses the trained property evaluation model to process the user image data of the candidate user, and outputs the personal property evaluation result of the candidate user.
  • the user image data of the seed user in the target area is used to train the property evaluation model, and the trained property evaluation model is used to evaluate the personal property of the candidate users in the target area, so that the personal property status evaluation result has high accuracy, objectivity and reliability. Sex.
  • the embodiment provides a non-transitory computer readable storage medium.
  • the non-transitory computer readable storage medium is for storing one or more computer executable instructions.
  • the computer-executable instructions are executed by one or more processors, such that the one or more processors perform the personal property status assessment method described in the first embodiment. To avoid repetition, details are not described herein again.
  • modules and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods for implementing the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present invention.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the modules is only a logical function division.
  • there may be another division manner for example, multiple modules or components may be combined. Or it can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or module, and may be electrical, mechanical or otherwise.
  • the modules described as separate components may or may not be physically separated.
  • the components displayed as modules may or may not be physical modules, that is, may be located in one place, or may be distributed to multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
  • the functions, if implemented in the form of software functional modules and sold or used as separate products, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Signal Processing (AREA)
  • Geometry (AREA)
  • Evolutionary Computation (AREA)
  • Computer Hardware Design (AREA)
  • Development Economics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种个人财产状态评估方法、装置、设备和存储介质。该个人财产评估方法包括：获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户（S10）；获取与所述目标用户相关联的用户画像数据（S20）；将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户（S30）；利用所述种子用户的用户画像数据训练所述财产评估模型（S40）；根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果（S50）。该个人财产评估方法中，获取的个人财产状态评估结果具有较高的准确性、客观性和可靠性。

Description

个人财产状态评估方法、装置、设备和存储介质 技术领域
本发明涉及信息处理技术领域，尤其涉及一种个人财产状态评估方法、装置、设备和存储介质。
背景技术
在金融机构给客户提供家庭消费管理、理财规划管理、资产配置管理和和投资管理等业务时，需全面评估客户的个人财产状态，并对个人财产状态进行整理和分析，及时发现客户的财务隐患，纠正不良理财习惯，提高抵抗金融风险的能力。现有金融机构主要利用客户的资产状况和消费流水等金融数据评估个人财产状态，评估数据来源单一，导致个人财产状态评估结果准确率较低。金融机构基于准确率较低的个人财产状态评估结果给客户提供服务时，提供与其个人财产状态不匹配的业务，可能导致金融风险。
发明内容
本发明提供一种个人财产状态评估方法、装置、设备和存储介质，以解决现有技术中个人财产状态评估时，评估数据来源单一且个人财产状态评估结果准确率较低的问题。
本发明解决其技术问题所采用的技术方案是：
第一方面，本发明提供一种个人财产状态评估方法，包括：
获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
获取与所述目标用户相关联的用户画像数据；
将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
利用所述种子用户的用户画像数据训练所述财产评估模型；
根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
第二方面，本发明提供一种个人财产状态评估装置，包括：
围栏信息获取模块，用于获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
画像数据获取模块，用于获取与所述目标用户相关联的用户画像数据；
数据集划分模块，用于将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
评估模型训练模块，用于利用所述种子用户的用户画像数据训练所述财产评估模型；
财产状态评估模块，用于根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
第三方面，本发明还提供一种个人财产状态评估设备，包括处理器及存储器，所述存储器存储有计算机可执行指令，所述处理器用于执行所述计算机可执行指令以执行如下步骤：
获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
获取与所述目标用户相关联的用户画像数据；
将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
利用所述种子用户的用户画像数据训练所述财产评估模型；
根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
第四方面，本发明还提供一种非易失性计算机可读存储介质，用于存储一个或多个计算机可执行指令，所述计算机可执行指令被一个或多个处理器执行，使得所述一个或多个处理器执行任一项所述的个人财产状态评估方法。
本发明与现有技术相比具有如下优点：本发明所提供的个人财产状态评估方法、装置、设备和存储介质中，先获取目标区域的地理位置围栏信息(包括目标用户)，并获取与目标用户相关联的用户画像数据；将目标用户划分成种子用户和候选用户；采用种子用户的用户画像数据训练财产评估模型，并利用训练好的财产评估模型对候选用户的用户画像数据进行处理，输出候选用户的个人财产评估结果。采用目标区域的种子用户的用户画像数据训练财产评估模型，利用训练好的财产评估模型对目标区域的候选用户进行个人财产评估，使得个人财产状态评估结果具有较高的准确性、客观性和可靠性。
附图说明
下面将结合附图及实施例对本发明作进一步说明，附图中：
图1是本发明第一实施例中个人财产状态评估方法的一流程图；
图2是本发明第二实施例中个人财产状态评估装置的一原理框图。
图3是本发明第三实施例中个人财产状态评估设备的一示意图。
具体实施方式
为了对本发明的技术特征、目的和效果有更加清楚的理解，现对照附图详细说明本发明的具体实施方式。
第一实施例
图1示出本实施例中个人财产状态评估方法的一流程图。该个人财产状态评估方法可在银行、保险等金融机构的个人财产状态评估设备上应用，用于评估任一用户的个人财产状态。如图1所示，该个人财产状态评估方法包括如下步骤：
S10：获取目标区域的地理位置围栏信息，地理位置围栏信息包括房屋信息和对应的目标用户。
其中，目标区域可以是任一住宅小区。房屋信息可以是目标区域(住宅小区)内任一房屋的房屋位置、房屋房号、房屋大小、房屋销售均价、房屋租金均价等信息。目标用户可以是该房屋信息对应的房屋所有权人。本实施例中，获取目标区域的地理位置围栏信息，以获取任一住宅小区中每一房屋的房屋信息和对应的目标用户，由于目标用户居住在同一住宅小区内，其个人财产状态具有一定的相似性，以便于基于地理位置围栏信息确定的目标用户进行个人财产状态评估。该目标区域对应的住宅小区优选为房屋销售均价较高的住宅小区，房屋销售均价超高，其对应的房屋户主(即目标用户)应当具有相应的个人财产状态。
具体地，步骤S10具体包括：采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位置围栏信息。
具体地，网络爬虫按预设规则自动抓取房产中介平台和/或房产登记平台中的房屋信息，并获取与房屋信息相关联的目标用户，将房屋信息和目标用户作为目标区域的地理位置信息输出。可以理解地，房产中介平台和/或房产登记平台中存储有目标区域任一房屋的房屋信息和目标用户，采用网络爬虫从房产中介平台和/或房产登记平台爬取目标区域的地理位置围栏信息，爬取数据内容明确，处理速度较快。
本实施例中，目标用户是与目标区域的任一房屋信息对应的房屋所有权人，同一目标区域的目标用户的个人财产状态具有一定的相似度。可以理解地，与目标区域的房屋销售均 价相同或相似的其他区域也可以作为同一目标区域，以扩大目标区域的目标用户范围。基于目标区域的目标用户进行个人财产评估，在一定程度上可提高个人财产状态评估结果的准确性和可靠性。
S20：获取与目标用户相关联的用户画像数据。
用户画像数据(即Persona数据)是真实用户的虚拟代表，是建立在一系统真实数据(Marketing Data/Usability Data)之上的目标用户模型。当前银行、保险等金融机构的个人财产状态评估设备中存储目标用户的用户画像数据，该用户画像数据包括但不限于用户姓名、身份识别特征、照片、联系方式、家庭住址、办公场所、职业和收入等。本实施例中，获取与目标用户相关联的用户画像数据中，每一用户画像数据对应的用户与地理信息围栏信息中的房屋信息和目标用户相关联，以使每一目标用户的用户画像数据均包含目标区域的房屋信息，以便基于与房屋信息相关的目标用户的用户画像数据进行处理，
具体地，用户画像数据包括基于位置服务的地理位置信息，地理位置信息包括与时间相关联的POI信息。
以目标用户一天的地理位置信息为例，该地理位置信息中包括0：00—24：00的POI信息，每一POI信息用于指示电子地图中的一点，包括POI点名称、经度和纬度等信息。通过对该目标用户在一段时间内每天的地理位置信息进行处理，可确定该目标用户的家庭住址、办公场所、上下班时间、常去的娱乐、消费、健身等等。可以理解地，用户画像数据还可以包括消费特征、投资特征或其他影响个人财产评估因素特征。可以理解地，若目标用户经常性出入高档消费场所，或者在金融机构有大额投资记录等信息，则该目标用户的个人财产较低，其获取的个人财产评估结果较高，以保证个人财产评估的准确性。基于位置服务的地理位置信息是目标用户的日常生活轨迹，具有客观性，基于地理位置信息进行个人财产状态评估，可有利于提高个人财产状态评估结果的客观性和准确性，避免仅依据目标用户自主提供的信息进行个人财产评估所导致的主观性强，评估结果准确性低的问题出现。
其中，基于位置服务(Location Based Service，简称LBS)是通过电信移动运营商的无线电通讯网络(如GSM网、CDMA网)或外部定位方式(如GPS)获取移动终端(即目标用户)的位置信息(地理坐标，或大地坐标)，在地理信息系统(Geographic Information System，简称GIS)平台的支持下，为目标用户提供相应服务的一种增值业务。总体来看，LBS由移动通信网络和计算机网络结合而成，两个网络之间通过网关实现交互。移动终端通过移动通信网络发出请求，经过网关传递给LBS服务平台；LBS服务平台根据目标用户请求和目标用户当前位置进行处理，并将结果通过网关返回给目标用户。POI(Point Of Interest,即兴 趣点或信息点)，包括名称、类型、经度、纬度等资料，以使POI可在电子地图上呈现，以标示电子地图上的某个地标、景点等地点信息。
本实施例中，基于位置服务的移动终端为智能手机，通过开启智能手机上的定位功能，以使LBS服务平台实时获取智能手机的地理位置信息，从而了解携带该智能手机的目标用户的地理位置信息。该LSB服务平台与个人财产状态评估设备相连，以使个人财产状态评估设备获取到目标用户的地理位置信息，并作为相应的用户画像数据存储。地理位置信息包括与时间相关联的POI信息中的时间包括日期和时刻，通过该地理位置信息可了解目标用户在任一时刻所处的POI信息。可以理解地，地理位置信息与目标用户的用户ID相关联，用户ID用于识别唯一识别用户，可以是身份证号或手机号。
S30：将所有目标用户划分成种子数据集和候选数据集；种子数据集包括至少一个种子用户，候选数据集包括至少一个候选用户。
本实施例中，对目标区域所有目标用户，依据是否进行过个人财产评估且具有评估的个人财产状态作为划分条件，将目标区域所有目标用户划分成种子数据集和候选数据集。其中，种子数据集中每一种子用户具有评估后的个人财产状态。候选数据集中每一候选用户不具有评估后的个人财产状态。
S40：利用种子用户的用户画像数据训练财产评估模型。
由于种子数据集中所有种子用户均具有评估后的个人财产状态，而每一种子用户均具有相应的用户画像数据，该用户画像数据包括但不限于用户姓名、身份识别特征、照片、联系方式、家庭住址、办公场所、职业和收入等，还包括体现目标用户日常生活轨迹的基于位置服务的地理位置数据。对所有种子用户的用户画像数据进行共性分析，获取种子用户的画像数据与个人财产状态之间的关联关系，以训练财产评估模型。
可以理解地，每一种子用户是目标区域的目标用户，利用种子用户的用户画像数据训练财产评估模型，在一定程度上可提高个人财产状态评估结果的准确性和可靠性。而且，用户画像数据包括用于体现种子用户日常生活轨迹的地理位置数据，具有客观性，基于地理位置信息训练财产评估模型，可有利于提高个人财产状态评估结果的客观性和准确性。
进一步地，步骤S40包括如下步骤：
S41：采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干子集群及每一所述子集群对应的共有画像数据。
其中，Look-alike,即相似人群扩展，是一种基于现有用户/设备ID，通过一定的算法评估模型，找到更多拥有潜在关联性的相似人群的技术。本实施例所采用look-alike算法 中采用种子用户的用户画像数据为正样本，训练分类模型以获取共有画像数据，以便于采用候选用户的用户画像数据为负样本，通过分类模型进行筛选。
具体地，采用look-alike算法对所有种子用户的用户画像数据进行分类过程中采用到基于PU-Learning(Learning from Positive and Unlabled Example，即正例和无标记样本学习)的分类方法进行分类，分类过程简单方便，可有效降低人工分类的预备工作量，提高分类精度。可以理解地，采用look-alike算法对所有种子用户的用户画像数据进行分类，获得的每一子集群具有相同的共有画像数据，是可用于评估个人财产状态的关联特征。
由于每一种子用户的用户画像数据包括体现目标用户日常生活轨迹的基于位置服务的地理位置数据，采用look-alike算法对所有种子用户的用户画像数据进行分类获取的每一子集群中的共有画像数据与基于位置服务的地理位置数据相关联，具有客观性和可靠性。
S42：获取每一种子用户的个人财产状态，并计算每一子集群中所有种子用户的个人财产均值。
由于种子数据集中每一种子用户具有评估后的个人财产状态，则采用look-alike算法对所有种子用户的用户画像数据进行分类而获取到的每一子集群中的种子用户也具有评估后的个人财产状态。本实施例中，计算每一子集群中所有种子用户的个人财产均值，采用个人财产均值构建财产评估模型。
S43：将每一子集群中的共有画像数据与个人财产均值进行逻辑回归处理，以获取财产评估模型。
本实施例中，将每一子集群的共有画像数据与该子集群的个人财产均值采用逻辑回归算法进行逻辑回归处理，以获取财产评估模型。该财产评估模型中，子集群中的共有画像数据与个人财产均值建立映射关系。其中，共有画像数据与基于位置服务的地理位置数据相关联，具有客观性和可靠性，使其形成的财产评估模型具有客观性和可靠性。
其中，逻辑回归(Logistic Regression)是当前业界比较常用的机器学习方法，用于估计某种事物的可能性。逻辑回归(Logistic Regression)是一个被logistic方程归一化后的线性回归。在逻辑回归(Logistic Regression)中，若设样本是{x,y}，y是0或者1，表示正类或者负类，x是我们的m维的样本特征向量。那么这个样本x属于正类，也就是y＝1的“概率”可以通过下面的逻辑函数来表示：
Figure PCTCN2017076463-appb-000001
其中，θ是模型参数，也就是回归系数，σ是sigmoid函数。实际上这个函数是由下 面的对数几率(也就是x属于正类的可能性和负类的可能性的比值的对数)变换得到的：
Figure PCTCN2017076463-appb-000002
S50：根据候选用户的用户画像数据，利用财产评估模型对候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
在练好的财产评估模型中，使用户画像数据与个人财产状态建立映射关系。对于任一不具有评估的个人财产状态的候选用户，只需将候选用户的用户画像数据输入训练好的财产评估模型进行个人财产状态评估，即可输出个人财产状态评估结果，评估过程简单方便，且输出的个人财产状态评估结果具有客观性和准确性，并且评估过程操作简单方便。
进一步地，步骤S50包括如下步骤：
S51：采用相似度算法计算候选用户的用户画像数据与每一子集群的共有画像数据的相似度。
本实施例中，采用文本相似度算法计算候选用户的用户画像数据与每一子集群的共有画像数据的相似度。采用文本相似度算法计算相似度包括如下过程：首先，对候选用户的用户画像数据进行分词、去停用词等预处理。然后，基于TF-IDF或者其他权重进行文本特征提取与加权。最后，采用向量空间模型VSM进行余弦值计算，以计算得到候选用户的用户画像数据与每一子集群的共有画像数据的相似度。其中，TF(Term frequency，即关键词词频)，是指一篇文章中关键词出现的频率；IDF(Inverse document frequency，即逆向文本频率)，是用于衡量关键词权重的指数。采用文本相似度算法计算相近度，具有计算过程简单，计算速度较快的优点。可以理解地，还可以采用基于语义相似度的文本相似度算法、基于拼音相似度的汉语模糊搜索算法等相似度算法进行处理。
S52：判断相似度是否大于预设的相似阈值。
其中，相似阈值是预先设置的，用于判断候选用户归属于任一子集群的数值，可自主设置。本实施例中，相似阈值设为70％。即候选用户的用户画像数据与一子集群的共有画像数据的相似度大于相似阈值(70％)时，则认为候选用户可归属于该子集群。
S53：若是，则将子集群对应的个人财产均值作为个人财产状态评估结果输出。
可以理解地，若候选用户的用户画像数据与一子集群的共有画像数据的相似度大于相似阈值时，认为候选用户可归属于该子集群，将该子集群对应的个人财产均值作为该候选用户的个人财产状态评估结果输出。本实施例中，任一候选用户的个人财产状态评估结果与其用户画像数据相关联，而用户画像数据包括基于位置服务的地理位置信息相关联，具有客观 性和可靠性。
本实施例所提供的个人财产评估方法中，先获取目标区域的地理位置围栏信息(包括目标用户)，并获取与目标用户相关联的用户画像数据；将目标用户划分成种子用户和候选用户；采用种子用户的用户画像数据训练财产评估模型，并利用训练好的财产评估模型对候选用户的用户画像数据进行处理，输出候选用户的个人财产评估结果。采用目标区域的种子用户的用户画像数据训练财产评估模型，利用训练好的财产评估模型对目标区域的候选用户进行个人财产评估，评估过程简单方便，且输出的个人财产状态评估结果具有较高的准确性、客观性和可靠性。
第二实施例
图2示出本实施例中个人财产状态评估装置的一流程图。该个人财产状态评估装置可以是银行、保险等金融机构中应用的个人财产状态评估设备，可用于评估任一用户的个人财产状态。如图2所示，该个人财产状态评估装置包括围栏信息获取模块10、画像数据获取模块20、数据集划分模块30、评估模型训练模块40和财产状态评估模块50。
围栏信息获取模块10，用于获取目标区域的地理位置围栏信息，地理位置围栏信息包括房屋信息和对应的目标用户。
其中，目标区域可以是任一住宅小区。房屋信息可以是目标区域(住宅小区)内任一房屋的房屋位置、房屋房号、房屋大小、房屋销售均价、房屋租金均价等信息。目标用户可以是该房屋信息对应的房屋所有权人。本实施例中，获取目标区域的地理位置围栏信息，以获取任一住宅小区中每一房屋的房屋信息和对应的目标用户，由于目标用户居住在同一住宅小区内，其个人财产状态具有一定的相似性，以便于基于地理位置围栏信息确定的目标用户进行个人财产状态评估。该目标区域对应的住宅小区优选为房屋销售均价较高的住宅小区，房屋销售均价超高，其对应的房屋户主(即目标用户)应当具有相应的个人财产状态。
具体地，围栏信息获取模块10，用于采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位置围栏信息。
具体地，网络爬虫按预设规则自动抓取房产中介平台和/或房产登记平台中的房屋信息，并获取与房屋信息相关联的目标用户，将房屋信息和目标用户作为目标区域的地理位置信息输出。可以理解地，房产中介平台和/或房产登记平台中存储有目标区域任一房屋的房屋信息和目标用户，采用网络爬虫从房产中介平台和/或房产登记平台爬取目标区域的地理位置围栏信息，爬取数据内容明确，处理速度较快。
本实施例中，目标用户是与目标区域的任一房屋信息对应的房屋所有权人，同一目标 区域的目标用户的个人财产状态具有一定的相似度。可以理解地，与目标区域的房屋销售均价相同或相似的其他区域也可以作为同一目标区域，以扩大目标区域的目标用户范围。基于目标区域的目标用户进行个人财产评估，在一定程度上可提高个人财产状态评估结果的准确性和可靠性。
画像数据获取模块20，用于获取与目标用户相关联的用户画像数据。
用户画像数据(即Persona数据)是真实用户的虚拟代表，是建立在一系统真实数据(Marketing Data/Usability Data)之上的目标用户模型。当前银行、保险等金融机构的个人财产状态评估设备中存储有目标用户的用户画像数据。该用户画像数据包括但不限于用户姓名、身份识别特征、照片、联系方式、家庭住址、办公场所、职业和收入等。本实施例中，获取与目标用户相关联的用户画像数据中，每一用户画像数据对应的用户与地理信息围栏信息中的房屋信息和目标用户相关联，以使每一目标用户的用户画像数据均包含目标区域的房屋信息，以便基于与房屋信息相关的目标用户的用户画像数据进行处理，
具体地，用户画像数据包括基于位置服务的地理位置信息，地理位置信息包括与时间相关联的POI信息。
以目标用户一天的地理位置信息为例，该地理位置信息中包括0：00—24：00的POI信息，每一POI信息用于指示电子地图中的一点，包括POI点名称、经度和纬度等信息。通过对该目标用户在一段时间内每天的地理位置信息进行处理，可确定该目标用户的家庭住址、办公场所、上下班时间、常去的娱乐、消费、健身等等。可以理解地，用户画像数据还可以包括消费特征、投资特征或其他影响个人财产评估因素特征。可以理解地，若目标用户经常性出入高档消费场所，或者在金融机构有大额投资记录等信息，则该目标用户的个人财产较低，其获取的个人财产评估结果较高，以保证个人财产评估的准确性。基于位置服务的地理位置信息是目标用户的日常生活轨迹，具有客观性，基于地理位置信息进行个人财产状态评估，可有利于提高个人财产状态评估结果的客观性和准确性，避免仅依据目标用户自主提供的信息进行个人财产评估所导致的主观性强，评估结果准确性低的问题出现。
其中，基于位置服务(Location Based Service，简称LBS)是通过电信移动运营商的无线电通讯网络(如GSM网、CDMA网)或外部定位方式(如GPS)获取移动终端(即目标用户)的位置信息(地理坐标，或大地坐标)，在地理信息系统(Geographic Information System，简称GIS)平台的支持下，为目标用户提供相应服务的一种增值业务。总体来看，LBS由移动通信网络和计算机网络结合而成，两个网络之间通过网关实现交互。移动终端通过移动通信网络发出请求，经过网关传递给LBS服务平台；LBS服务平台根据目标用户请求和目标用 户当前位置进行处理，并将结果通过网关返回给目标用户。POI(Point Of Interest,即兴趣点或信息点)，包括名称、类型、经度、纬度等资料，以使POI可在电子地图上呈现，以标示电子地图上的某个地标、景点等地点信息。
本实施例中，基于位置服务的移动终端为智能手机，通过开启智能手机上的定位功能，以使LBS服务平台实时获取智能手机的地理位置信息，从而了解携带该智能手机的目标用户的地理位置信息。该LSB服务平台与个人财产状态评估设备相连，以使个人财产状态评估设备获取到目标用户的地理位置信息，并作为相应的用户画像数据存储。地理位置信息包括与时间相关联的POI信息中的时间包括日期和时刻，通过该地理位置信息可了解目标用户在任一时刻所处的POI信息。可以理解地，地理位置信息与目标用户的用户ID相关联，用户ID用于识别唯一识别用户，可以是身份证号或手机号。
数据集划分模块30，用于将所有目标用户划分成种子数据集和候选数据集；种子数据集包括至少一个种子用户，候选数据集包括至少一个候选用户。
本实施例中，对目标区域所有目标用户，依据是否进行过个人财产评估且具有评估的个人财产状态作为划分条件，将目标区域所有目标用户划分成种子数据集和候选数据集。其中，种子数据集中每一种子用户具有评估后的个人财产状态。候选数据集中每一候选用户不具有评估后的个人财产状态。
评估模型训练模块40，用于利用种子用户的用户画像数据训练财产评估模型。
由于种子数据集中所有种子用户均具有评估后的个人财产状态，而每一种子用户均具有相应的用户画像数据，该用户画像数据包括但不限于用户姓名、身份识别特征、照片、联系方式、家庭住址、办公场所、职业和收入等，还包括体现目标用户日常生活轨迹的基于位置服务的地理位置数据。对所有种子用户的用户画像数据进行共性分析，获取种子用户的画像数据与个人财产状态之间的关联关系，以训练财产评估模型。
可以理解地，每一种子用户是目标区域的目标用户，利用种子用户的用户画像数据训练财产评估模型，在一定程度上可提高个人财产状态评估结果的准确性和可靠性。而且，用户画像数据包括用于体现种子用户日常生活轨迹的地理位置数据，具有客观性，基于地理位置信息训练财产评估模型，可有利于提高个人财产状态评估结果的客观性和准确性。
进一步地，评估模型训练模块40具体包括画像数据分类单元41、财产均值计算单元42和评估模型处理单元43。
画像数据分类单元41，用于采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干子集群及每一所述子集群对应的共有画像数据。
其中，Look-alike,即相似人群扩展，是一种基于现有用户/设备ID，通过一定的算法评估模型，找到更多拥有潜在关联性的相似人群的技术。本实施例所采用look-alike算法中采用种子用户的用户画像数据为正样本，训练分类模型以获取共有画像数据，以便于采用候选用户的用户画像数据为负样本，通过分类模型进行筛选。
具体地，采用look-alike算法对所有种子用户的用户画像数据进行分类过程中采用到基于PU-Learning(Learning from Positive and Unlabled Example，正例和无标记样本学习)的分类方法进行分类，分类过程简单方便，可有效降低人工分类的预备工作量，提高分类精度。可以理解地，采用look-alike算法对所有种子用户的用户画像数据进行分类，获得的每一子集群具有相同的共有画像数据，是可用于评估个人财产状态的关联特征。
由于每一种子用户的用户画像数据包括体现目标用户日常生活轨迹的基于位置服务的地理位置数据，采用look-alike算法对所有种子用户的用户画像数据进行分类获取的每一子集群中的共有画像数据与基于位置服务的地理位置数据相关联，具有客观性和可靠性。
财产均值计算单元42，用于获取每一种子用户的个人财产状态，并计算每一子集群中所有种子用户的个人财产均值。
由于种子数据集中每一种子用户具有评估后的个人财产状态，则采用look-alike算法对所有种子用户的用户画像数据进行分类而获取到的每一子集群中的种子用户也具有评估后的个人财产状态。本实施例中，计算每一子集群中所有种子用户的个人财产均值，采用个人财产均值构建财产评估模型。
评估模型处理单元43，用于将每一子集群中的共有画像数据与个人财产均值进行逻辑回归处理，以获取财产评估模型。
本实施例中，将每一子集群的共有画像数据与该子集群的个人财产均值采用逻辑回归算法进行逻辑回归处理，以获取财产评估模型。该财产评估模型中，子集群中的共有画像数据与个人财产均值建立映射关系。其中，共有画像数据与基于位置服务的地理位置数据相关联，具有客观性和可靠性，使其形成的财产评估模型具有客观性和可靠性。
其中，逻辑回归(Logistic Regression)是当前业界比较常用的机器学习方法，用于估计某种事物的可能性。逻辑回归(Logistic Regression)是一个被logistic方程归一化后的线性回归。在逻辑回归(Logistic Regression)中，若设样本是{x,y}，y是0或者1，表示正类或者负类，x是我们的m维的样本特征向量。那么这个样本x属于正类，也就是y＝1的“概率”可以通过下面的逻辑函数来表示：
Figure PCTCN2017076463-appb-000003
其中，θ是模型参数，也就是回归系数，σ是sigmoid函数。实际上这个函数是由下面的对数几率(也就是x属于正类的可能性和负类的可能性的比值的对数)变换得到的：
Figure PCTCN2017076463-appb-000004
财产状态评估模块50，用于根据候选用户的用户画像数据，利用财产评估模型对候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
在练好的财产评估模型中，使用户画像数据与个人财产状态建立映射关系。对于任一不具有评估的个人财产状态的候选用户，只需将候选用户的用户画像数据输入训练好的财产评估模型进行个人财产状态评估，即可输出个人财产状态评估结果，评估过程简单方便，且输出的个人财产状态评估结果具有客观性和准确性，并且评估过程操作简单方便。
进一步地，财产状态评估模块50具体包括相似度计算单元51、相似度比较单元52和评估结果输出单元53。
相似度计算单元51，用于采用相似度算法计算候选用户的用户画像数据与每一子集群的共有画像数据的相似度。
本实施例中，采用文本相似度算法计算候选用户的用户画像数据与每一子集群的共有画像数据的相似度。采用文本相似度算法计算相似度包括如下过程：首先，对候选用户的用户画像数据进行分词、去停用词等预处理。然后，基于TF-IDF或者其他权重进行文本特征提取与加权。最后，采用向量空间模型VSM进行余弦值计算，以计算得到候选用户的用户画像数据与每一子集群的共有画像数据的相似度。其中，TF(Term frequency，即关键词词频)，是指一篇文章中关键词出现的频率；IDF(Inverse document frequency，即逆向文本频率)，是用于衡量关键词权重的指数。采用文本相似度算法计算相近度，具有计算过程简单，计算速度较快的优点。可以理解地，还可以采用基于语义相似度的文本相似度算法、基于拼音相似度的汉语模糊搜索算法等相似度算法进行处理。
相似度比较单元52，用于判断相似度是否大于预设的相似阈值。
其中，相似阈值是预先设置的，用于判断候选用户归属于任一子集群的数值，可自主设置。本实施例中，相似阈值设为70％。即候选用户的用户画像数据与一子集群的共有画像数据的相似度大于相似阈值(70％)时，则认为候选用户可归属于该子集群。
评估结果输出单元53，用于若是，则将子集群对应的个人财产均值作为个人财产状态 评估结果输出。
可以理解地，若候选用户的用户画像数据与一子集群的共有画像数据的相似度大于相似阈值时，认为候选用户可归属于该子集群，将该子集群对应的个人财产均值作为该候选用户的个人财产状态评估结果输出。本实施例中，任一候选用户的个人财产状态评估结果与其用户画像数据相关联，而用户画像数据包括基于位置服务的地理位置信息相关联，具有客观性和可靠性。
本实施例所提供的个人财产评估装置中，先获取目标区域的地理位置围栏信息(包括目标用户)，并获取与目标用户相关联的用户画像数据；将目标用户划分成种子用户和候选用户；采用种子用户的用户画像数据训练财产评估模型，并利用训练好的财产评估模型对候选用户的用户画像数据进行处理，输出候选用户的个人财产评估结果。采用目标区域的种子用户的用户画像数据训练财产评估模型，利用训练好的财产评估模型对目标区域的候选用户进行个人财产评估，评估过程简单方便，且输出的个人财产状态评估结果具有较高的准确性、客观性和可靠性。
第三实施例
图3是本发明第三实施例的个人财产状态评估设备300的结构示意图。其中，设备300可以为手机、平板电脑、个人数字助理(PersonalDigital Assistant，PDA)和或车载电脑等具有一定的数据处理能力的移动终端、或者台式电脑、服务器等终端。如图3所示，设备300包括射频(RadioFrequency，RF)电路301、存储器302、输入模块303、显示模块304、处理器305、音频电路306、WiFi(WirelessFidelity)模块307和电源308。
输入模块303和显示模块304作为设备300的用户交互装置，用于实现用户与设备300之间的交互，例如，接收用户输入的财产评估指令并显示对应的个人财产状态评估结果。输入模块303用于接收用户输入的财产评估指令，并将所述财产评估指令发送给所述处理器305，所述财产评估指令包括候选用户的用户画像数据。可以理解地，该候选用户的用户画像数据是指需要进行个人财产评估的用户画像数据。所述处理器305用于接收所述财产评估指令，并基于所述财产评估指令获取所述个人财产状态评估结果，并将所述个人财产状态评估结果发送给所述显示模块304。显示模块304接收并显示个人财产状态评估结果。
在一些实施例中，输入模块303可用于接收用户输入的数字或字符信息，以及产生与设备300的用户设置以及功能控制有关的信号输入。在一些实施例中，该输入模块303可以包括触控面板3031。触控面板3031，也称为触摸屏，可收集用户在其上或附近的触摸操作(比如用户使用手指、触笔等任何适合的物体或附件在触控面板3031上的操作)，并根 据预先设定的程式驱动相应的连接装置。可选地，触控面板3031可包括触摸检测装置和触摸控制器两个部分。其中，触摸检测装置检测用户的触摸方位，并检测触摸操作带来的信号，将信号传送给触摸控制器；触摸控制器从触摸检测装置上接收触摸信息，并将它转换成触点坐标，再送给该处理器305，并能接收处理器305发来的命令并加以执行。此外，可以采用电阻式、电容式、红外线以及表面声波等多种类型实现触控面板3031。除了触控面板3031，输入模块303还可以包括其他输入设备3032，其他输入设备3032可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆等中的一种或多种。
在一些实施例中，显示模块304可用于显示由用户输入的信息或提供给用户的信息以及设备300的各种菜单界面。显示模块304可包括显示面板3041，可选地，可以采用LCD或有机发光二极管(Organic Light-Emitting Diode，OLED)等形式来配置显示面板3041。
可以理解地，触控面板3031可以覆盖显示面板3041，形成触摸显示屏，当该触摸显示屏检测到在其上或附近的触摸操作后，传送给处理器305以确定触摸事件的类型，随后处理器305根据触摸事件的类型在触摸显示屏上提供相应的视觉输出。
触摸显示屏包括应用程序界面显示区及常用控件显示区。该应用程序界面显示区及该常用控件显示区的排列方式并不限定，可以为上下排列、左右排列等可以区分两个显示区的排列方式。该应用程序界面显示区可以用于显示应用程序的界面。每一个界面可以包含至少一个应用程序的图标和/或widget桌面控件等界面元素。该应用程序界面显示区也可以为不包含任何内容的空界面。该常用控件显示区用于显示使用率较高的控件，例如，设置按钮、界面编号、滚动条、电话本图标等应用程序图标等。
WiFi模块307作为设备300的网络接口，可以实现设备300与其他设备的数据交互，本实施例中，网络接口可与远端存储设备和外部显示设备通过网络通信相连。所述网络接口用于接收所述远端存储设备发送的所述地理位置围栏信息和所述用户画像数据，并将所述地理位置围栏信息和所述用户画像数据发送给所述处理器305；还用于接收所述处理器305发送的所述个人财产状态评估结果，并将所述个人财产状态评估结果发送给所述外部显示设备。外部显示设备可接收并显示所述个人财产状态评估结果。本实施例中，与该网络接口通过WiFi网络相连的远端存储设备可以是云服务器或其他数据库，该远端存储设备上存储有所述地理位置围栏信息和所述用户画像数据，可将所述所述地理位置围栏信息和所述用户画像数据可通过WiFi网络发送给WiFi模块307，WiFi模块307将获取到的所述所述地理位置围栏信息和所述用户画像数据发送给所述处理器305，并将所述个人财产状态评估结果发送给所述外部显示设备。远端存储设备可以是存储有地理位置围栏信息的房产中介平台和/或房 产登记平台中存储有地理位置围栏信息，还可以是存储有目标用户的用户画像数据的银行、保险等金融机构的服务平台。
存储器302包括第一存储器3021及第二存储器3022。在一些实施例中，第一存储器3021可为非易失性计算机可读存储介质，其上存储有操作系统、数据库及计算机可执行指令。计算机可执行指令可被处理器305所执行，用于实现如图1所示的实施例的个人财产状态评估方法。存储器302上的数据库用于存储各类数据，例如，上述个人财产状态评估方法中所涉及的各种数据，如地理位置信息、用户画像数据和取财产评估模型。第二存储器3021可为设备300的内存储器，为非易失性计算机可读存储介质中的操作系统、数据库和计算机可执行指令提供高速缓存的运行环境。
在本实施例中，处理器305是设备300的控制中心，利用各种接口和线路连接整个手机的各个部分，通过运行或执行存储在第一存储器3021内的计算机可执行搜集和/或数据库内的数据，执行设备300的各种功能和处理数据，从而对设备300进行整体监控。可选地，处理器305可包括一个或多个处理模块。
在本实施例中，通过执行存储该第一存储器3021内的计算机可执行指令和/或数据库内的数据，处理器305用于：获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；获取与所述目标用户相关联的用户画像数据；将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；利用所述种子用户的用户画像数据训练所述财产评估模型；根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
优选地，所述利用所述种子用户的用户画像数据训练所述财产评估模型，包括：
采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干所述子集群及每一所述子集群对应的共有画像数据；
获取每一所述种子用户的个人财产状态，并计算每一所述子集群中所有种子用户的个人财产均值；
将每一所述子集群中所述共有画像数据与所述个人财产均值进行逻辑回归处理，以获取所述财产评估模型。
优选地，所述根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果，包括：
采用相似度算法计算所述候选用户的用户画像数据与每一所述子集群的所述共有画 像数据的相似度；
判断所述相似度是否大于预设的相似阈值；
若是，则将子集群对应的个人财产均值作为所述个人财产状态评估结果输出。
优选地，所述用户画像数据包括基于位置服务的地理位置信息，所述地理位置信息包括与时间相关联的POI信息。
优选地，所述获取目标区域的地理位置围栏信息，包括：
采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位置围栏信息。
本发明实施例的个人财产状态评估设备300，处理器305先获取目标区域的地理位置围栏信息(包括目标用户)，并获取与目标用户相关联的用户画像数据；将目标用户划分成种子用户和候选用户；采用种子用户的用户画像数据训练财产评估模型，并利用训练好的财产评估模型对候选用户的用户画像数据进行处理，输出候选用户的个人财产评估结果。采用目标区域的种子用户的用户画像数据训练财产评估模型，利用训练好的财产评估模型对目标区域的候选用户进行个人财产评估，使得个人财产状态评估结果具有较高的准确性、客观性和可靠性。
第四实施例
本实施例提供一种非易失性计算机可读存储介质。该非易失性计算机可读存储介质用于存储一个或多个计算机可执行指令。具体地，计算机可执行指令被一个或多个处理器执行，使得所述一个或多个处理器执行第一实施例所述个人财产状态评估方法，为避免重复，这里不再赘述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的模块及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述模块的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个模块或组件可以结合 或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或模块的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理模块，即可以位于一个地方，或者也可以分布到多个网络模块上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能模块可以集成在一个处理模块中，也可以是各个模块单独物理存在，也可以两个或两个以上模块集成在一个模块中。
所述功能如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以权利要求的保护范围为准。

Claims (19)

  1. 一种个人财产状态评估方法，其特征在于，包括：
    获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
    获取与所述目标用户相关联的用户画像数据；
    将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
    利用所述种子用户的用户画像数据训练所述财产评估模型；
    根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
  2. 根据权利要求1所述的个人财产状态评估方法，其特征在于，所述利用所述种子用户的用户画像数据训练所述财产评估模型，包括：
    采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干所述子集群及每一所述子集群对应的共有画像数据；
    获取每一所述种子用户的个人财产状态，并计算每一所述子集群中所有种子用户的个人财产均值；
    将每一所述子集群中所述共有画像数据与所述个人财产均值进行逻辑回归处理，以获取所述财产评估模型。
  3. 根据权利要求2所述的个人财产状态评估方法，其特征在于，所述根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果，包括：
    采用相似度算法计算所述候选用户的用户画像数据与每一所述子集群的所述共有画像数据的相似度；
    判断所述相似度是否大于预设的相似阈值；
    若是，则将子集群对应的个人财产均值作为所述个人财产状态评估结果输出。
  4. 根据权利要求1所述的个人财产状态评估方法，其特征在于，所述用户画像数据包括基于位置服务的地理位置信息，所述地理位置信息包括与时间相关联的POI信息。
  5. 根据权利要求1所述的个人财产状态评估方法，其特征在于，所述获取目标区域的地理位置围栏信息，包括：
    采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位置围栏信息。
  6. 一种个人财产状态评估装置，其特征在于，包括：
    围栏信息获取模块，用于获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
    画像数据获取模块，用于获取与所述目标用户相关联的用户画像数据；
    数据集划分模块，用于将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
    评估模型训练模块，用于利用所述种子用户的用户画像数据训练所述财产评估模型；
    财产状态评估模块，用于根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
  7. 根据权利要求6所述的个人财产状态评估装置，其特征在于，所述评估模型训练模块包括：
    画像数据分类单元，用于采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干所述子集群及每一所述子集群对应的共有画像数据；
    财产均值计算单元，用于获取每一所述种子用户的个人财产状态，并计算每一所述子集群中所有种子用户的个人财产均值；
    评估模型处理单元，用于将每一所述子集群中所述共有画像数据与所述个人财产均值进行逻辑回归处理，以获取所述财产评估模型。
  8. 根据权利要求7所述的个人财产状态评估装置，其特征在于，所述财产状态评估模块包括：
    相似度计算单元，用于采用相似度算法计算所述候选用户的用户画像数据与每一所述子集群的所述共有画像数据的相似度；
    相似度比较单元，用于判断所述相似度是否大于预设的相似阈值；
    评估结果输出单元，用于若是，则将子集群对应的个人财产均值作为所述个人财产状态评估结果输出。
  9. 根据权利要求6所述的个人财产状态评估装置，其特征在于，所述用户画像数据包括基于位置服务的地理位置信息，所述地理位置信息包括与时间相关联的POI信息。
  10. 根据权利要求6所述的个人财产状态评估装置，其特征在于，所述围栏信息获取模块，用于采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位 置围栏信息。
  11. 一种个人财产状态评估设备，其特征在于，包括处理器及存储器，所述存储器存储有计算机可执行指令，所述处理器用于执行所述计算机可执行指令以执行如下步骤：
    获取目标区域的地理位置围栏信息，所述地理位置围栏信息包括房屋信息和对应的目标用户；
    获取与所述目标用户相关联的用户画像数据；
    将所有目标用户划分成种子数据集和候选数据集；所述种子数据集包括至少一个种子用户，所述候选数据集包括至少一个候选用户；
    利用所述种子用户的用户画像数据训练所述财产评估模型；
    根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果。
  12. 根据权利要求11所述的设备，其特征在于，所述利用所述种子用户的用户画像数据训练所述财产评估模型，包括：
    采用look-alike算法对所有种子用户的用户画像数据进行分类，获取若干所述子集群及每一所述子集群对应的共有画像数据；
    获取每一所述种子用户的个人财产状态，并计算每一所述子集群中所有种子用户的个人财产均值；
    将每一所述子集群中所述共有画像数据与所述个人财产均值进行逻辑回归处理，以获取所述财产评估模型。
  13. 根据权利要求12所述的设备，其特征在于，所述根据所述候选用户的用户画像数据，利用所述财产评估模型对所述候选用户的个人财产状态进行评估，以输出个人财产状态评估结果，包括：
    采用相似度算法计算所述候选用户的用户画像数据与每一所述子集群的所述共有画像数据的相似度；
    判断所述相似度是否大于预设的相似阈值；
    若是，则将子集群对应的个人财产均值作为所述个人财产状态评估结果输出。
  14. 根据权利要求11所述的设备，其特征在于，所述用户画像数据包括基于位置服务的地理位置信息，所述地理位置信息包括与时间相关联的POI信息。
  15. 根据权利要求11所述的设备，其特征在于，所述获取目标区域的地理位置围栏信息，包括：
    采用网络爬虫爬取房产中介平台和/或房产登记平台，以获取目标区域的地理位置围栏信息。
  16. 根据权利要求11所述的设备，其特征在于，所述设备还包括与所述处理器相连的网络接口；所述网络接口与远端存储设备和外部显示设备相连；所述网络接口用于接收所述远端存储设备发送的所述地理位置围栏信息和所述用户画像数据，并将所述地理位置围栏信息和所述用户画像数据发送给所述处理器；还用于接收所述处理器发送的所述个人财产状态评估结果，并将所述个人财产状态评估结果发送给所述外部显示设备。
  17. 根据权利要求11所述的设备，其特征在于，所述设备还包括与所述处理器相连的用户交互装置，所述用户交互装置用于接收用户输入的财产评估指令，并将所述财产评估指令发送给所述处理器，所述财产评估指令包括候选用户的用户画像数据；
    所述处理器，用于接收所述财产评估指令，并基于所述财产评估指令获取所述个人财产状态评估结果，并将所述个人财产状态评估结果发送给所述用户交互装置；
    所述用户交互装置，还用于接收并显示所述个人财产状态评估结果。
  18. 根据权利要求11所述的设备，其特征在于，所述存储器中存储有数据库，用于存储所述地理位置围栏信息、所述用户画像数据和所述财产评估模型。
  19. 一种非易失性计算机可读存储介质，其特征在于，用于存储一个或多个计算机可执行指令，所述计算机可执行指令被一个或多个处理器执行，使得所述一个或多个处理器执行权利要求1-5任一项所述个人财产状态评估方法。
PCT/CN2017/076463 2016-12-29 2017-03-13 个人财产状态评估方法、装置、设备和存储介质 WO2018120425A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611246227.X 2016-12-29
CN201611246227.XA CN107666649A (zh) 2016-12-29 2016-12-29 个人财产状态评估方法及装置

Publications (1)

Publication Number Publication Date
WO2018120425A1 true WO2018120425A1 (zh) 2018-07-05

Family

ID=61122475

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/076463 WO2018120425A1 (zh) 2016-12-29 2017-03-13 个人财产状态评估方法、装置、设备和存储介质

Country Status (2)

Country Link
CN (1) CN107666649A (zh)
WO (1) WO2018120425A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109858957A (zh) * 2019-01-17 2019-06-07 深圳壹账通智能科技有限公司 理财产品推荐方法、装置、计算机设备及存储介质
CN110442761A (zh) * 2019-06-21 2019-11-12 深圳中琛源科技股份有限公司 一种用户画像构建方法、电子设备及存储介质
CN111049664A (zh) * 2018-10-11 2020-04-21 中兴通讯股份有限公司 一种网络告警处理方法、装置及存储介质
CN111506733A (zh) * 2020-05-29 2020-08-07 广东太平洋互联网信息服务有限公司 对象画像的生成方法、装置、计算机设备和存储介质
CN112925982A (zh) * 2021-03-12 2021-06-08 上海意略明数字科技股份有限公司 用户重定向方法及装置、存储介质、计算机设备
CN117271905A (zh) * 2023-11-21 2023-12-22 杭州小策科技有限公司 基于人群画像的侧向需求分析方法及系统

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110781374A (zh) * 2018-07-13 2020-02-11 北京字节跳动网络技术有限公司 用户数据匹配方法、装置、电子设备和计算机可读介质
CN110781221A (zh) * 2019-09-27 2020-02-11 同济大学 一种法院被执行人隐匿财产估算决策支持系统架构
CN112383544B (zh) * 2020-11-13 2023-03-24 西安热工研究院有限公司 适用于电力scada的基于业务行为画像的反爬虫方法
CN113065739B (zh) * 2021-02-24 2023-07-04 广州互联网法院 被执行人的履行能力评估方法、装置及电子设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104866969A (zh) * 2015-05-25 2015-08-26 百度在线网络技术（北京）有限公司 个人信用数据处理方法和装置
CN104965913A (zh) * 2015-07-03 2015-10-07 重庆邮电大学 一种基于gps地理位置数据挖掘的用户分类方法
CN105894089A (zh) * 2016-04-21 2016-08-24 百度在线网络技术（北京）有限公司 一种征信模型的建立方法、征信确定方法及对应装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104866969A (zh) * 2015-05-25 2015-08-26 百度在线网络技术（北京）有限公司 个人信用数据处理方法和装置
CN104965913A (zh) * 2015-07-03 2015-10-07 重庆邮电大学 一种基于gps地理位置数据挖掘的用户分类方法
CN105894089A (zh) * 2016-04-21 2016-08-24 百度在线网络技术（北京）有限公司 一种征信模型的建立方法、征信确定方法及对应装置

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111049664A (zh) * 2018-10-11 2020-04-21 中兴通讯股份有限公司 一种网络告警处理方法、装置及存储介质
CN109858957A (zh) * 2019-01-17 2019-06-07 深圳壹账通智能科技有限公司 理财产品推荐方法、装置、计算机设备及存储介质
CN110442761A (zh) * 2019-06-21 2019-11-12 深圳中琛源科技股份有限公司 一种用户画像构建方法、电子设备及存储介质
CN111506733A (zh) * 2020-05-29 2020-08-07 广东太平洋互联网信息服务有限公司 对象画像的生成方法、装置、计算机设备和存储介质
CN112925982A (zh) * 2021-03-12 2021-06-08 上海意略明数字科技股份有限公司 用户重定向方法及装置、存储介质、计算机设备
CN112925982B (zh) * 2021-03-12 2023-04-07 上海意略明数字科技股份有限公司 用户重定向方法及装置、存储介质、计算机设备
CN117271905A (zh) * 2023-11-21 2023-12-22 杭州小策科技有限公司 基于人群画像的侧向需求分析方法及系统
CN117271905B (zh) * 2023-11-21 2024-02-09 杭州小策科技有限公司 基于人群画像的侧向需求分析方法及系统

Also Published As

Publication number Publication date
CN107666649A (zh) 2018-02-06

Similar Documents

Publication Publication Date Title
WO2018120425A1 (zh) 个人财产状态评估方法、装置、设备和存储介质
WO2018120424A1 (zh) 基于位置服务的人群分类方法、装置、设备和存储介质
WO2018120427A1 (zh) 基于位置服务的风险评估方法、装置、设备和存储介质
US11301729B2 (en) Systems and methods for inferential sharing of photos
JP6759844B2 (ja) 画像を施設に対して関連付けるシステム、方法、プログラム及び装置
WO2018120428A1 (zh) 个性化场景预测方法、装置、设备和存储介质
WO2018120426A1 (zh) 基于位置服务的个人健康状态评估方法、装置、设备和存储介质
CN109918669B (zh) 实体确定方法、装置及存储介质
US10922206B2 (en) Systems and methods for determining performance metrics of remote relational databases
JP2017045435A (ja) ソーシャルメディアメッセージ及び施設の間のリンクを推定する方法、コンピュータシステム、及びプログラム
US11663282B2 (en) Taxonomy-based system for discovering and annotating geofences from geo-referenced data
JP6911603B2 (ja) ユーザによって訪問される施設のカテゴリの予測モデルを生成する方法、プログラム、サーバ装置、及び処理装置
US11538121B2 (en) Location-based verification of user requests and generation of notifications on mobile devices
KR20170124581A (ko) 특정 컨텍스트에 대한 사용자 요구의 예측
US20210217093A1 (en) A system and method for protection plans and warranty data analytics
WO2021120875A1 (zh) 搜索方法、装置、终端设备及存储介质
CN111914113A (zh) 一种图像检索的方法以及相关装置
CN113505256B (zh) 特征提取网络训练方法、图像处理方法及装置
WO2020000715A1 (zh) 基于指数特征提取的股指预测方法、服务器及存储介质
CN112000264B (zh) 菜品信息展示方法、装置、计算机设备及存储介质
CN107807940B (zh) 信息推荐方法和装置
CN116307394A (zh) 产品用户体验评分方法、装置、介质及设备
CN115758271A (zh) 数据处理方法、装置、计算机设备和存储介质
US20210248177A1 (en) Keyword Localization Digital Image Search
WO2021000084A1 (zh) 数据分类方法及相关产品

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: 17889097

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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 1205 DATED 13/08/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 17889097

Country of ref document: EP

Kind code of ref document: A1