Method, device and storage medium for predicting permanent residence of guest user

Abstract

This invention relates to user data management technology in the financial industry, and discloses a method for predicting the permanent residence of customers. The method includes: extracting latitude and longitude information and corresponding time-series information from the customer's embedded data; performing address matching on the latitude and longitude information using a pre-set map library to obtain address information, and extracting location features from the address information; classifying the address information according to the time-series information and location features to obtain a first address dataset; performing clustering operations on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and filtering out location center data from the second address dataset; updating historical location data based on the location center data to obtain the customer's permanent residence. Furthermore, this invention also relates to blockchain technology, where customer historical data can be stored on blockchain nodes. This invention also proposes a device, electronic device, and storage medium for predicting the permanent residence of customers. This invention can improve the accuracy of predicting the permanent residence of financial and insurance users.

Description

Technical Field

[0001] This invention relates to the field of user data management technology, and in particular to a method, apparatus, electronic device, and computer-readable storage medium for predicting the permanent residence of a customer. Background Technology

[0002] Customer location information is crucial for the financial industry, serving as the foundation for subsequent business operations. For financial institutions, comprehensive, accurate, and timely access to customer location information is beneficial for building strong relationships with customers, thereby improving customer satisfaction with policy maintenance in financial and insurance services. Therefore, it is necessary to predict customer locations.

[0003] Existing methods for predicting permanent residence are mostly based on the permanent residence information that customers fill in themselves when purchasing products or registering applications, such as the address information filled in when purchasing financial services such as insurance and securities. In practice, the permanent residence information filled in by customers is somewhat arbitrary and cannot be verified. Furthermore, when the permanent residence of customers changes, the information cannot be updated in a timely manner, which may result in low accuracy when predicting the permanent residence of customers. Summary of the Invention

[0004] This invention provides a method, apparatus, and computer-readable storage medium for predicting the permanent residence of customers, with the main objective of solving the problem of low accuracy in predicting the permanent residence of financial and insurance customers.

[0005] To achieve the above objectives, the present invention provides a method for predicting the permanent residence of a customer, comprising:

[0006] Acquire customer tracking data, and extract latitude and longitude information and corresponding time series information from the tracking data;

[0007] The latitude and longitude information is matched with addresses using a preset map library to obtain address information, and the location features are extracted from the address information.

[0008] The address information is classified according to the time series information to obtain a time series address dataset. The address information is classified according to the location features to obtain a location address dataset. The time series address dataset and the location address dataset are then merged into a first address dataset.

[0009] Clustering operations are performed on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and location center data are filtered out from the second address dataset;

[0010] Historical location data is extracted from preset customer historical data, and the historical location data is updated according to the location center data to obtain the customer's permanent residence.

[0011] Optionally, extracting latitude and longitude information and corresponding time series information from the embedded data includes:

[0012] The location string is extracted from the embedded data according to the preset location keywords;

[0013] The latitude and longitude strings are extracted from the location string, and the latitude and longitude strings are converted to obtain latitude and longitude information;

[0014] The timing string is extracted from the positioning string, and the timing string is converted to obtain timing information.

[0015] Optionally, extracting location features from the address information includes:

[0016] Select each address text in the address information as the target address text;

[0017] The target address text is segmented to obtain a target address word set;

[0018] The target address word set is vectorized to obtain an address word vector set;

[0019] The address word vector set is matched with a preset location word vector library to obtain location features.

[0020] Optionally, the step of matching the address word vector set with a preset location word vector library to obtain location features includes:

[0021] One by one, the address word vectors in the address word vector set are selected as the target address word vectors, and one by one, the place word vectors in the place word vector library are selected as the target place word vectors;

[0022] Calculate the similarity between the target address word vector and the target location word vector;

[0023] Select the target location word vector with the highest similarity to the target address word vector as the similar word vector, and determine whether the similarity between the target address word vector and the similar word vector is greater than a preset similarity threshold;

[0024] When the similarity is less than or equal to the similarity threshold, return to the step of selecting address word vectors from the address word vector set one by one as the target address word vector;

[0025] When the similarity is greater than the similarity threshold, the similar word vectors are used as the address features of the target address word vectors, and all the address features are aggregated into location features.

[0026] Optionally, classifying the address information based on the time-series information to obtain a time-series address dataset includes:

[0027] Extract the year, month, and day information from the time sequence information, and divide the address information into weekday addresses and weekend addresses based on the year, month, and day information;

[0028] Hour, minute, and second information is extracted from the time sequence information. The weekday address is divided into a time sequence work address, a first leisure address, and a first home address based on the hour, minute, and second information. The rest day address is divided into a second leisure address and a second home address based on the hour, minute, and second information.

[0029] The first leisure address and the second leisure address are merged into a time-series leisure address, the first home address and the second home address are merged into a time-series home address, and the time-series work address, time-series leisure address and time-series home address are aggregated into a time-series address dataset.

[0030] Optionally, classifying the address information based on the location features to obtain a location address dataset includes:

[0031] The characteristics of these locations are categorized into workplace characteristics, family-oriented characteristics, and entertainment-oriented characteristics.

[0032] The address information corresponding to the characteristics of the workplace is used as the workplace address, the address information corresponding to the characteristics of the home is used as the home address, and the address information corresponding to the characteristics of the entertainment venue is used as the entertainment address.

[0033] The venue's work address, home address, and entertainment address are compiled into a venue address dataset.

[0034] Optionally, the step of clustering the latitude and longitude information based on the first address dataset to obtain the second address dataset includes:

[0035] Extract the latitude and longitude point set from the latitude and longitude information, and divide the latitude and longitude point set into multiple location latitude and longitude point groups according to the first address dataset;

[0036] Each of the location latitude and longitude points is selected as the target latitude and longitude point group. One of the latitude and longitude points in the target latitude and longitude point group is randomly selected as the initial latitude and longitude center point. All the initial latitude and longitude center points are then combined into an initial center point set.

[0037] Select latitude and longitude points in the set of latitude and longitude points one by one as target latitude and longitude points, and calculate the clustering distance between the target latitude and longitude points and each initial latitude and longitude center point in the initial center point set;

[0038] The initial latitude and longitude center point with the smallest clustering distance to the target latitude and longitude point is selected as the nearest center point of the target latitude and longitude point, and the target latitude and longitude point is divided into the location latitude and longitude point group corresponding to the nearest center point. When the target latitude and longitude point is the last latitude and longitude point in the latitude and longitude point set, all the clustered location latitude and longitude point groups are gathered into a standard latitude and longitude point set.

[0039] Select the location latitude and longitude point group in the standard latitude and longitude point set one by one as the target standard point group, calculate the secondary center point of the target standard point group, and calculate the center distance between the secondary center point and the initial latitude and longitude center point of the target standard point group. The sum of all the center distances is used as the modulus center distance.

[0040] Determine whether the center distance of the model is greater than a preset center distance threshold;

[0041] When the distance between the model centers is greater than the center distance threshold, the secondary center point is used as the initial latitude and longitude center point, and the process of selecting latitude and longitude points in the latitude and longitude point set one by one as the target latitude and longitude point is returned.

[0042] When the center distance of the modulus is less than or equal to the center distance threshold, the standard latitude and longitude point set at this time is used as the second address dataset.

[0043] To address the above problems, the present invention also provides a device for predicting the permanent residence of a customer, the device comprising:

[0044] The data acquisition module is used to acquire the embedded data of customer users and extract latitude and longitude information and corresponding time series information from the embedded data.

[0045] The location feature module is used to perform address matching on the latitude and longitude information using a preset map library to obtain address information, and extract location features from the address information;

[0046] The address classification module is used to classify the address information according to the time series information to obtain a time series address dataset, classify the address information according to the location features to obtain a location address dataset, and merge the time series address dataset and the location address dataset into a first address dataset.

[0047] The address clustering module is used to perform clustering operations on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and to filter out the location center data from the second address dataset.

[0048] The address update module is used to extract historical location data from preset customer historical data, update the historical location data according to the location center data, and obtain the customer's permanent residence.

[0049] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0050] At least one processor; and,

[0051] A memory communicatively connected to the at least one processor; wherein,

[0052] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the aforementioned method for predicting the permanent residence of customers.

[0053] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the aforementioned method for predicting the permanent residence of a customer.

[0054] This invention, through acquiring customer tracking data, extracts latitude and longitude information and corresponding time-series information from the tracking data to obtain the customer's location and time data, facilitating subsequent prediction of their permanent residence. By using a preset map library to match the latitude and longitude information with addresses, address information is obtained, enabling the acquisition of text information about the customer's long-term residence area. By extracting location features from the address information, it is convenient to determine the address category based on the location features, resulting in a more accurate customer movement trajectory. By classifying the address information according to the time-series information, a time-series address dataset is obtained. By classifying the address information according to the location features, a location address dataset is obtained. The time-series address dataset and the location address dataset are then merged into a first address dataset. This allows for address classification in both the time and address feature dimensions, resulting in more accurate address classification. By clustering the latitude and longitude information based on the first address dataset, a second address dataset is obtained, providing more accurate information about the customer's permanent residence. By filtering location center data from the second address dataset, the center of the customer's permanent residence can be determined. By extracting historical location data from preset customer historical data and updating the historical location data based on the location center data, the customer's permanent residence is obtained. This improves the accuracy and authenticity of customer address data, facilitating subsequent recommendations of financial products, policy maintenance, and the creation of user profiles. Therefore, the customer's permanent residence prediction method, device, electronic device, and computer-readable storage medium proposed in this invention can solve the problem of low accuracy in predicting the permanent residence of financial insurance users. Attached Figure Description

[0055] Figure 1 This is a flowchart illustrating a method for predicting the permanent residence of customers according to an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the process for generating location features according to an embodiment of the present invention;

[0057] Figure 3 This is a schematic diagram of the process for generating a second address dataset according to an embodiment of the present invention;

[0058] Figure 4 A functional block diagram of a customer's permanent residence prediction device provided in an embodiment of the present invention;

[0059] Figure 5 This is a schematic diagram of the structure of an electronic device for implementing the customer's permanent residence prediction method according to an embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0062] This application provides a method for predicting the permanent residence of a customer. The execution entity of the method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for predicting the permanent residence of a customer can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0063] Reference Figure 1 The diagram shown is a flowchart illustrating a method for predicting the permanent residence of customers according to an embodiment of the present invention. In this embodiment, the method for predicting the permanent residence of customers includes:

[0064] S1. Obtain the customer's embedded data, and extract latitude and longitude information and corresponding time sequence information from the embedded data.

[0065] Specifically, the embedded data includes location data, mobile phone number data, and device model data.

[0066] In this embodiment of the invention, the customer user can be a financial insurance user. The process of acquiring the customer user's event tracking data includes: monitoring the customer user's application using a preset monitoring interface; when the application triggers an event tracking event, receiving the return value of the event tracking event using the monitoring interface, and aggregating all the return values ​​into event tracking data.

[0067] Specifically, the listening interface can be the IntersectionObserver interface.

[0068] Specifically, the application can be a financial securities or financial insurance application, such as a financial banking, financial health insurance, or financial securities application.

[0069] Specifically, the event tracking can be an application open event, an application use event, or an application close event. For example, when Pocket Bank is opened, the application automatically obtains location information.

[0070] In detail, the return value of the event tracking point can be received using the IntersectionObserver.takeRecords() method of the listening interface.

[0071] In detail, the step of extracting latitude and longitude information and corresponding time sequence information from the embedded data includes: extracting a positioning string from the embedded data according to a preset positioning keyword; extracting a latitude and longitude string from the positioning string, performing data conversion on the latitude and longitude string to obtain latitude and longitude information; and extracting a time sequence string from the positioning string, performing data conversion on the time sequence string to obtain time sequence information.

[0072] Specifically, the positioning keyword can be GPRMC.

[0073] In detail, regular expressions can be used to extract the location string from the embedded data based on preset location keywords.

[0074] Specifically, extracting the latitude and longitude string from the positioning string means taking the 3rd and 4th strings in the positioning string as the latitude string and the 5th and 6th strings as the longitude string, and combining the latitude and longitude strings into the latitude and longitude string.

[0075] In detail, latitude and longitude data processing tools such as kotlinDLLatLngUtil can be used to convert the latitude and longitude strings to obtain latitude and longitude information.

[0076] Specifically, extracting the time sequence string from the positioning string means using the 1st and 9th elements of the positioning string as the time sequence string. More specifically, performing data conversion on the time sequence string to obtain time sequence information means converting the time sequence string from the format of hours, minutes, seconds, day, month, and year into the format of year, month, day, hour, minute, and second.

[0077] In this embodiment of the invention, by acquiring the tracking data of customers in industries such as finance and insurance, latitude and longitude information and corresponding time series information are extracted from the tracking data, the movement location and time data of customers in industries such as finance and insurance can be obtained, which facilitates subsequent prediction of permanent residence.

[0078] S2. Use a preset map library to perform address matching on the latitude and longitude information to obtain address information, and extract location features from the address information.

[0079] In this embodiment of the invention, the map database can be a publicly available map database, such as an online map or an offline map.

[0080] In this embodiment of the invention, the step of using a preset map database to perform address matching on the latitude and longitude information to obtain address information refers to inputting the latitude and longitude information into the map database to obtain the corresponding address information.

[0081] In detail, the step of extracting location features from the address information includes: selecting address text from the address information one by one as target address text; performing text segmentation on the target address text to obtain a target address word set; performing vectorization on the target address word set to obtain an address word vector set; and matching the address word vector set with a preset location word vector library to obtain location features.

[0082] In detail, the target address text can be segmented using a bidirectional maximum matching algorithm to obtain a target address word set.

[0083] Specifically, word2vec or one-hot encoding can be used to vectorize the target address word set to obtain an address word vector set.

[0084] In detail, the location word vector library refers to a database containing word vectors of multiple location keywords, such as "residential community," "primary school," and "square," which are words related to locations.

[0085] Specifically, refer to Figure 2 As shown, the step of matching the address word vector set with a preset location word vector library to obtain location features includes:

[0086] S21. Select address word vectors from the address word vector set one by one as target address word vectors, and select place word vectors from the place word vector library one by one as target place word vectors;

[0087] S22. Calculate the similarity between the target address word vector and the target location word vector;

[0088] S23. Select the target location word vector with the highest similarity to the target address word vector as the similar word vector;

[0089] S24. Determine whether the similarity between the target address word vector and the similar word vector is greater than a preset similarity threshold;

[0090] S25. When the similarity is less than or equal to the similarity threshold, return to the step of selecting address word vectors from the address word vector set one by one as the target address word vector;

[0091] S26. When the similarity is greater than the similarity threshold, the similar word vector is used as the address feature of the target address word vector, and all the address features are aggregated into location features.

[0092] For example, in this embodiment of the invention, the address word vector corresponding to "Sunflower Community" can be selected from the address word vector set as the target address word vector, and the location word vector corresponding to "XX Community" can be selected sequentially from the location word vector library as the target location word vector. Furthermore, this embodiment of the invention can use the cosine similarity formula to calculate the similarity between the target address word vector and the target location word vector.

[0093] Specifically, the similarity threshold can be 0.85 or 0.9.

[0094] In this embodiment of the invention, by using a preset map library to match the latitude and longitude information to obtain address information, text information of the area where the customer stays for a long time can be obtained. By extracting the location features from the address information, it is convenient to determine the category of the address based on the location features, and obtain a more accurate customer movement trajectory.

[0095] S3. Classify the address information according to the time series information to obtain a time series address dataset, classify the address information according to the location characteristics to obtain a location address dataset, and merge the time series address dataset and the location address dataset into a first address dataset.

[0096] In this embodiment of the invention, classifying the address information according to the time-series information to obtain a time-series address dataset includes: extracting year, month, and day information from the time-series information; dividing the address information into weekday addresses and weekend addresses according to the year, month, and day information; extracting hour, minute, and second information from the time-series information; dividing the weekday addresses into time-series work addresses, a first leisure address, and a first home address according to the hour, minute, and second information; dividing the weekend addresses into a second leisure address and a second home address according to the hour, minute, and second information; merging the first leisure address and the second leisure address into a time-series leisure address; merging the first home address and the second home address into a time-series home address; and aggregating the time-series work addresses, time-series leisure addresses, and time-series home addresses into a time-series address dataset.

[0097] In detail, the year, month, and day information can be extracted from the time series information using the select language or regular expressions.

[0098] Specifically, dividing the address information into weekday addresses and rest day addresses based on the year, month, and day information means taking the address information corresponding to a weekday as the weekday address and the address information corresponding to a rest day as the rest day address.

[0099] In detail, the weekday address is divided into a time-series work address, a first leisure address, and a first home address based on the time, minute, and second information. For example, the weekday addresses corresponding to 8 to 12 o'clock and 14 to 19 o'clock are used as time-series work addresses, the weekday addresses corresponding to 12 to 14 o'clock are used as first leisure addresses, and the weekday addresses corresponding to 21 to 8 o'clock the next day are used as first home addresses.

[0100] Specifically, the rest day address is divided into a second leisure address and a second home address based on the time, minute, and second information. For example, the rest day address corresponding to 10:00 to 21:00 is used as the second leisure address, and the rest day address corresponding to 21:00 to 10:00 is used as the second home address.

[0101] In detail, the step of classifying the address information according to the location features to obtain a location address dataset includes: classifying the location features into work-type location features, home-type location features, and entertainment-type location features; using the address information corresponding to the work-type location features as the location work address, the address information corresponding to the home-type location features as the location home address, and the address information corresponding to the entertainment-type location features as the location entertainment address; and aggregating the location work addresses, location home addresses, and location entertainment addresses into a location address dataset.

[0102] In detail, the location features are divided into workplace features, family features, and entertainment features. For example, the word vectors of words such as "building" and "limited company" are used as workplace features, the word vectors of words such as "residential community" and "residential garden" are used as family features, and the word vectors of words such as "shopping center" and "park" are used as entertainment features.

[0103] In detail, the step of merging the time-series address dataset and the location address dataset into a first address dataset includes: using overlapping addresses from the time-series work addresses and the location work addresses as initial work addresses; using overlapping addresses from the time-series home addresses and the location home addresses as initial home addresses; using overlapping addresses from the time-series entertainment addresses and the location entertainment addresses as initial entertainment addresses; and aggregating the initial work addresses, the initial home addresses, and the initial entertainment addresses into a first address dataset.

[0104] In this embodiment of the invention, the address information is classified according to the time-series information to obtain a time-series address dataset, and the address information is classified according to the location features to obtain a location address dataset. The time-series address dataset and the location address dataset are then merged into a first address dataset. This allows for address classification in both the time dimension and the address feature dimension, resulting in more accurate address classification.

[0105] S4. Perform clustering operation on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and filter out the location center data from the second address dataset.

[0106] In this embodiment of the invention, reference is made to Figure 3 As shown, the step of clustering the latitude and longitude information based on the first address dataset to obtain the second address dataset includes:

[0107] S31. Extract the latitude and longitude point set from the latitude and longitude information, and divide the latitude and longitude point set into multiple location latitude and longitude point groups according to the first address dataset;

[0108] S32. Select each of the location latitude and longitude points as the target latitude and longitude point group, randomly select one of the latitude and longitude points in the target latitude and longitude point group as the initial latitude and longitude center point, and gather all the initial latitude and longitude center points into an initial center point set;

[0109] S33. Select latitude and longitude points in the latitude and longitude point set one by one as target latitude and longitude points, and calculate the clustering distance between the target latitude and longitude points and each initial latitude and longitude center point in the initial center point set.

[0110] S34. Select the initial latitude and longitude center point with the smallest clustering distance to the target latitude and longitude point as the nearest center point of the target latitude and longitude point, and divide the target latitude and longitude point into the location latitude and longitude point group corresponding to the nearest center point, until the target latitude and longitude point is the last latitude and longitude point in the latitude and longitude point set, and then gather all the location latitude and longitude point groups after clustering into a standard latitude and longitude point set.

[0111] S35. Select the location latitude and longitude point group in the standard latitude and longitude point set one by one as the target standard point group, calculate the secondary center point of the target standard point group, and calculate the center distance between the secondary center point and the initial latitude and longitude center point of the target standard point group. The sum of all the center distances is used as the modulus center distance.

[0112] S36. Determine whether the center distance of the mold taking is greater than a preset center distance threshold;

[0113] S37. When the distance between the model centers is greater than the center distance threshold, the secondary center point is used as the initial latitude and longitude center point, and the process of selecting latitude and longitude points in the latitude and longitude point set one by one as the target latitude and longitude point is returned.

[0114] S38. When the center distance of the modulus is less than or equal to the center distance threshold, the standard latitude and longitude point set at this time is used as the second address dataset.

[0115] In detail, the location latitude and longitude point group refers to the combination of latitude and longitude points of the same location class.

[0116] Specifically, the latitude and longitude point set is divided into multiple location latitude and longitude point groups according to the first address dataset. For example, the latitude and longitude point corresponding to the initial working address of the first address dataset in the latitude and longitude set is taken as the working latitude and longitude point.

[0117] In detail, the secondary center point refers to the latitude and longitude point that has the same cluster distance as all latitude and longitude points in the target latitude and longitude point group.

[0118] In detail, the step of filtering out the location center data from the second address dataset means taking all the initial latitude and longitude center points in the second address dataset as location center points and aggregating all the location center points into location center data.

[0119] In this embodiment of the invention, by performing clustering operations on the latitude and longitude information based on the first address dataset to obtain the second address dataset, more accurate information about the permanent residence of the customer can be obtained. By filtering out the location center data from the second address dataset, the center of the customer's permanent residence can be determined.

[0120] S5. Extract historical location data from preset customer historical data, update the historical location data according to the location center data, and obtain the customer's permanent residence.

[0121] In this embodiment of the invention, the customer historical data refers to customer user information uploaded by customers or employees and stored in data storage media such as blockchain or servers. For example, customer user information uploaded by customers to financial insurance servers when purchasing financial products or insurance.

[0122] Specifically, the historical location data refers to the address-related data in the customer's historical data.

[0123] In this embodiment of the invention, updating the historical location data based on the location center data to obtain the permanent residence of the customer refers to extracting regional information from the location center data, updating the historical location data using the regional information, and using the updated historical location data as the permanent residence.

[0124] In this embodiment of the invention, by extracting historical location data from preset customer historical data and updating the historical location data according to the location center data, the permanent residence of the customer can be obtained, which can improve the accuracy and authenticity of customer address data, thereby enhancing the operational efficiency of insurance and other businesses.

[0125] This invention, through acquiring customer tracking data, extracts latitude and longitude information and corresponding time-series information from the tracking data to obtain the customer's location and time data, facilitating subsequent prediction of their permanent residence. By using a preset map library to match the latitude and longitude information with addresses, address information is obtained, enabling the acquisition of text information about the customer's long-term residence area. By extracting location features from the address information, it is convenient to determine the address category based on the location features, resulting in a more accurate customer movement trajectory.

[0126] By classifying the address information according to the time-series information, a time-series address dataset is obtained. Similarly, by classifying the address information according to the location features, a location address dataset is obtained. The time-series address dataset and the location address dataset are then merged into a first address dataset. This allows for address classification across both time and address feature dimensions, resulting in more accurate address classification. By clustering the latitude and longitude information using the first address dataset, a second address dataset is obtained, providing more accurate information about the customer's permanent residence. By filtering location center data from the second address dataset, the center of the customer's permanent residence can be determined. Furthermore, by extracting historical location data from preset customer historical data and updating the historical location data based on the location center data, the customer's permanent residence is obtained. This improves the accuracy and authenticity of customer address data. Accurate customer address data facilitates building good relationships with customers, thereby increasing customer satisfaction with policy maintenance in financial and insurance services. Therefore, the customer's permanent residence prediction method proposed in this invention solves the problem of low accuracy in predicting customer permanent residence.

[0127] like Figure 4 The diagram shown is a functional block diagram of a customer's permanent residence prediction device provided in an embodiment of the present invention.

[0128] The customer's permanent residence prediction device 100 of the present invention can be installed in an electronic device. Depending on the functions implemented, the customer's permanent residence prediction device 100 may include a data acquisition module 101, a location feature module 102, an address classification module 103, an address clustering module 104, and an address update module 105. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0129] In this embodiment, the functions of each module / unit are as follows:

[0130] The data acquisition module 101 is used to acquire the embedded data of customer users and extract latitude and longitude information and corresponding time series information from the embedded data.

[0131] The location feature module 102 is used to perform address matching on the latitude and longitude information using a preset map library to obtain address information and extract location features from the address information.

[0132] The address classification module 103 is used to classify the address information according to the time series information to obtain a time series address dataset, classify the address information according to the location features to obtain a location address dataset, and merge the time series address dataset and the location address dataset into a first address dataset.

[0133] The address clustering module 104 is used to perform clustering operations on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and to filter out the location center data from the second address dataset.

[0134] The address update module 105 is used to extract historical location data from preset customer historical data, update the historical location data according to the location center data, and obtain the customer's permanent residence.

[0135] In detail, the modules in the customer's permanent residence prediction device 100 described in this embodiment of the invention employ the same methods as described above during use. Figures 1 to 3 The method used is the same as the one described above for predicting the permanent residence of customers, and it can produce the same technical effect, so it will not be repeated here.

[0136] like Figure 5 The diagram shown is a schematic representation of an electronic device for implementing a method for predicting the permanent residence of customers according to an embodiment of the present invention.

[0137] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13. It may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a customer's permanent residence prediction program.

[0138] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a customer's permanent residence prediction program) and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0139] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a customer's permanent residence prediction program, but also to temporarily store data that has been output or will be output.

[0140] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0141] The communication interface 13 is used for communication between the aforementioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish communication connections between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), or, optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0142] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0143] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0144] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0145] The customer's permanent residence prediction program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions, which, when run in the processor 10, can achieve the following:

[0146] Acquire customer tracking data, and extract latitude and longitude information and corresponding time series information from the tracking data;

[0147] The latitude and longitude information is matched with addresses using a preset map library to obtain address information, and the location features are extracted from the address information.

[0148] The address information is classified according to the time series information to obtain a time series address dataset. The address information is classified according to the location features to obtain a location address dataset. The time series address dataset and the location address dataset are then merged into a first address dataset.

[0149] Clustering operations are performed on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and location center data are filtered out from the second address dataset;

[0150] Historical location data is extracted from preset customer historical data, and the historical location data is updated according to the location center data to obtain the customer's permanent residence.

[0151] Specifically, the specific implementation method of the processor 10 for the above instructions can be referred to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, and will not be repeated here.

[0152] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0153] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0154] Acquire customer tracking data, and extract latitude and longitude information and corresponding time series information from the tracking data;

[0155] The latitude and longitude information is matched with addresses using a preset map library to obtain address information, and the location features are extracted from the address information.

[0156] The address information is classified according to the time series information to obtain a time series address dataset. The address information is classified according to the location features to obtain a location address dataset. The time series address dataset and the location address dataset are then merged into a first address dataset.

[0157] Clustering operations are performed on the latitude and longitude information based on the first address dataset to obtain a second address dataset, and location center data are filtered out from the second address dataset;

[0158] Historical location data is extracted from preset customer historical data, and the historical location data is updated according to the location center data to obtain the customer's permanent residence.

[0159] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0160] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0161] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0162] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0163] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0164] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0165] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0166] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for predicting the permanent residence of a customer, characterized in that, The method includes: Acquire customer tracking data, and extract latitude and longitude information and corresponding time series information from the tracking data; The latitude and longitude information is matched with a pre-set map library to obtain address information. Address text is selected one by one from the address information as target address text. The target address text is segmented into words to obtain a target address word set. The target address word set is vectorized to obtain an address word vector set. The address word vector set is matched with a pre-set location word vector library to obtain the location features of the target address text. The location word vector library is a database containing word vectors of multiple location feature keywords. The address information is classified according to the time-series information to obtain a time-series address dataset, which includes time-series work addresses, time-series leisure addresses, and time-series home addresses. The address information is also classified according to the location characteristics to obtain a location address dataset, which includes location work addresses, location home addresses, and location entertainment addresses. Overlapping addresses between the time-series work addresses and location work addresses are used as initial work addresses. Overlapping addresses between the time-series home addresses and location home addresses are used as initial home addresses. Overlapping addresses between the time-series leisure addresses and location entertainment addresses are used as initial entertainment addresses. The initial work addresses, initial home addresses, and initial entertainment addresses are then aggregated into a first address dataset. Based on the first address dataset, the latitude and longitude information is classified into locations and clustered for each category to obtain a second address dataset. Then, the location center data is filtered out from the second address dataset. Historical location data is extracted from preset customer historical data, and the historical location data is updated according to the location center data to obtain the customer's permanent residence.

2. The method for predicting the permanent residence of customers as described in claim 1, characterized in that, The extraction of latitude and longitude information and corresponding time series information from the embedded data includes: The location string is extracted from the embedded data according to the preset location keywords; The latitude and longitude strings are extracted from the location string, and the latitude and longitude strings are converted to obtain latitude and longitude information; The timing string is extracted from the positioning string, and the timing string is converted to obtain timing information.

3. The method for predicting the permanent residence of customers as described in claim 1, characterized in that, The step of matching the address word vector set with a preset location word vector library to obtain location features includes: One by one, the address word vectors in the address word vector set are selected as the target address word vectors, and one by one, the place word vectors in the place word vector library are selected as the target place word vectors; Calculate the similarity between the target address word vector and the target location word vector; Select the target location word vector with the highest similarity to the target address word vector as the similar word vector, and determine whether the similarity between the target address word vector and the similar word vector is greater than a preset similarity threshold; When the similarity is less than or equal to the similarity threshold, return to the step of selecting address word vectors from the address word vector set one by one as the target address word vector; When the similarity is greater than the similarity threshold, the similar word vectors are used as the address features of the target address word vectors, and all the address features are aggregated into location features.

4. The method for predicting the permanent residence of customers as described in claim 1, characterized in that, The step of classifying the address information according to the time-series information to obtain a time-series address dataset includes: Extract the year, month, and day information from the time sequence information, and divide the address information into weekday addresses and weekend addresses based on the year, month, and day information; Hour, minute, and second information is extracted from the time sequence information. The weekday address is divided into a time sequence work address, a first leisure address, and a first home address based on the hour, minute, and second information. The rest day address is divided into a second leisure address and a second home address based on the hour, minute, and second information. The first leisure address and the second leisure address are merged into a time-series leisure address, the first home address and the second home address are merged into a time-series home address, and the time-series work address, time-series leisure address and time-series home address are aggregated into a time-series address dataset.

5. The method for predicting the permanent residence of customers as described in claim 1, characterized in that, The step of classifying the address information according to the location features to obtain a location address dataset includes: The characteristics of these locations are categorized into workplace characteristics, family-oriented characteristics, and entertainment-oriented characteristics. The address information corresponding to the characteristics of the workplace is used as the workplace address, the address information corresponding to the characteristics of the home is used as the home address, and the address information corresponding to the characteristics of the entertainment venue is used as the entertainment address. The venue's work address, home address, and entertainment address are compiled into a venue address dataset.

6. The method for predicting the permanent residence of customers as described in claim 1, characterized in that, The step of clustering the latitude and longitude information based on the first address dataset to obtain the second address dataset includes: Extract the latitude and longitude point set from the latitude and longitude information, and divide the latitude and longitude point set into multiple location latitude and longitude point groups according to the first address dataset; Each of the location latitude and longitude points is selected as the target latitude and longitude point group. One of the latitude and longitude points in the target latitude and longitude point group is randomly selected as the initial latitude and longitude center point. All the initial latitude and longitude center points are then combined into an initial center point set. Select latitude and longitude points in the set of latitude and longitude points one by one as target latitude and longitude points, and calculate the clustering distance between the target latitude and longitude points and each initial latitude and longitude center point in the initial center point set; The initial latitude and longitude center point with the smallest clustering distance to the target latitude and longitude point is selected as the nearest center point of the target latitude and longitude point, and the target latitude and longitude point is divided into the location latitude and longitude point group corresponding to the nearest center point. When the target latitude and longitude point is the last latitude and longitude point in the latitude and longitude point set, all the clustered location latitude and longitude point groups are gathered into a standard latitude and longitude point set. Select the location latitude and longitude point group in the standard latitude and longitude point set one by one as the target standard point group, calculate the secondary center point of the target standard point group, and calculate the center distance between the secondary center point and the initial latitude and longitude center point of the target standard point group. The sum of all the center distances is used as the modulus center distance. Determine whether the center distance of the model is greater than a preset center distance threshold; When the distance between the model centers is greater than the center distance threshold, the secondary center point is used as the initial latitude and longitude center point, and the process of selecting latitude and longitude points in the latitude and longitude point set one by one as the target latitude and longitude point is returned. When the center distance of the modulus is less than or equal to the center distance threshold, the standard latitude and longitude point set at this time is used as the second address dataset.

7. A device for predicting the permanent residence of a customer, characterized in that, The device includes: The data acquisition module is used to acquire the embedded data of customer users and extract latitude and longitude information and corresponding time series information from the embedded data. The location feature module is used to perform address matching on the latitude and longitude information using a preset map library to obtain address information, select address text from the address information one by one as target address text, perform text segmentation on the target address text to obtain a target address word set, perform vectorization operation on the target address word set to obtain an address word vector set, and match the address word vector set with a preset location word vector library to obtain the location features of the target address text. The location word vector library is a database containing word vectors of multiple location feature keywords. The address classification module is used to classify the address information according to the time-series information to obtain a time-series address dataset, which includes time-series work addresses, time-series leisure addresses, and time-series home addresses. It also classifies the address information according to the location characteristics to obtain a location address dataset, which includes location work addresses, location home addresses, and location entertainment addresses. The module uses overlapping addresses between the time-series work addresses and location work addresses as initial work addresses, overlapping addresses between the time-series home addresses and location home addresses as initial home addresses, and overlapping addresses between the time-series leisure addresses and location entertainment addresses as initial entertainment addresses. Finally, it aggregates the initial work addresses, initial home addresses, and initial entertainment addresses into a first address dataset. The address clustering module is used to classify the latitude and longitude information into locations based on the first address dataset and perform clustering operations on each category to obtain a second address dataset, and to filter out the location center data from the second address dataset. The address update module is used to extract historical location data from preset customer historical data, update the historical location data according to the location center data, and obtain the customer's permanent residence.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the customer's place of residence prediction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the customer's permanent residence prediction method as described in any one of claims 1 to 6.

Images