Multi-source data fusion-based real estate accurate customer extension method and multi-source data fusion-based real estate accurate customer extension system

Abstract

The invention relates to the technical field of marketing customer extension, and discloses a multi-source data fusion real estate accurate customer extension method and system, and the method comprises the steps: collecting access user information which comprises user basic information and user browsing data; preliminarily comparing the access user with a potential client library based on the user basic information, and establishing an access user portrait based on the corresponding information of the access user in the potential client library, the user basic information and the user browsing data when judging that the access user exists in the potential client library; when it is judged that the access user does not exist in the potential customer library, according to the user basic information and the user browsing data, carrying out coincidence comparison with preset user portraits, and selecting the preset user portraits with the highest coincidence degree as the access user portraits; and determining a corresponding advertisement strategy according to the access user portrait. According to the method and the system, repeated marketing can be reduced, and the transaction probability can be improved in marketing scheme selection.

Description

Technical Field

[0001] This application relates to the technical field of marketing and customer acquisition, and in particular to a method and system for precise customer acquisition in the real estate industry that integrates multi-source data. Background Technology

[0002] As competition in the real estate market intensifies, appropriate marketing and customer acquisition strategies are crucial for enhancing industry competitiveness. However, the accumulation of customer information can easily lead to a lack of focus in marketing and customer acquisition efforts, resulting in poor effectiveness. Therefore, it is necessary to integrate and analyze multiple data sources during the customer acquisition process to obtain precise customer acquisition methods.

[0003] Existing real estate customer acquisition methods employ different approaches depending on the source of information. For users acquired through software, public accounts, mini-programs, and websites, advertising is used for marketing. For external user data obtained in cooperation with compliant data service providers, methods such as telephone sales and information dissemination are employed. However, information from different sources often overlaps, and fragmented data cannot accurately assess the needs of potential customers, thus failing to guarantee the effectiveness of marketing and customer acquisition. Therefore, how to integrate multi-source data to improve the accuracy of marketing and customer acquisition is the fundamental problem that this invention aims to solve. Summary of the Invention

[0004] In order to improve the accuracy of marketing and customer acquisition by integrating multi-source data, this application provides a method and system for precise customer acquisition in the real estate industry through multi-source data fusion.

[0005] Firstly, this application provides a method for precise customer acquisition in the real estate market through multi-source data fusion, employing the following technical solution:

[0006] A method for precise customer acquisition in the real estate market that integrates multi-source data includes:

[0007] Collect access user information, which includes basic user information and user browsing data;

[0008] Based on the user's basic information, the visiting user is initially compared with the potential customer database. When it is determined that the visiting user exists in the potential customer database, a user profile is built based on the corresponding information of the visiting user in the potential customer database, the user's basic information, and the user's browsing data. When it is determined that the visiting user does not exist in the potential customer database, the user's basic information and user browsing data are compared with the preset user profile, and the preset user profile with the highest degree of overlap is selected as the visiting user profile.

[0009] Determine the corresponding advertising strategy based on the user profile of the visitor.

[0010] Optionally, when it is determined that the visiting user exists in the potential customer database, the process of building a visiting user profile includes:

[0011] Pre-define user needs;

[0012] Based on the corresponding information of the visiting users in the potential customer database and the user's basic information, extract user demand tags for each type of user demand direction;

[0013] Analyze user browsing data to obtain the weight coefficients for each type of user demand.

[0014] The demand percentage for each user demand direction is determined based on the user demand tags and weight coefficients for each user demand direction.

[0015] A user profile is created based on the proportion of all user needs.

[0016] Optionally, the process of obtaining the demand percentage for each user demand category includes:

[0017] Based on AI, the similarity of any two sets of user demand tags in each type of user demand direction is compared to obtain the similarity value;

[0018] Through equations Calculate the demand factor g for each type of user needs. i ; through equations Calculate the demand percentage p for each user demand direction. i ;

[0019] Where e is the natural constant, λ i Let n be the weight coefficient for the i-th type of user demand direction. i Let r be the number of user demand tags for the i-th type of user demand direction. j The similarity between any two groups of user demand tags in each user demand direction, 0≤r j <1, where γ is the adjustment coefficient; γ>1, where Z is the number of user demand directions.

[0020] Optionally, the process of obtaining the weight coefficients for each type of user demand includes:

[0021] The user browsing data is broken down to obtain the segmented browsing pages and their corresponding browsing durations;

[0022] Through equations Calculate the weight coefficient λ for the i-th type of user demand direction. i ;

[0023] Where, λt i Let d be the basic weight for the i-th type of user demand direction, Q be the number of pages to be browsed, k∈[1,Q], and d be the weight for the i-th type of user demand direction. ikt is the preset association value between the k-th split browsing page and the i-th type of user demand direction. k This represents the browsing time for the k-th split page.

[0024] Optionally, when it is determined that the accessing user does not exist in the potential customer database, the process of performing a coincidence comparison includes:

[0025] Through equations Calculate the deviation V between the accessing user and the x-th preset user profile. x Select the minimum deviation value V x The corresponding preset user profile is used as the visiting user profile;

[0026] Where B is the amount of basic user information, y∈[1,B], r xy Let β be the similarity between the basic information of the y-th user and the basic information of the user corresponding to the x-th preset user profile. y Let λ be the weight coefficient of the basic information of the y-th user, τ be the correction coefficient, i∈[1,Z], and λ be the weight coefficient of the basic information of the y-th user. ix Let α be the base value for the i-th user demand direction in the x-th preset user profile. i Let be the weight value for the i-th user's demand direction.

[0027] Optionally, the process of determining the corresponding advertising strategy based on the user profile includes:

[0028] Pre-set different advertising strategies based on different user needs;

[0029] The final advertising strategy is determined based on the proportion of demand for each user category in the user profile.

[0030] Optionally, the process of determining the final advertising strategy based on the user profile of the visitor also includes:

[0031] Through equations Calculate the selection coefficient u for the i-th user demand direction i According to the maximum selection coefficient u i The corresponding advertising strategy serves as the final advertising strategy;

[0032] Where Z-1 represents the number of user demand directions other than the i-th user demand direction, s∈[1, Z-1], p s ω represents the proportion of demand for the s-th user demand direction. is Let be the correlation between the s-th user demand direction and the i-th user demand direction.

[0033] Secondly, this application provides a multi-source data fusion-based real estate precision customer acquisition system, which adopts the following technical solution:

[0034] A multi-source data fusion-based real estate customer acquisition system includes:

[0035] An information collection unit is used to collect access user information, which includes basic user information and user browsing data.

[0036] The user analysis unit is used to make a preliminary comparison between the visiting user and the potential customer database based on the user's basic information. When it is determined that the visiting user exists in the potential customer database, a user profile is built based on the corresponding information of the visiting user in the potential customer database, the user's basic information, and the user's browsing data. When it is determined that the visiting user does not exist in the potential customer database, the user's basic information and the user's browsing data are compared with the preset user profile, and the preset user profile with the highest degree of overlap is selected as the visiting user profile.

[0037] The promotion unit is used to determine the corresponding advertising strategy based on the user profile of the visitor.

[0038] In summary, this application includes at least one of the following beneficial technical effects:

[0039] This invention reduces repetitive marketing while providing more accurate and comprehensive user profiles that reflect user needs, thereby increasing the conversion rate in marketing strategy selection. Furthermore, it uses big data to accurately match user browsing data, further improving conversion rates during customer acquisition. Finally, it determines corresponding advertising strategies based on user profiles, achieving targeted recommendations of different ads based on user needs and preferences, thus enhancing the precision of customer acquisition marketing. Attached Figure Description

[0040] Figure 1 This is a flowchart of the real estate precision customer acquisition method based on multi-source data fusion of the present invention;

[0041] Figure 2 This is a logical diagram of the real estate precision customer acquisition system based on multi-source data fusion of the present invention. Detailed Implementation

[0042] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0043] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0044] This application discloses a method for precise customer acquisition in the real estate market based on multi-source data fusion, referring to... Figure 1 The process includes: collecting user information, which includes basic user information and browsing data; performing a preliminary comparison between the user and a potential customer database based on the basic user information; when the user is found to exist in the potential customer database, establishing a user profile based on the user's corresponding information in the database, basic user information, and browsing data; when the user is found not to exist in the database, performing a comparison of the user's basic information and browsing data with a preset user profile, and selecting the preset user profile with the highest overlap as the user profile; and determining a corresponding advertising strategy based on the user profile. This customer acquisition method enables the comparison of multiple data sources to determine whether the user's information overlaps with the information in the potential customer database. When the user is found to exist in the database, a corresponding advertising strategy is established based on the user profile. By accessing users' information in the potential customer database, their basic information, and their browsing data to create user profiles, we can reduce duplicate marketing and obtain more accurate and comprehensive user profiles that reflect user needs. This, in turn, increases the likelihood of conversion when choosing marketing strategies. When a user is determined not to exist in the potential customer database, we compare their basic information and browsing data with preset user profiles and select the one with the highest overlap. This process leverages big data to make more accurate demand matching based on the user's browsing data, thereby increasing the conversion rate in customer acquisition marketing. Finally, we determine corresponding advertising strategies based on the user profiles to achieve targeted recommendations of different ads based on user needs and preferences, improving the precision of marketing and customer acquisition.

[0045] In another embodiment, a process for building a user profile when it is determined that the visiting user exists in the potential customer database is provided. This includes: pre-classifying user demand directions based on common real estate needs, such as school district needs, transportation needs, and commercial needs; then extracting user demand tags for each user demand direction based on the corresponding information of the visiting user in the potential customer database and the user's basic information. The user demand tag extraction process is implemented based on existing AI large-scale models; then analyzing user browsing data to obtain the weight coefficient of each user demand direction; and determining the demand percentage of each user demand direction based on the user demand tags and weight coefficients. Therefore, the demand percentage of each user demand direction can reflect the degree of user demand for the corresponding user demand direction. The higher the demand percentage, the stronger the corresponding demand. Thus, forming a user profile based on the demand percentage of all user demands can improve the accuracy of advertising in subsequent advertising promotion processes.

[0046] The process of obtaining the demand percentage for each user demand direction includes: using AI to compare the similarity of any two sets of user demand tags for each user demand direction and obtaining the similarity value; this process is implemented based on existing large AI models, and the obtained similarity range is [0, 1). When the two sets of user demand tags are relatively similar, the similarity value approaches 1, and when the two sets of user demand tags are completely opposite, the similarity value approaches 0. Then, the similarity is obtained through equations. Calculate the demand factor g for each type of user needs. i Then through the equation Calculate the demand percentage p for each user demand direction. i Where e is the natural constant, and λ i Let n be the weight coefficient for the i-th type of user demand direction. i Let r be the number of user demand tags for the i-th type of user demand direction. j The similarity between any two groups of user demand tags in each user demand direction, 0≤r j <1, γ is the adjustment coefficient, γ>1, the adjustment coefficient γ is set according to the test data fitting, Z is the number of user demand directions, therefore, through the demand factor g of each type of user demand. i The size of the tag can be adjusted as the number of user demand tags in a certain category of user demand increases (n). i The larger the value, and the higher the overall consistency of user demand tags ( The smaller the value, the better the assessment of the level of demand, achieved through the equation. Calculate the demand percentage p for each user demand direction. i This allows for the creation of user profiles based on the proportion of each user's needs, thereby improving the accuracy of advertising in subsequent promotional processes.

[0047] In addition, the process of obtaining the weight coefficients for each type of user demand includes: splitting user browsing data to obtain the split browsing pages and corresponding browsing durations; and using equations... Calculate the weight coefficient λ for the i-th type of user demand direction. i ;wherein, λt i The basic weights for the i-th type of user demand direction are pre-set based on the degree of user demand for different user demand directions in empirical data, Q is the number of pages to be browsed, k∈[1,Q], d ik Let t be the preset association value between the k-th split browsing page and the i-th type of user demand direction. Since the number of browsing pages and user demand directions is finite, the preset association value can be set in advance based on the degree of association between the content of the browsing page and the user demand direction. k Let be the browsing time of the k-th segmented browsing page. Therefore, by analyzing the correlation between the segmented browsing page and the user's demand direction, as well as the time the user spends on the segmented browsing page, we can determine the user's demand level in different user demand directions. This is achieved through the weight coefficient λ of the i-th type of user demand direction. i This enables the realization of demand factor g. i The calculation.

[0048] In one embodiment, when it is determined that the accessing user does not exist in the potential customer database, the process of performing a coincidence comparison includes: using an equation... Calculate the deviation V between the accessing user and the x-th preset user profile. x Where B is the amount of basic user information, y∈[1,B], r xy Let β be the similarity between the basic information of the y-th user and the basic information of the user corresponding to the x-th preset user profile. y Let y be the weight coefficient for the y-th user's basic information, which is set according to the category of the user's basic information. τ is a correction coefficient, which is determined based on the test data. The numerical range is adaptively set, Z is the number of user demand directions, i∈[1,Z], λ ix The base value for the i-th user demand direction in the x-th preset user profile is set according to the characteristics when the preset user profile is created, α. i Let V be the weight value for the i-th user demand direction, which is pre-set according to the importance of the i-th user demand direction. Therefore, the deviation value V x The smaller the value, the higher the similarity between the user's basic information and the x-th preset user profile, and the closer the overall state of the user's needs is to the x-th preset user profile. Therefore, the minimum deviation value V is selected. xThe corresponding preset user profile serves as the visitor profile, enabling more accurate demand matching based on the visitor's browsing data, thereby increasing the conversion rate during customer acquisition and marketing.

[0049] In one embodiment, a process for determining a corresponding advertising strategy based on the user profile of the visitor is also proposed, including: pre-setting different advertising strategies according to different user demand directions; determining the final advertising strategy based on the demand proportion of each type of user demand direction in the user profile of the visitor, and using an equation. Calculate the selection coefficient u for the i-th user demand direction i Where Z-1 is the number of user demand directions other than the i-th user demand direction, s∈[1,Z-1], p s ω represents the proportion of demand for the s-th user demand direction. is Let be the correlation between the s-th user demand direction and the i-th user demand direction. Since different user demand directions have a certain correlation, a corresponding correlation degree is pre-set for different user demand directions. The correlation degree range is between [0, 1). Then, by selecting the coefficient u... i The size determines the corresponding advertising strategy, based on the maximum selection coefficient u. i The corresponding advertising strategy, as the final advertising strategy, can achieve the effect of precise advertising promotion, that is, improve the accuracy of marketing and customer acquisition.

[0050] This application also discloses a multi-source data fusion real estate precision customer acquisition system, referring to... Figure 2 The system includes: an information collection unit for collecting access user information, including basic user information and user browsing data; a user analysis unit for performing a preliminary comparison between the access user and a potential customer database based on the basic user information; when it is determined that the access user exists in the potential customer database, establishing an access user profile based on the corresponding information of the access user in the potential customer database, the basic user information, and the user browsing data; when it is determined that the access user does not exist in the potential customer database, performing an overlap comparison between the basic user information and the user browsing data and a preset user profile, selecting the preset user profile with the highest overlap as the access user profile; and a promotion unit for determining the corresponding advertising strategy based on the access user profile.

[0051] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for precise customer acquisition in the real estate market through multi-source data fusion, characterized in that, include: Collect access user information, which includes basic user information and user browsing data; Based on the user's basic information, the visiting user is initially compared with the potential customer database. When it is determined that the visiting user exists in the potential customer database, a user profile is built based on the corresponding information of the visiting user in the potential customer database, the user's basic information, and the user's browsing data. When it is determined that the visiting user does not exist in the potential customer database, the user's basic information and user browsing data are compared with the preset user profile, and the preset user profile with the highest degree of overlap is selected as the visiting user profile. Determine the corresponding advertising strategy based on the user profile of the visitor.

2. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 1, characterized in that, When it is determined that a visiting user exists in the potential customer database, the process of creating a visiting user profile includes: Pre-define user needs; Based on the corresponding information of the visiting users in the potential customer database and the user's basic information, extract user demand tags for each type of user demand direction; Analyze user browsing data to obtain the weight coefficients for each type of user demand. The demand percentage for each user demand direction is determined based on the user demand tags and weight coefficients for each user demand direction. A user profile is created based on the proportion of all user needs.

3. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 2, characterized in that, The process of obtaining the demand percentage for each user demand category includes: Based on AI, the similarity of any two sets of user demand tags in each type of user demand direction is compared to obtain the similarity value; Through equations Calculate the demand factor g for each type of user needs. i ; through equations Calculate the demand percentage p for each user demand direction. i ; Where e is the natural constant, λ i Let n be the weight coefficient for the i-th type of user demand direction. i Let r be the number of user demand tags for the i-th type of user demand direction. j The similarity between any two groups of user demand tags in each user demand direction, 0≤r j <1, where γ is the adjustment coefficient; γ>1, where Z is the number of user demand directions.

4. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 3, characterized in that, The process of obtaining the weight coefficients for each type of user demand includes: The user browsing data is broken down to obtain the segmented browsing pages and their corresponding browsing durations; Through equations Calculate the weight coefficient λ for the i-th type of user demand direction. i ; Where, λt i Let d be the basic weight for the i-th type of user demand direction, Q be the number of pages to be browsed, k∈[1,Q], and d be the weight for the i-th type of user demand direction. ik t is the preset association value between the k-th split browsing page and the i-th type of user demand direction. k This represents the browsing time for the k-th split page.

5. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 4, characterized in that, When it is determined that the visiting user does not exist in the potential customer database, the process of performing a coincidence comparison includes: Through equations Calculate the deviation V between the accessing user and the x-th preset user profile. x Select the minimum deviation value V x The corresponding preset user profile is used as the visiting user profile; Where B is the amount of basic user information, y∈[1,B], r xy Let β be the similarity between the basic information of the y-th user and the basic information of the user corresponding to the x-th preset user profile. y Let λ be the weight coefficient of the basic information of the y-th user, τ be the correction coefficient, Z be the number of user demand directions, i∈[1,Z], and λ be the weight coefficient of the basic information of the y-th user. ix Let α be the base value for the i-th user demand direction in the x-th preset user profile. i Let be the weight value for the i-th user's demand direction.

6. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 5, characterized in that, The process of determining the corresponding advertising strategy based on the user profile includes: Pre-set different advertising strategies based on different user needs; The final advertising strategy is determined based on the proportion of demand for each user category in the user profile.

7. The method for precise customer acquisition in real estate through multi-source data fusion according to claim 6, characterized in that, The process of determining the final advertising strategy based on user profiles also includes: Through equations Calculate the selection coefficient u for the i-th user demand direction i According to the maximum selection coefficient u i The corresponding advertising strategy serves as the final advertising strategy; Where Z-1 represents the number of user demand directions other than the i-th user demand direction, s∈[1, Z-1], p s ω represents the proportion of demand for the s-th user demand direction. is Let be the correlation between the s-th user demand direction and the i-th user demand direction.

8. A real estate precision customer acquisition system that integrates multi-source data, characterized in that, The system employs a multi-source data fusion method for precise customer acquisition in the real estate market, as described in any one of claims 1-7, comprising: An information collection unit is used to collect access user information, which includes basic user information and user browsing data. The user analysis unit is used to make a preliminary comparison between the visiting user and the potential customer database based on the user's basic information. When it is determined that the visiting user exists in the potential customer database, a user profile is built based on the corresponding information of the visiting user in the potential customer database, the user's basic information, and the user's browsing data. When it is determined that the visiting user does not exist in the potential customer database, the user's basic information and the user's browsing data are compared with the preset user profile, and the preset user profile with the highest degree of overlap is selected as the visiting user profile. The promotion unit is used to determine the corresponding advertising strategy based on the user profile of the visitor.

Images