Business district merchant analysis method

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a business district merchant analysis method which can adjust merchant information in a business district in real time according to consumption records of users so as to improve the accuracy and real-time performance of business district merchant division.

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a method for analyzing merchants in a business district. Background Art

[0002] A business district refers to a gathering place consisting of a large number of merchants and consumers in an area. If you can accurately find the merchants in the surrounding business district when the merchant marketing activities are carried out, and provide consumers with marketing activity information of merchants in the same business district, you can greatly increase the participation rate of merchants in marketing activities.

[0003] The registered address of a business license usually includes the street, shopping mall, etc. Due to the possibility of errors in the text of the registered address and the inconsistency between the registered address of the business license and the actual business address, it is necessary to divide the business according to the registered address of the business license, and dispatch business personnel to confirm the business address, business status, marketing activities, etc. of the business on the spot. Even so, it is impossible to update the business information of the business district in time when the business status, business address, etc. of the business change, resulting in a delay in the division of businesses in the same business district, affecting the accuracy of marketing activity push. Summary of the invention

[0004] The purpose of the present invention is to provide a business district merchant analysis method, which can effectively improve the accuracy of business district merchant division.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is:

[0006] A method for analyzing merchants in a business district includes step S130, determining whether two merchants belong to the same business district based on transaction frequency and transaction time difference, and specifically includes the following steps:

[0007] S131, extracting from the database the time difference between the preset time and the preset time 1 The consumption records of the same user in two merchants are recorded as merchant a and merchant b. The consumption records include the transaction account of merchant a, the transaction account of merchant b, the previous transaction time t a , at the post-trading time t b ;

[0008] S132: Count the consumption records with the same two merchant accounts in the consumption records extracted in step S131, and obtain the consumption records with a preset time difference of -T. 1 The total number of consumption records of a single user with merchant a and merchant b respectively is n;

[0009] S133. Calculate the time weight W of merchant a and merchant b e ,

[0010]

[0011] or

[0012]

[0013] Among them, α and β are constant preset coefficients;

[0014] S134: Determine the time weight W calculated in step S133 e Is it within the preset time weight threshold range? If W e Within the preset threshold range, it is determined that merchants a and b belong to the same business district. If W e If it is not within the preset threshold range, it is determined that merchant a and merchant b belong to different business districts.

[0015] Compared with the prior art, the present invention has the following technical effects: it can adjust the merchant information in the business district in real time according to the user's consumption record, thereby improving the accuracy and real-time performance of the business district merchant division. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The following is a brief description of the contents and symbols in the drawings of this specification:

[0017] Figure 1 It is a schematic diagram of the time-frequency topology graph;

[0018] Figure 2 It is a schematic diagram of the spatial topology map. DETAILED DESCRIPTION

[0019] The specific implementation of the present invention is further described in detail below through description of embodiments in conjunction with the accompanying drawings.

[0020] Embodiment 1

[0021] After integrating the merchant information, the merchant information belonging to the same commercial gathering place is included in the same set, and then a business district set can be constructed. In order to monitor and update the information of the business district in real time, the following business district merchant analysis method can be used:

[0022] S130: Determine whether the two merchants belong to the same business district based on the transaction frequency and the transaction time difference.

[0023] The transaction frequency refers to the number of transactions between different users who consume at two merchants. The more the number of transactions, the greater the transaction frequency between the two merchants, which means that more users consume at the two merchants one after another, and the greater the probability that the two merchants are located in the same business district.

[0024] The transaction time difference is the time difference between the same user making purchases at two merchants. The smaller the time difference, the faster the user can make purchases at the two merchants. If the distance between the two merchants is small, then the probability that the two merchants are located in the same business district is high.

[0025] Therefore, the transaction frequency and transaction time difference can be combined to determine whether two merchants belong to the same business district.

[0026] This embodiment uses the following steps to make a determination:

[0027] S131. Extract consumption records from the database of the financial service provider.

[0028] The consumption record is a time difference of 1T 1 The consumption records of the same user at two merchants.

[0029] The two merchants are called merchant a and merchant b. The consumption records include the transaction account of merchant a, the transaction account of merchant b, and the previous transaction time t a , at the post-trading time t b ;

[0030] S132: Count the consumption records with the same two merchant accounts in the consumption records extracted in step S131, and obtain the consumption records with a preset time difference of -T. 1 The total number of consumption records n of a single user conducting transactions with merchant a and merchant b respectively.

[0031] When counting consumption records, as long as the merchant accounts of two transactions are consistent, they can be included without determining the order in which the user conducts transactions at the two merchants.

[0032] S133. Calculate the time weight W of merchant a and merchant b e ,

[0033]

[0034] or

[0035]

[0036] Among them, α and β are preset coefficients, which are constants.

[0037] It can be known from the above calculation formula that if the transaction time difference is small, the time weight value is large, and the correlation between the two merchants is strong; if the transaction time difference is large, the time weight value is small, and the correlation between the two merchants is weak. However, if the correlation between the two merchants is judged only based on the time difference between the two transactions, there may be a correlation judgment error caused by one of the transactions being an online transaction. Therefore, it is necessary to increase the transaction frequency to judge the strength of the correlation between the two merchants, and introduce the preset coefficients α and β to balance the impact of the transaction frequency and the transaction time difference on the time weight value.

[0038] S134: Determine the time weight W calculated in step S133 e Is it within the preset time weight threshold range? e Within the preset threshold range, it is determined that merchants a and b belong to the same business district; if W e If it is not within the preset threshold range, it is determined that merchant a and merchant b belong to different business districts.

[0039] If the time weight value is too small, it is possible that at least one of the two transactions is a long-distance online transaction rather than a physical offline transaction. If the time weight value is too large, it may be because the two merchants are far apart and do not belong to the same business district, so it is necessary to set a time weight threshold range to eliminate abnormal situations.

[0040] Embodiment 2

[0041] The difference between this embodiment and the first embodiment is that, in order to improve the accuracy of determining merchants in the business district, after completing step S130, step S140 needs to be performed to exclude merchants that interfere with the business at a long distance. Exclude the occurrence of misjudgment of merchants in the business district caused by the fact that two merchants are far apart offline, but short-time difference transactions often occur between the two merchants, and at least one of the transactions is an online transaction.

[0042] This embodiment specifically uses the following steps to eliminate abnormal situations:

[0043] S141, obtaining the location coordinate information of the two merchant transaction accounts determined to be located in the same business district in step S134, including the longitude coordinates and latitude coordinates of the two merchants.

[0044] S142, calculating the distance between the two position coordinate information obtained in step S141.

[0045] The spacing calculation formula is as follows:

[0046] dis (a，b)=6371*acos(cos(radians(b.LAT))*cos(radians(a.LAT)))*cos(radians(a.LNG)-radians(b.LNG))+sin(radians(b.LAT))*sin(radians(a.LAT)))

[0047] Among them, dis (a，b) is the distance between merchant a and merchant b, dis (a，b) The smaller the value, the closer the distance between the two merchants; a.LAT is the latitude coordinate of merchant a, b.LAT is the latitude coordinate of merchant b, a.LNG is the longitude coordinate of merchant a, and b.LNG is the longitude coordinate of merchant b.

[0048] S143, the distance dis calculated in step S142 (a，b) Compare with the set distance threshold D. (a，b) <D, that is, the distance between merchant a and merchant b is less than the set distance threshold D, which can be used to determine that merchant a and merchant b belong to the same business district; if dis (a，b) >D, that is, the distance between merchant a and merchant b is greater than the set distance threshold D, based on which it can be determined that merchant a and merchant b do not belong to the same business district.

[0049] Embodiment 3

[0050] The difference between this embodiment and the first and second embodiments is that the business district set 10 is first constructed based on the transaction data, and then the business district merchant analysis method disclosed in the first or second embodiment is applied to update the business district set 10.

[0051] Specifically, the following steps need to be completed before step S130:

[0052] S120, constructing a business district set 10 based on the transaction time sequence.

[0053] S121, extracting the same user account from the database at a time difference of 2T 2 The consumption records with the number of consumption times greater than k include the transaction time and merchant transaction account of each consumption. This consumption record indicates that the user has made multiple transactions within a certain period of time, that is, the user has short-term high-frequency transactions. Based on this, it can be judged that the user may have made consumption activities in the business district. Therefore, the user's consumption trajectory can be analyzed based on this type of consumption behavior, and the association relationship between the merchants involved can be obtained through integration.

[0054] S122: Arrange the merchant transaction accounts according to the order of transaction time for the consumption records extracted in step S121 to obtain the association track of the merchant transaction accounts corresponding to the merchants. The approximate spatial position relationship of the merchants involved can be obtained from the association track.

[0055] S123, selecting a seed merchant, screening all consumption records involving the seed merchant, repeating steps S121 and S122, obtaining association tracks of all consumption records involving the seed merchant, and integrating the obtained association tracks to obtain an association network of all merchants involved;

[0056] Alternatively, steps S121 and S122 are repeated to obtain the association tracks of consumption records of all user accounts, and the obtained association tracks are integrated to obtain the association relationship network of all merchants involved.

[0057] S124, including all the merchant information in the associated relationship network obtained in step S123 into the same information set to construct a business district set 10.

[0058] Then, step S130 is entered to determine whether the merchants in the business district set 10 obtained in step S124 belong to the same business district, that is, the consumption records selected in step S131 are consumption records of the same user in two merchants, and the two merchants are both merchants in the business district set 10.

[0059] A time-frequency topology graph can be drawn based on the consumption records of the same user account in two merchants. The specific drawing method is to first draw two nodes 11 to represent merchants, and then build an edge 12 between the nodes 11 representing the two merchants. Figure 1 Shown is a time-frequency topology diagram obtained based on multiple consumption records.

[0060] In the time-frequency topology graph, the length of the edge 12 connecting the two nodes 11 is the time weight W calculated in step S133. e , or the length of the edge 12 connecting the two nodes 11 and the time weight W calculated in step S133 e Positive correlation.

[0061] Furthermore, based on a set of consumption records extracted in step S121, a single-chain time-frequency topology map can be obtained to intuitively express the association trajectory of the merchants; based on the association relationship network obtained in step S123, a mesh-like time-frequency topology map can be obtained to express the association trajectory of merchants in the business district.

[0062] Embodiment 4

[0063] The difference between this embodiment and the first and second embodiments is that the business district set 10 is first constructed by applying the coordinate information of the merchants, and then the business district merchant analysis method disclosed in the first or second embodiment is applied to update the business district set 10.

[0064] Specifically, the following steps need to be completed before step S130:

[0065] S110, constructing a business district set 10 based on the business location coordinates.

[0066] S111, obtain the location coordinate information of the merchant from the database, select the seed merchant, where the seed merchant is recorded as merchant c. The seed merchant mentioned here can be a merchant selected manually, or a merchant whose coordinates are close to the center of the business district, or a merchant with a high user consumption frequency obtained by screening based on consumption records. In specific implementation, multiple seed merchants can be selected to construct the business district set 10.

[0067] S112, calculating the distance between other merchants and the seed merchant, and the other merchants are recorded as merchant d.

[0068] The spacing calculation formula is as follows:

[0069] dis (c，d) =6371*acos(cos(radians(d.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(d.LNG))+sin(radians(d.LAT))*sin(radians(c.LAT)))

[0070] Among them, dis (c，d) is the distance between merchant c and merchant d, dis (c，d) The smaller the value, the closer the distance between the two merchants; c.LAT is the latitude coordinate of merchant c, d.LAT is the latitude coordinate of merchant d, c.LNG is the longitude coordinate of merchant c, and d.LNG is the longitude coordinate of merchant d.

[0071] S113, setting a distance threshold D, and selecting merchants whose location coordinate distance from the seed merchant is less than or equal to D to form a business district set 10.

[0072] Then, the process proceeds to step S130, wherein at least one merchant in the consumption records extracted in step S131 is a merchant in the business district set 10 obtained in step S113. That is, in step S131, merchant a or merchant b is a merchant in the business district set 10 obtained in step S113. In this way, step S130 can add or remove merchants from the business district set 10 according to the user consumption records, thereby achieving the purpose of improving the accuracy of the business district set 10 and timely updating the merchant information in the business district set 10.

[0073] A spatial topology map can be drawn based on the merchant location coordinates. The specific drawing method is to first draw a central node 21 to represent the seed merchant, then draw a spacing ring 22 with the central node 21 as the center and the spacing threshold D as the radius, and when the surrounding merchants that meet the requirements of the spacing with the seed merchant are screened, mark the surrounding nodes 23 on the spacing ring 22 to represent the surrounding merchants, and the spatial topology map can be obtained.

[0074] In specific implementation, a preferred solution can be selected, that is, in step S113, multiple spacing thresholds D are set, and D = (D1, D2, ... Dn). When constructing the business district set 10, merchants whose location coordinates are less than or equal to Dn from the seed merchant are included in the same set to form a subset of the business district set 10. The final business district set 10 has at least n subsets. The spatial topology map drawn based on the above preferred solution is shown in the attached figure. Figure 2 As shown, after drawing the central node 21, n spacing rings 22 are drawn with the central node 21 as the center and the spacing threshold value D = (D1, D2, ... Dn) as the radius. When the surrounding merchants that meet the requirements of the spacing with the seed merchant are screened, the surrounding nodes 23 are marked on the corresponding spacing rings 22 to refer to the surrounding merchants, and the spatial topology map can be obtained.

[0075] When the same user consumes at two merchants included in the business circle set 10, an edge can be constructed between the two merchant nodes in the spatial topology graph to indicate that there is a temporal correlation between the two based on the consumption records.

[0076] Embodiment 5

[0077] The difference between this embodiment and the first and second embodiments is that a business district set 10 is first constructed based on the coordinate information of the merchants, and then a business district supplement set 20 is constructed based on the transaction data. After the business district merchant analysis method disclosed in the first or second embodiment is applied to eliminate abnormal merchants in the business district supplement set 20, the merchants included in the business district supplement set 20 are added to the business district set 10 to obtain an updated and supplemented business district total set 30, thereby achieving the purpose of real-time updating of the information of the business district set 10 based on consumption records.

[0078] The complete business district merchant analysis method includes the following steps:

[0079] S110, constructing a business district set 10 based on the business location coordinates.

[0080] S111, obtain the location coordinate information of the merchant from the database, select a seed merchant, and record the seed merchant as merchant c. The seed merchant mentioned here can be a merchant selected manually, or a merchant whose coordinates are close to the center of the business district, or a merchant with a high user consumption frequency obtained by screening based on consumption records. Multiple seed merchants can be selected to construct the business district set 10.

[0081] S112, calculate the distance between other merchants and the seed merchant, and the other merchants are recorded as merchant d. The distance calculation formula is as follows:

[0082] dis (c，d)=6371*acos(cos(radians(d.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(d.LNG))+sin(radians(d.LAT))*sin(radians(c.LAT)))

[0083] Among them, dis (c，d) is the distance between merchant c and merchant d, dis (c，d) The smaller the value, the closer the distance between the two merchants; c.LAT is the latitude coordinate of merchant c, d.LAT is the latitude coordinate of merchant d, c.LNG is the longitude coordinate of merchant c, and d.LNG is the longitude coordinate of merchant d.

[0084] S113, setting a distance threshold D, screening merchants whose location coordinate distance from the seed merchant is less than or equal to D to form a business district set 10 constructed based on the merchant coordinates.

[0085] After the business district set 10 is constructed, it can be used to provide users with marketing information of the same business district. In order to improve the merchants in the business district, the merchants in the business district can be updated based on the user's historical consumption records or new consumption records while the marketing system is operating and applying. The specific steps are as follows.

[0086] S120, expanding the business district set 10 based on the transaction time sequence to obtain the merchant space location, that is, constructing the business district supplement set 20 based on the transaction time sequence to obtain the merchant space location.

[0087] S121, extracting the same user account from the database at a time difference of 2T 2 The consumption records with the number of consumption times greater than k include the transaction time and merchant transaction account of each transaction. It should be noted that, in the extracted consumption records, the merchant transaction account of at least one transaction is the transaction account of the merchant in the business district set 10, so as to ensure that the added merchant has a certain relevance to the business district set 10.

[0088] S122. Arrange the merchant transaction accounts according to the order of transaction time for the consumption records extracted in step S121, and record the transaction time difference between two adjacent merchant transaction accounts to obtain the association track between the merchant transaction accounts and the merchants.

[0089] S123, repeat steps S121 and S122 to obtain the association tracks of all selected consumption records, and integrate the obtained association tracks to obtain the association network of all merchants involved. The selected consumption records mentioned here are all consumption records that meet the following two conditions at the same time: first, the same user account has a time difference of 2T 2The number of in-circle consumption is greater than k; secondly, the merchant transaction account of at least one transaction is the transaction account of the merchant in the business circle set 10.

[0090] S124, all merchants in the associated relationship network obtained in step S123 are included in the same information set to construct a business district supplementary set 20.

[0091] In order to further improve the accuracy of merchant judgment in the business district supplementary set 20, it is necessary to remove abnormal merchants in the set, and the following steps are specifically adopted.

[0092] S130: Eliminate interference items in the business district supplementary set 20 based on transaction frequency and transaction time difference.

[0093] S131, extracting from the database the time difference between the preset time and the preset time 1 The consumption records of the same user in two merchants. One of the merchants is a merchant in the business district set 10, recorded as merchant a, and the other merchant is a merchant in the business district supplementary set 20, recorded as merchant b. In this way, the position relationship between the merchants in the business district supplementary set 20 and the merchants in the business district set 10 can be determined based on the location coordinates of the merchants. In the preferred solution, merchant a is a seed merchant.

[0094] Consumption records include merchant a's transaction account, merchant b's transaction account, and the previous transaction time t a , at the post-trading time t b .

[0095] S132: Count the consumption records with the same transaction account number of two merchants in the consumption records extracted in step S131, and obtain the number of consumption records with a preset time difference of -T. 1 The total number of consumption records n of a single user conducting transactions with merchant a and merchant b respectively.

[0096] S133. Calculate the time weight W of merchant a and merchant b e ,

[0097]

[0098] or

[0099]

[0100] Among them, α and β are constant preset coefficients.

[0101] S134: Determine the time weight W calculated in step S133 e Is it within the preset time weight threshold range? If W e Within the preset threshold range, it is determined that merchants a and b belong to the same business district, and merchant b is retained in the business district supplementary set 20. If W eIf the value is not within the preset threshold range, it is determined that merchant a and merchant b belong to different business districts, and merchant b is removed from the business district supplementary set 20 .

[0102] S140, excluding long-distance interference items in the business district supplement set 20.

[0103] S141, obtaining location coordinate information of each merchant in the business district supplementary set 20.

[0104] S142, calculating the distance between each merchant in the business district supplement set 20 and the seed merchant in sequence, the seed merchant is recorded as merchant c, and the merchant in the business district supplement set 20 is recorded as merchant e.

[0105] The spacing calculation formula is as follows:

[0106] dis (c，e) =6371*acos(cos(radians(e.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(e.LNG))+sin(radians(e.LAT))*sin(radians(c.LAT)))

[0107] Among them, dis (c，e) is the distance between merchant c and merchant e, dis (c，e) The smaller the value, the closer the distance between the two merchants; c.LAT is the latitude coordinate of merchant c, e.LAT is the latitude coordinate of merchant e, c.LNG is the longitude coordinate of merchant c, and e.LNG is the longitude coordinate of merchant e.

[0108] S143, the distance dis calculated in step S142 (s，e) Compare with the set distance threshold D. If the distance (c，e) <D, it is determined that the two merchants belong to the same business district, and merchant b is retained in the business district supplement set 20. (c，e) >D, it is determined that the two merchants belong to different business districts, and merchant b is removed from the business district supplement set 20.

[0109] S150, the business district supplementary set 20 and the business district set 10 are combined to obtain the business district total set 30, and the information update of the business district information is completed.

[0110] This embodiment comprehensively considers the spatial distance of merchants and information such as consumption frequency and consumption time difference, so that the update of business district information is more comprehensive and the business district division result is more accurate and reliable.

Claims

1. A method for analyzing merchants in a business district, characterized by: The method comprises step S130, determining whether two merchants belong to the same business district based on the transaction frequency and the transaction time difference, and specifically comprises the following steps: S131, extracting from the database the consumption records of the same user at two merchants within a preset time difference of -T1, recording the two merchants as merchant a and merchant b, and the consumption records including the transaction account of merchant a, the transaction account of merchant b, the previous transaction time t1, a , at the post-trading time t b ; S132, counting the consumption records of the two merchant accounts that are consistent in the consumption records extracted in step S131, and obtaining the total number n of consumption records of a single user conducting transactions with merchant a and merchant b respectively within a preset time difference of -T1; S133. Calculate the time weight W of merchant a and merchant b e , or Among them, α and β are constant preset coefficients; S134: Determine the time weight W calculated in step S133 e Is it within the preset time weight threshold range? If W e Within the preset threshold range, it is determined that merchants a and b belong to the same business district. If W e If it is not within the preset threshold range, it is determined that merchant a and merchant b belong to different business districts.

2. The business district merchant analysis method according to claim 1, characterized in that: The step S140 is also included to eliminate the long-distance interference merchants, which specifically includes the following steps: S141, obtaining location coordinate information of two businesses determined to be located in the same business district in step S134; S142, calculating the distance between the two position coordinate information obtained in step S141, the distance calculation formula is as follows: dis (a，b) =6371*acos(cos(radians(b.LAT))*cos(radians(a.LAT)))*cos(radians(a.LNG)-radians(b.LNG))+sin(radians(b.LAT))))*sin(radians(a.LAT))) Among them, dis (a，b) is the distance between merchant a and merchant b, dis (a，b) The smaller it is, the closer the distance between the two merchants is; a.LAT is the latitude coordinate of merchant a, b.LAT is the latitude coordinate of merchant b, a.LNG is the longitude coordinate of merchant a, b.LNG is the longitude coordinate of merchant b; S143, the distance dis calculated in step S142 (a，b) Compare with the set distance threshold D. If the distance (a，b) <D, determine that merchants a and b belong to the same business district. If dis (a，b) >D, determine that merchants a and b belong to different business districts.

3. The business district merchant analysis method according to claim 1 or 2, characterized in that: Before step S130, there is step S120, which constructs a business district set (10) based on the transaction sequence, which specifically includes the following steps: S121, extracting from the database the consumption records of the same user account with a consumption number greater than k within the time difference T2, the consumption records including the transaction time and merchant account of each consumption; S122, arranging the merchant accounts according to the order of transaction time for the consumption records extracted in step S121, and recording the transaction time difference between two adjacent merchant accounts, to obtain the association track of the merchant accounts corresponding to the merchants; S123, selecting a seed merchant, screening all consumption records involving the seed merchant, repeating steps S121 and S122, obtaining association tracks of all consumption records involving the seed merchant, and integrating the obtained association tracks to obtain an association network of all merchants involved; Or repeat steps S121 and S122 to obtain the association tracks of consumption records of all user accounts, and integrate the obtained association tracks to obtain the association network of all merchants involved; S124, adding all merchants in the associated relationship network obtained in step S123 into the same business district set (10); Then, the process proceeds to step S130 to determine whether the merchants in the business district set (10) obtained in step S124 belong to the same business district.

4. The business district merchant analysis method according to claim 3 is characterized by: Draw a time-frequency topology graph based on the consumption records of the same user account in two merchants. Draw two nodes (11) to represent merchants, and construct an edge (12) between the nodes (11) representing the two merchants. The length of the edge (12) connecting the two nodes (11) is the time weight W calculated in step S133. e , or the length of the edge (12) connecting the two nodes (11) and the time weight W calculated in step S133 e Positive correlation.

5. The business district merchant analysis method according to claim 1 or 2, characterized in that: Before step S130, there is step S110, which constructs a business district set (10) based on the business location coordinates, which specifically includes the following steps: S111, obtaining the location coordinate information of merchants from the database, selecting a seed merchant, and recording the seed merchant as merchant c; S112, calculate the distance between other merchants and the seed merchant, other merchants are recorded as merchant d, and the distance calculation formula is as follows: dis (c，d) =6371*acos(cos(radians(d.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(d.LNG))+sin(radians(d.LAT))*sin(radians(c.LAT))) Among them, dis (c，d) is the distance between merchant c and merchant d. The smaller the value, the closer the distance between the two merchants. c.LAT is the latitude coordinate of merchant c, d.LAT is the latitude coordinate of merchant d, c.LNG is the longitude coordinate of merchant c, and d.LNG is the longitude coordinate of merchant d. S113, setting a distance threshold D, and selecting merchants whose location coordinate distance from the seed merchant is less than or equal to D to form a business district set (10); Then, the process proceeds to step S130 , wherein at least one merchant in the consumption records extracted in step S131 is a merchant in the business district set ( 10 ) obtained in step S113 .

6. The method for analyzing business district merchants according to claim 5, characterized in that: In step S113, multiple distance thresholds D are set, namely D = (D1, D2, ... Dn), and merchants whose location coordinate distance from the seed merchant is less than or equal to Dn constitute a subset of the business district set (10), and the business district set (10) has at least n subsets.

7. The business district merchant analysis method according to claim 6, characterized in that: Draw a spatial topology map based on the merchant location coordinates, A central node (21) is drawn to represent the seed merchant, and a number of spacing rings (22) are drawn with the central node (21) as the center and a spacing threshold D = (D1, D2, ... Dn) as the radius. When peripheral merchants that meet the spacing requirements with the seed merchant are screened, peripheral nodes (23) are marked on the corresponding spacing rings (22) to represent the peripheral merchants, and a spatial topology diagram is obtained.

8. A method for analyzing merchants in a business district, characterized by: S110, constructing a business district set based on the business location coordinates (10); S111, obtaining the location coordinate information of merchants from the database, selecting a seed merchant, and recording the seed merchant as merchant c; S112, calculate the distance between other merchants and the seed merchant, other merchants are recorded as merchant d, and the distance calculation formula is as follows: dis (c，d) =6371*acos(cos(radians(d.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(d.LNG))+sin(radians(d.LAT))*sin(radians(c.LAT))) Among them, dis (c，d) is the distance between merchant c and merchant d. The smaller the value, the closer the distance between the two merchants. c.LAT is the latitude coordinate of merchant c, d.LAT is the latitude coordinate of merchant d, c.LNG is the longitude coordinate of merchant c, and d.LNG is the longitude coordinate of merchant d. S113, setting a distance threshold D, and selecting merchants whose location coordinate distance from the seed merchant is less than or equal to D to form a business district set (10); S120, obtaining the merchant space location and expanding the business district set based on the transaction time sequence (10); S121, extracting from the database consumption records of the same user account with a consumption number greater than k within the time difference T2, the consumption records including the transaction time and merchant account of each transaction, and the merchant account of at least one transaction in the consumption records is the transaction account of the merchant in the business circle set (10); S122, arranging the merchant accounts according to the order of transaction time for the consumption records extracted in step S121, and recording the transaction time difference between two adjacent merchant accounts, to obtain the association track of the merchant accounts corresponding to the merchants; S123, repeating steps S121 and S122 to obtain the association tracks of consumption records of all user accounts, and integrating the obtained association tracks to obtain the association network of all merchants involved; S124, adding all merchants in the associated relationship network obtained in step S123 into the business district supplementary set (20); S130, eliminating interference items in the business district supplementary set (20) based on transaction frequency and transaction time difference; S131, extracting from the database the consumption records of the same user at two merchants within a preset time difference of -T1, wherein one merchant is a merchant in the business district set (10), denoted as merchant a, and the other merchant is a merchant in the business district supplementary set (20), denoted as merchant b, and the consumption records include the transaction account of merchant a, the transaction account of merchant b, the prior transaction time t a , at the post-trading time t b ; S132, counting the consumption records of the two merchant accounts that are consistent in the consumption records extracted in step S131, and obtaining the total number n of consumption records of a single user conducting transactions with merchant a and merchant b respectively within a preset time difference of -T1; S133. Calculate the time weight W of merchant a and merchant b e , or Among them, α and β are constant preset coefficients; S134: Determine the time weight W calculated in step S133 e Is it within the preset time weight threshold range? If W e Within the preset threshold range, it is determined that merchants a and b belong to the same business district, and merchant b is retained in the business district supplement set (20). If W e If it is not within the preset threshold range, it is determined that merchant a and merchant b belong to different business districts, and merchant b is removed from the business district supplement set (20); S140, exclude long-distance interference items in the business district supplement set (20); S141, obtaining the location coordinate information of each merchant in the business district supplementary set (20); S142, sequentially calculating the distance between each merchant in the supplementary set (20) of the business circle and the seed merchant, the seed merchant is recorded as merchant c, and the merchant in the supplementary set (20) of the business circle is recorded as merchant e. The spacing calculation formula is as follows: dis (c，e) =6371*acos(cos(radians(e.LAT))*cos(radians(c.LAT)))*cos(radians(c.LNG)-radians(e.LNG))+sin(radians(e.LAT))*sin(radians(c.LAT))) Among them, dis (c，e) is the distance between merchant c and merchant e, dis (c，e) The smaller the value, the closer the distance between the two merchants; c.LAT is the latitude coordinate of merchant c, e.LAT is the latitude coordinate of merchant e, c.LNG is the longitude coordinate of merchant c, and e.LNG is the longitude coordinate of merchant e; S143, the distance dis calculated in step S142 (c，e) Compare with the set distance threshold D. If the distance (c，e) <D, it is determined that the two merchants belong to the same business district, and merchant b is retained in the business district supplement set (20). (c，e) >D, determine that the two merchants belong to different business districts, and remove merchant b from the business district supplement set (20). S150, the business district supplementary set (20) and the business district set (10) are combined to obtain the business district total set (30).

9. The business district merchant analysis method according to claim 8, characterized in that: In step S113, a plurality of distance thresholds D are set, namely D=(D1, D2, ..., Dn), and merchants whose distance from the location coordinates of the seed merchant is less than or equal to Dn constitute a subset of the business district set (10), and the business district set (10) has at least n subsets; In step S143, the distance threshold D takes the maximum value D max .

10. The business district merchant analysis method according to claim 8, characterized in that: In step S131, among the consumption records extracted from the database, one of the merchants is a seed merchant in the business district set (10), recorded as merchant a, and the other merchant is a merchant in the business district supplementary set (20), recorded as merchant b.

Images