Exhibition service method and system based on potential buyer analysis model

Abstract

The present application relates to an exhibition service method and system based on a potential buyer analysis model. A first sample set is established by retrieving the buyer data of local buyers over the years, so that the first sample set is used as a training sample based on the potential buyer analysis model. It is worth noting that the data of all buyers in the buyer area in a continuous time period in the first sample set is screened to form a second sample set, and the second sample set is used to establish a potential buyer analysis model to be trained. The present application uses the overall fluctuation trend of the industry, agriculture or service industry in the location of buyers who are highly related to the purchase of goods at the exhibition as a reference, and adjusts all potential buyer analysis models to be trained by the influence parameter method, ensuring that the transaction scale data calculated by the trained potential buyer analysis model can match the buyer's actual purchase demand. The present application solves the defects of lack of accuracy in the collation and application of data based on the potential buyer analysis model.

Description

Technical Field

[0001] The present application relates to the field of digital exhibition technology, and in particular to an exhibition service method and system based on a potential buyer analysis model. Background Art

[0002] Based on the potential buyer analysis model, such as Google's LOOK-LIKE, it is currently a common technical means for sellers to obtain buyers' consumption tendencies and portray Internet users' consumption portraits. However, this type of traditional target feature extraction application model mainly uses the customer's own accumulated access data for model training. This training method is not suitable for the working conditions of short-term application software of exhibition buyers and sellers, and will cause inaccurate output results. Specifically, this type of traditional target feature extraction application model is difficult to analyze the capacity of the entire market and the general trend of target features using short-term macro data. This type of traditional target feature extraction application model can meet the needs of sophisticated users, but lacks accuracy in macro data collation and application. To this end, it is necessary to propose an exhibition service method and system based on a potential buyer analysis model in view of the defect that the traditional target feature extraction application model is difficult to cope with the working conditions of short-term application software, resulting in the lack of accuracy in macro data collation and application. Summary of the invention

[0003] Based on this, it is necessary to propose an exhibition service method and system based on the potential buyer analysis model, in view of the fact that the traditional target feature extraction application model is difficult to cope with the working conditions of short-term application software, resulting in the defects of macro data collation and lack of application accuracy.

[0004] This application provides an exhibition service method based on a potential buyer analysis model, including:

[0005] Retrieving the local buyer data of previous years to generate a first sample set;

[0006] Select a buyer region;

[0007] Using the buyer region as an index, filtering the data of all buyers in the buyer region within a continuous time period in the first sample set to form a second sample set;

[0008] Based on the second sample set, generating a potential buyer analysis model to be trained;

[0009] Returning to the step of selecting a buyer region until all buyer regions are selected, all potential buyer analysis models to be trained are adjusted using the influence parameter method;

[0010] Using the first sample set, training the adjusted potential buyer analysis model to be trained to obtain a potential buyer analysis model;

[0011] Receive buyer data, incorporate buyer data into the potential buyer analysis model, and generate data on the size of transactions to be made;

[0012] Arrange exhibition booths and deploy advertisements based on the data on the scale of transactions to be completed.

[0013] The present application also provides an exhibition service system based on a potential buyer analysis model, including:

[0014] A server, used to execute the exhibition service method based on the potential buyer analysis model mentioned above;

[0015] The memory is communicatively connected with the server.

[0016] The present application relates to an exhibition service method and system based on a potential buyer analysis model. By retrieving the buyer data of local buyers over the years, a first sample set is established so that the first sample set is used as a training sample based on the potential buyer analysis model. It is worth noting that the data of all buyers in the buyer area in a continuous time period in the first sample set is screened to form a second sample set, and the second sample set is used to establish a potential buyer analysis model to be trained. Since the data samples of exhibitions are characterized by short-term intensive macro data rather than continuous micro data, the present application uses the overall fluctuation trend of the industry, agriculture or service industry where the buyers are located, which is highly related to the purchase of goods at the exhibition, as a reference, and the influence parameter method is used to adjust all the potential buyer analysis models to be trained, ensuring that the scale data to be traded calculated based on the potential buyer analysis model after training can match the actual purchase demand of the buyer. Based on the scale data to be traded, the exhibition booth is carefully arranged and the advertising is deployed, thereby solving the problem that the potential buyer analysis model is difficult to cope with the working conditions of short-term application software, resulting in the lack of accuracy in macro data collation and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The illustrative embodiment drawings and their descriptions of this application are used to explain this application and do not constitute an improper limitation on this application.

[0018] Figure 1 A method flow chart of an exhibition service method based on a potential buyer analysis model provided in one embodiment of the present application.

[0019] Figure 2 A schematic diagram of the structure of an exhibition service system based on a potential buyer analysis model provided in one embodiment of the present application.

[0020] Reference numerals:

[0021] 100-server; 200-storage. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0023] The present application provides an exhibition service method based on a potential buyer analysis model.

[0024] like Figure 1 As shown, in one embodiment of the present application, an exhibition service method based on a potential buyer analysis model includes:

[0025] S100, retrieve local buyer data of previous years to generate a first sample set.

[0026] S200, select a buyer region.

[0027] S300, using the buyer area as an index, filtering the data of all buyers in the buyer area within a continuous time period in the first sample set to form a second sample set.

[0028] S400: Based on the second sample set, a potential buyer analysis model to be trained is generated.

[0029] S500, returning to the step of selecting a buyer region until all buyer regions are selected, and adjusting all potential buyer analysis models to be trained using an influence parameter method.

[0030] S600: Using the first sample set, the adjusted potential buyer analysis model to be trained is trained to obtain a potential buyer analysis model.

[0031] S700, receiving buyer data, incorporating the buyer data into a potential buyer analysis model, and generating data on the scale of transactions to be made.

[0032] S800, arranges exhibition booths and deploys advertisements based on the transaction scale data.

[0033] Specifically, by retrieving the buyer data of local buyers over the years, a first sample set is established, so that the first sample set is used as a training sample based on the potential buyer analysis model. It is worth noting that the data of all buyers in the buyer area in a continuous time period in the first sample set is screened to form a second sample set, and the potential buyer analysis model to be trained is established using the second sample set.

[0034] This embodiment relates to an exhibition service method and system based on a potential buyer analysis model. The characteristics of the data samples of the exhibition category are short-term intensive macro data, rather than continuous micro data. Therefore, this embodiment uses the overall fluctuation trend of the industry, agriculture or service industry in the location of the buyer who is highly related to the purchase of goods at the exhibition as a reference. The influence parameter method adjusts all the potential buyer analysis models to be trained, ensuring that the transaction scale data calculated based on the potential buyer analysis model after training can match the actual purchase demand of the buyer. Based on the transaction scale data, the exhibition booth arrangement and advertising deployment are carefully carried out, thereby solving the problem that the potential buyer analysis model is difficult to cope with the working conditions of short-term application software, resulting in the lack of accuracy in macro data collation and application.

[0035] In one embodiment of the present application, S400 includes:

[0036] S411, select a purchasing category.

[0037] S412, selecting the purchasing category, purchasing volume and visit volume of each buyer in a year of each buyer in the second sample set.

[0038] S413, determining the purchase volume and visit volume of the corresponding buyer based on the purchase category.

[0039] S414, determining the ratio of the purchase volume of the corresponding buyer of the purchase category to the purchase volume in the second sample set to form a first ratio value, and the ratio of the visit volume of the corresponding buyer of the purchase category to the visit volume in the second sample set to form a second ratio value.

[0040] S415, returning the purchasing category, purchasing volume and visit volume of each buyer in a year selected in the second sample set until the data of all years are selected.

[0041] S416, returning to the step of selecting a purchasing category until all purchasing categories are selected.

[0042] Specifically, in order to accurately determine the actual purchase volume that the buyer can accommodate, this embodiment uses the ratio of the purchase volume of the corresponding buyer of the purchase category to the purchase volume of the second sample set to form a first ratio value, and then determines whether the first ratio value matches other buyers of the purchase category in this area. There are three situations for this matching result: match, buyer over-purchase, and buyer low purchase volume.

[0043] More specifically, there may be buyers from multiple companies in a region. When the proportion of a buyer's purchase volume in the second sample set is too high or too low, the second proportion value can be used to determine whether the amount of advertising is reasonable.

[0044] At this time, the proportion of the visit volume of the corresponding buyer of the procurement category in the visit volume of the second sample set is used to form a second ratio value. Based on the second ratio value, the reasons for excessive purchase by the buyer and low purchase volume by the buyer are judged: when the second ratio value is too low, it may be that the amount of advertising is too low, resulting in too low information acquisition rate of the buyer for the exhibition. Generally speaking, the buyer will purchase excessive samples for testing and finally select mass-produced products, so the second ratio value can also reflect the rationality of advertising.

[0045] This embodiment relates to an application method of the visit volume and purchase volume of a buyer. The first ratio value is used to determine whether the purchase volume of the buyer matches the data of other buyers in the purchase category in this area, and the ratio of the visit volume of the corresponding buyer in the purchase category to the visit volume in the second sample set is further used to determine the rationality of advertising delivery. This is conducive to improving the rationality of advertising delivery and improving the efficiency of advertising delivery.

[0046] In one embodiment of the present application, S400 includes:

[0047] S421, determining the purchasing category of each buyer in the second sample set, the purchasing volume of each buyer, and the number of visits of each buyer.

[0048] S422, based on the purchasing category and purchasing volume of each buyer in the second sample set, determine the purchasing characteristics of the buyer's area.

[0049] S423, based on the purchasing characteristics of the buyer region, obtain the industry data of the buyer region over the years.

[0050] Specifically, a purchasing category actually corresponds to a specific application industry such as industry, agriculture or service industry in a buyer's area. The application industry data of this area can more accurately reflect the region's macro demand for the goods in this purchasing category.

[0051] More specifically, for example, if the purchasing category is agricultural drip irrigation pipes, and the buyer is located in a desert, then it can be determined that the purchase volume of agricultural drip irrigation pipes in that area is large.

[0052] It is worth mentioning that the introduction of procurement categories is to expand the granularity of advertising. Simply put, agricultural drip irrigation pipes can be in the field of agricultural equipment or in the field of construction.

[0053] This embodiment involves obtaining the industry data of the buyer region over the years. Based on the purchasing characteristics of the buyer region in the purchasing category, the source of the industry data of the buyer region over the years can be expanded, and the granularity of the advertisement can also be improved.

[0054] In one embodiment of the present application, S400 includes:

[0055] S431, using the industry data of the buyer region over the years and the data of the purchasing characteristics of the buyer region, determine whether the data fluctuation of the purchasing characteristics of the buyer region is consistent with the data fluctuation of the industry data of the buyer region over the years.

[0056] S432: If the data fluctuation of the purchasing characteristics of the buyer region is consistent with the fluctuation of the industry data of the buyer region over the years, a first potential buyer analysis model is generated based on the fluctuation trend of the purchasing characteristics of the buyer region.

[0057] S433: If the data fluctuation of the purchasing characteristics of the buyer region does not match the fluctuation of the industry data of the buyer region over the years, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer region.

[0058] Specifically, the most ideal matching state between the industry data of the buyer's region over the years and the data of the purchasing characteristics of the buyer's region is that the data fluctuations of the purchasing characteristics of the buyer's region are consistent with the data fluctuations of the industry data of the buyer's region over the years.

[0059] Simple example: in the agricultural technology industry in the buyer's area, listed companies will have annual financial reports, which will indicate the purchase volume of agricultural drip irrigation pipes, the proportion of suppliers, and the market share of the listed company in the local industry. This information can determine the supply ratio of exogenous suppliers in the purchase category of agricultural drip irrigation pipes. Based on the supply ratio and the market size of the entire agricultural drip irrigation pipe market, the data fluctuations of the procurement characteristics can be used to determine the approximate number of agricultural drip irrigation pipes to be purchased by the buyer at the exhibition. At this time, first based on the potential buyer analysis model, the scale data to be traded can be relatively accurately determined.

[0060] In extreme cases, the data fluctuation of the purchasing characteristics of the buyer area does not match the fluctuation of the industry data of the buyer area over the years. In this case, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer area.

[0061] This embodiment relates to a potential buyer analysis model. The first potential buyer analysis model and the second potential buyer analysis model both belong to a potential buyer analysis model. The potential buyer analysis model can generate transaction scale data based on buyer data.

[0062] In one embodiment of the present application, S433 includes:

[0063] S441, using the purchasing characteristic data of the buyer area as the vertical coordinate.

[0064] S442, the industry data of the buyer region over the years is used as the horizontal axis.

[0065] S443, establishing a first association model.

[0066] S444, select a year.

[0067] S445, adding the purchasing characteristic data of the buyer region and the industry data of the buyer region for one year to the first association model to obtain a first relationship point.

[0068] S446, returning to the step of selecting a year until all the years are selected to obtain a plurality of first relationship points.

[0069] S447: Fit the first relationship point into a relationship curve based on a linear regression model or a data discrete model.

[0070] S448, incorporating the relationship curve into a second potential buyer analysis model; the second potential buyer analysis model is a potential buyer analysis model to be trained.

[0071] Specifically, in fact, the first association model is the association model of the data of the purchasing characteristics of the buyer area and the industry data of the buyer area over the years. More specifically, the association model of the purchasing volume of the buyer whose data fluctuation of the purchasing characteristics of all buyer areas of the same purchasing category does not coincide with the fluctuation of the industry data of the buyer area over the years and the industry data of the buyer area in the same year is taken as a unit of one year. One first relationship point is one year. These first relationship points are distributed in the two-dimensional mathematical model to form the first association model.

[0072] These first relationship points are fitted into a relationship curve using a linear regression model or a data discrete model. This relationship curve can obtain the industry data of the buyer region in the buyer data for one year and the purchase volume of the buyer in the corresponding year.

[0073] This embodiment involves the establishment of a first association model. Since the first association model is an association model between the data of purchasing characteristics of the buyer's area and the industry data of the buyer's area over the years, the purchase volume of the buyer in the corresponding year can be obtained through the industry data of the buyer's area in the buyer data for one year. It is worth noting that the purchase volume of the buyer in this corresponding year is the initial data that has not been adjusted by the influencing parameter method. In order to ensure that the purchase volume of the buyer in the subsequent corresponding year output based on the potential buyer analysis model meets the expectation that the data fluctuations of the purchasing characteristics of the buyer's area are consistent with the fluctuations of the industry data of the buyer's area over the years. A second association model can also be introduced. Simply, by adjusting the advertising placement, the data fluctuations of the purchasing characteristics of the buyer's area are consistent with the fluctuations of the industry data of the buyer's area over the years.

[0074] In one embodiment of the present application, S433 further includes:

[0075] S451, select a buyer.

[0076] S452, calling the buyer's purchase volume and the buyer's visit volume.

[0077] S453, the buyer's visit volume is used as the vertical coordinate.

[0078] S454, the purchase volume of the buyer is used as the horizontal axis.

[0079] S455, establishing a second association model.

[0080] S456, select a year.

[0081] S457, adding the purchase volume of the buyer and the visit volume of the buyer to the second association model to obtain a second relationship point.

[0082] S458, returning to the step of selecting a year until all the years are selected to obtain a plurality of second relationship points.

[0083] S459, fitting the relationship points into a correlation curve based on a linear regression model or a data discrete model.

[0084] Specifically, in fact, the second association model is an association model between the buyer's visit volume and the buyer's purchase volume. More specifically, the association model between the buyer's visit volume and the buyer's purchase volume in the same year is formed when the data fluctuation of the purchase characteristics of all buyer regions of the same purchase category does not coincide with the fluctuation of the industry data of the buyer region over the years. One second relationship point is one year. These second relationship points are distributed in a two-dimensional mathematical model to form a second association model.

[0085] These second relationship points are fitted into a correlation curve using a linear regression model or a data discrete model. This correlation curve can reflect the relationship between the buyer's visit volume and the purchase volume when the data fluctuation of the purchase characteristics of the buyer's region does not coincide with the fluctuation of the industry data of the buyer's region over the years.

[0086] This embodiment involves the establishment of a second association model. The establishment of the second association model is conducive to finding the relationship between advertising investment and the buyer's purchase volume. It is worth mentioning that the number of visits and advertising investment are positively correlated. When it is necessary to adjust the buyer's purchase volume so that the data fluctuations of the purchase characteristics of the buyer's area are consistent with the industry data fluctuations of the buyer's area over the years, the advertising investment based on the potential buyer analysis model can be adjusted.

[0087] In one embodiment of the present application, S433 further includes:

[0088] S461, select an association curve.

[0089] S462: Determine whether the correlation curve is a positive correlation curve based on the distribution trend of the correlation curve.

[0090] S463, if the correlation curve is a positive correlation curve, then the relationship curve corresponding to the correlation curve is gained to achieve that the data fluctuation of the purchasing characteristics of the buyer region is consistent with the fluctuation of the industry data of the buyer region over the years.

[0091] S464, if the correlation curve is not a positive correlation curve, it is determined that the correlation curve is a negative correlation curve, and the relationship curve corresponding to the correlation curve is reduced to achieve that the data fluctuation of the purchasing characteristics of the buyer area is consistent with the industry data fluctuation of the buyer area over the years.

[0092] S465, returning to the step of selecting a correlation curve until all correlation curves are selected.

[0093] Specific:

[0094] F(X i , Y i ) is the first association model.

[0095] Among them, X i is the industry data of the buyer region in the i-th year, Y i is the data of the buyer’s purchasing characteristics in the i-th year, where i is the year.

[0096] G(Y i ,Z i ) is the second association model.

[0097] Among them, Z i is the buyer’s visits in the i-th year.

[0098] F 目标 (X i , Y i ) is the pre-adjusted purchase volume. It is worth mentioning that the pre-adjusted purchase volume can be derived by the first potential buyer analysis model based on the industry data of the buyer's region over the years.

[0099] F 推算 (X i , Y i ) is the estimated purchase volume of the first association model.

[0100] Then we have:

[0101]

[0102] It is the average of industry data of the target buyer region over the years.

[0103] To estimate the average industry data of the buyer's region over the years.

[0104] R=F 目标 (X i , Y i ) and F 推算 (X i , Y i ) correlation coefficient.

[0105] It can be seen that: Z i目标 =R·Z i

[0106] Z i The target is the revised target visit volume for the buyer in year i.

[0107] This embodiment relates to a method for obtaining a relationship curve corresponding to a gain or loss correlation curve. In fact, R is the coefficient of gain or loss. By adjusting R, the number of buyers' visits can be adjusted, and then the possibility of buyers purchasing the number of exhibits can be adjusted.

[0108] In one embodiment of the present application, S800 includes:

[0109] S811, receiving transaction scale data.

[0110] S812, select a purchasing category.

[0111] S813, based on the pending transaction scale data, determine the proportion of the transaction scale data corresponding to the procurement category to the total transaction scale data.

[0112] S814, allocating the number of booths according to the ratio of the transaction scale data corresponding to the procurement category to the total transaction scale data.

[0113] S815, returning to the step of selecting a purchasing category until all purchasing categories are selected.

[0114] This embodiment relates to a method for planning the number of booths. The number of booths is allocated based on the ratio of the pending transaction scale data to the total transaction scale data, which is conducive to improving the efficiency of booth planning.

[0115] In one embodiment of the present application, S800 further includes:

[0116] S821, select a purchasing category.

[0117] S822, receiving the gain value or loss value of the procurement category.

[0118] S823, adjusting the advertising density of the procurement category based on the gain value or loss value of the procurement category.

[0119] This embodiment relates to a method for adjusting the density of advertisement placement, which adjusts the density of advertisement placement of a procurement category based on a gain value or a loss value of the procurement category, thereby helping to improve the efficiency of advertisement placement.

[0120] The present application provides an exhibition service system based on a potential buyer analysis model.

[0121] like Figure 2 As shown, in one embodiment of the present application, an exhibition service system based on a potential buyer analysis model includes a server 100 and a storage 200 .

[0122] The server 100 is used to execute the exhibition service method based on the potential buyer analysis model mentioned in any one of the above embodiments.

[0123] The storage 200 is in communication connection with the server 100 .

[0124] This embodiment relates to an exhibition service system based on a potential buyer analysis model. The server 100 establishes a first sample set by retrieving the buyer data of the buyers in the local memory 200 over the years, so that the first sample set is used as a training sample based on the potential buyer analysis model. It is worth noting that the data of all buyers in the buyer area in a continuous time period in the first sample set are screened to form a second sample set, and the server 100 uses the second sample set to establish a potential buyer analysis model to be trained. Since the data samples of the exhibition category are characterized by short-term intensive macro data rather than continuous micro data, the application uses the overall fluctuation trend of the industry, agriculture or service industry where the buyers are located, which is highly related to the purchase of goods at the exhibition, as a reference, and the influence parameter method is used to adjust all the potential buyer analysis models to be trained, ensuring that the scale data to be traded calculated based on the potential buyer analysis model after training can match the actual purchase demand of the buyers. Based on the scale data to be traded, the exhibition booths are carefully arranged and the advertising is deployed, thereby solving the problem that the potential buyer analysis model is difficult to cope with the working conditions of short-term application software, resulting in the lack of accuracy in macro data collation and application.

[0125] The technical features of the above-described embodiments may be arbitrarily combined, and the execution order of the method steps is not limited. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An exhibition service method based on a potential buyer analysis model, characterized in that: include: Retrieving the local buyer data of previous years to generate a first sample set; Select a buyer region; Using the buyer region as an index, filtering the data of all buyers in the buyer region within a continuous time period in the first sample set to form a second sample set; Based on the second sample set, generating a potential buyer analysis model to be trained; Returning to the step of selecting a buyer region until all buyer regions are selected, all potential buyer analysis models to be trained are adjusted using the influence parameter method; Using the first sample set, training the adjusted potential buyer analysis model to be trained to obtain a potential buyer analysis model; Receive buyer data, incorporate buyer data into the potential buyer analysis model, and generate data on the size of transactions to be made; Arrange exhibition booths and deploy advertisements based on the transaction size data; The step of generating a potential buyer analysis model to be trained based on the second sample set includes: Select a purchasing category; Select the purchasing categories, purchasing volume and visit volume of each buyer in one year of the second sample set; Determine the purchase volume and visit volume of the corresponding buyer based on the purchase category; Determine the ratio of the purchase volume of the corresponding buyer of the purchase category to the purchase volume in the second sample set to form a first ratio value, and the ratio of the visit volume of the corresponding buyer of the purchase category to the visit volume in the second sample set to form a second ratio value; Return the purchasing category, purchasing volume and visit volume of each buyer in a year of the second sample set until the data of all years are selected; Return to selecting a purchasing category until all purchasing categories are selected.

2. The exhibition service method based on the potential buyer analysis model according to claim 1 is characterized in that: The generating a potential buyer analysis model to be trained based on the second sample set also includes: Determine the purchase category, purchase volume and visit volume of each buyer in the second sample set; Based on the purchasing categories and purchasing volumes of each buyer in the second sample set, determining the purchasing characteristics of the buyer region; Based on the purchasing characteristics of the buyer's region, obtain the industry data of the buyer's region over the years.

3. The exhibition service method based on the potential buyer analysis model according to claim 2 is characterized in that: The generating a potential buyer analysis model to be trained based on the second sample set also includes: Using the industry data of the buyer region over the years and the data on the purchasing characteristics of the buyer region, determine whether the data fluctuations of the purchasing characteristics of the buyer region are consistent with the fluctuations of the industry data of the buyer region over the years; If the data fluctuation of the purchasing characteristics of the buyer region is consistent with the fluctuation of the industry data of the buyer region over the years, a first potential buyer analysis model is generated based on the fluctuation trend of the purchasing characteristics of the buyer region; If the data fluctuation of the purchasing characteristics of the buyer area does not match the fluctuation of the industry data of the buyer area over the years, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer area.

4. The exhibition service method based on the potential buyer analysis model according to claim 3 is characterized in that: If the data fluctuation of the purchasing characteristics of the buyer region does not match the fluctuation of the industry data of the buyer region over the years, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer region, including: The data of the purchasing characteristics of the buyer area is used as the vertical axis; The industry data of the buyer region over the years is used as the horizontal axis; Establishing a first correlation model; Select a year; Add the purchasing characteristic data of the buyer region and the industry data of the buyer region for one year to the first association model to obtain a first relationship point; return to the step of selecting a year until all years are selected to obtain multiple first relationship points; Based on a linear regression model or a data discrete model, the first relationship point is fitted into a relationship curve; the relationship curve is incorporated into a second potential buyer analysis model; the second potential buyer analysis model is a potential buyer analysis model to be trained.

5. The exhibition service method based on the potential buyer analysis model according to claim 4 is characterized in that: If the data fluctuation of the purchasing characteristics of the buyer area does not match the fluctuation of the industry data of the buyer area over the years, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer area, and further includes: Select a buyer; Call the buyer's purchase volume and the buyer's visit volume; The buyer's visit volume is used as the vertical axis; The purchase volume of this buyer is used as the horizontal axis; Establishing a second association model; Select a year; Adding the purchase volume of the buyer and the visit volume of the buyer to the second association model to obtain a second relationship point; Returning to the step of selecting a year until all the years are selected, so as to obtain a plurality of second relationship points; Based on the linear regression model or data discrete model, the relationship points are fitted into an association curve.

6. The exhibition service method based on the potential buyer analysis model according to claim 5 is characterized in that: If the data fluctuation of the purchasing characteristics of the buyer area does not match the fluctuation of the industry data of the buyer area over the years, a second potential buyer analysis model to be trained is generated based on the fluctuation trend of the purchasing characteristics of the buyer area, and further includes: Select a correlation curve; Based on the distribution trend of the correlation curve, determining whether the correlation curve is a positive correlation curve; If the correlation curve is a positive correlation curve, the relationship curve corresponding to the correlation curve is added to achieve that the data fluctuation of the purchasing characteristics of the buyer region is consistent with the fluctuation of the industry data of the buyer region over the years; If the correlation curve is not a positive correlation curve, it is determined that the correlation curve is a negative correlation curve, and the relationship curve corresponding to the correlation curve is reduced to achieve that the data fluctuation of the purchasing characteristics of the buyer area is consistent with the fluctuation of the industry data of the buyer area over the years; Return to the step of selecting an associated curve until all associated curves are selected.

7. The exhibition service method based on the potential buyer analysis model according to claim 6 is characterized in that: Arrange exhibition booths and deploy advertisements based on the transaction scale data, including: Receive data on the size of transactions to be made; Select a purchasing category; Based on the pending transaction scale data, determine the proportion of the transaction scale data corresponding to the procurement category to the total transaction scale data; The number of booths will be allocated based on the proportion of the transaction scale data corresponding to the procurement category to the total transaction scale data; Return to selecting a purchasing category until all purchasing categories are selected.

8. The exhibition service method based on the potential buyer analysis model according to claim 7 is characterized in that: Arrange exhibition booths and deploy advertisements based on the transaction scale data, and also include: Select a purchasing category; Receive the gain or loss value of the procurement category; Based on the gain value or loss value of the procurement category, adjust the advertising density of the procurement category.

9. An exhibition service system based on a potential buyer analysis model, characterized in that: include: A server, used to execute the exhibition service method based on the potential buyer analysis model as described in any one of claims 1 to 8; The memory is communicatively connected with the server.

Images