Merchant asset liquidation method and system based on data analysis

Abstract

The invention relates to the technical field of asset management, in particular to a merchant asset liquidation method and system based on data analysis, and the method comprises the following steps: obtaining a geographic association degree based on asset list coordinates, a traffic network, commercial division and population census data; and establishing a multi-dimensional asset feature set by combining the geographic association degree with financial account parameters. According to the method, geographic correlation degree analysis is carried out based on asset list coordinates in combination with a traffic network, commercial division and population census data, data dimensions are expanded on the basis of traditional asset management, quantification of asset values by geographic positions is realized, and asset evaluation is enabled to better fit an actual commercial environment; and meanwhile, through quantitative association of factors such as traffic proximity in the asset feature set, the market influence of real estate in the region is reflected, and an asset space value factor is formed, so that the asset estimation can more accurately reflect the market supply and demand features and the region competition situation.

Description

Technical Field

[0001] The present invention relates to the technical field of asset management, and in particular to a merchant asset liquidation method and system based on data analysis. Background Art

[0002] Asset management is a comprehensive management activity that effectively plans, allocates, operates, monitors, and disposes of assets. It involves processes such as asset valuation, risk management, portfolio optimization, and asset liquidity maintenance. Merchant asset liquidation methods, on the other hand, involve the systematic evaluation, inventory, valuation, and allocation and disposal of various assets held by merchants when they cease operations or undergo significant business changes.

[0003] Existing asset management technologies typically rely on processes such as asset planning, allocation, operation, monitoring, and disposal, but fail to effectively incorporate quantitative analysis of geographic factors and regional economic environments. Asset valuations often rely on subjective judgments or single financial indicators, failing to accurately reflect the actual value fluctuations of assets in specific market environments. When formulating asset turnover strategies, there is a general lack of in-depth analysis of regional market characteristics and buyer profiles. This leads to asset liquidation strategies being highly generalizable and difficult to meet the actual needs of specific regions, resulting in inefficient asset allocation and loss of market value. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a merchant asset liquidation method and system based on data analysis.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a merchant asset liquidation method based on data analysis, comprising the following steps:

[0006] Obtaining geographic relevance based on asset inventory coordinates and transportation network, commercial zoning, and census data; combining the geographic relevance with financial account parameters to establish a multi-dimensional asset feature set;

[0007] Based on the quantitative correlation of traffic proximity in the multi-dimensional asset feature set, the basic weight of real estate value in the region is calculated to obtain the spatial comprehensive impact results. Based on the spatial comprehensive impact results and the analysis of the impact of the distribution of competing and complementary business formats on asset management, the asset space value factor is established;

[0008] Calculate the asset base valuation based on the asset space value factor and asset list type, determine the intensity of demand for the target movable assets in different regions, and obtain a location-based liquidation premium valuation. Based on the location-based liquidation premium valuation and combined with balance sheet requirements and identification of potential buyer regions, generate a regionalized liquidation strategy basis;

[0009] Based on the income and expenditure forecasts based on the regionalized clearing strategy, the statistical mean and standard deviation of the financial accounts are set, the distribution range of transaction amount and frequency is obtained, the expected turnover fluctuation benchmark is calculated, and the clearing funds flow monitoring baseline is established.

[0010] Preferably, the steps for obtaining the geographical relevance are:

[0011] Extract the longitude and latitude information of each asset point in the asset list coordinates, call the transportation network node coordinates, the commercial district boundary point set and the census unit center coordinates, calculate the direction vector and distance value of each asset point to the nearest point in the three types of target points, and form the asset list spatial direction-distance basis set;

[0012] Based on the spatial direction-distance basis set of the asset list, calculate the spatial fusion response value of each asset point based on the three-dimensional vectors of transportation accessibility, commercial proximity and population coverage;

[0013] Based on the spatial fusion response values, all asset points with spatial fusion response values lower than the average are screened, and coordinate mappings with transportation network nodes, commercial district boundaries, and census centers are established to generate geographic relevance.

[0014] Preferably, the steps for obtaining the multi-dimensional asset feature set are:

[0015] Recalling the mapping results between the asset inventory coordinates recorded in the geographic association and the transportation network node coordinates, the commercial district boundary coordinates, and the census unit centroid coordinates; based on the mapped coordinate relationships, calculating the Euclidean distance between each asset point and the corresponding transportation network node; performing adjacency determination on points within the commercial district boundary envelope; and marking whether the census unit number overlaps with the coordinates of the asset point, thereby generating an asset inventory spatial basic association information group;

[0016] Based on the asset list spatial basic association information group, the operating income field, transaction frequency field and expenditure details field recorded in the financial account parameters are extracted, and the primary key correspondence verification is performed between the asset list number field and the asset list spatial basic association information group, and a one-to-one correspondence is established between the asset point coordinates, the geographic mapping results and the financial account parameters to generate a multi-dimensional asset feature set.

[0017] Preferably, the steps for obtaining the spatial comprehensive impact result are:

[0018] Extract the multi-dimensional asset feature set, perform geographical zoning attribution judgment on all asset locations, and generate a geographical zoning comprehensive basic parameter set;

[0019] Calculate the basic weight of the value of real estate within the region based on the comprehensive basic parameter set of the geographical division;

[0020] Based on the real estate value basic weight, the geographical division comprehensive basic parameter set is numbered and sorted and matched with the real estate value basic weight to generate a spatial comprehensive impact result.

[0021] Preferably, the steps for obtaining the asset space value factor are:

[0022] Based on the spatial comprehensive impact results, the asset coordinate points contained in each geographical zone are extracted, and the asset type identifiers of the corresponding competitive and complementary business formats are read. The total distribution and spatial coordinate positions of each type of asset point within the zone are statistically analyzed by type, and a spatial structure set of the zone asset type is generated;

[0023] According to the partitioned asset type spatial structure set, the distance value between each asset point and the nearest competing business format, the total number of complementary business formats and the number of heterogeneous non-complementary business formats in the current block are retrieved to generate a set of relative asset point spatial attributes;

[0024] Based on the asset point spatial relative attribute set, an asset space value factor is calculated.

[0025] Preferably, the steps for obtaining the location liquidation premium valuation are:

[0026] Based on the asset space value factor and the asset list type field, the asset space value factor and asset type code corresponding to each asset point are extracted, and the asset code and asset type code are grouped and integrated to confirm the point distribution and type distribution density of each type of asset within the corresponding geographical block, and generate an asset type block distribution set;

[0027] Extract the asset space value factor, asset type code, valuation sequence of the last twelve months and the maximum valuation value in the last natural month of each asset point from the asset type block distribution set to generate an asset type valuation attribute group;

[0028] Based on the asset type valuation attribute group, a location liquidation premium valuation is calculated.

[0029] Preferably, the steps for obtaining the regionalized liquidation strategy are:

[0030] Based on the location liquidation premium valuation, combined with the distribution of the asset list covered by each geographical block, the asset current book value field and the matured liability record field in the balance sheet are retrieved to establish a matching relationship between the valuation and financial items, and generate an asset valuation and book liability comparison group;

[0031] Based on the asset valuation and book liability comparison, the coordinates of asset points in each geographic block where the book value is lower than the premium valuation are extracted. At the same time, regional customer visit records, address distribution of completed buyers, and consultation frequency records are read. Each buyer information is categorized and organized according to the administrative division field to generate a set of high-potential buyer concentration area identifiers.

[0032] Based on the set of identifiers of high-potential buyer clusters, combined with the locations of assets in the geographic blocks where the book value is lower than the premium valuation, a cross-attribution matching is performed between the asset locations and the buyer regions, and a geographic block number and a recommended liquidation ranking result are generated according to the matching results to generate a basis for a regionalized liquidation strategy.

[0033] Preferably, the steps for obtaining the clearing funds flow monitoring baseline are:

[0034] Based on the regionalized liquidation strategy, the transaction amount and transaction frequency details of each asset point are extracted, the time field and asset number are aligned, and abnormal records are removed to generate a time series set of asset transaction behavior;

[0035] Calculate the expected turnover fluctuation benchmark for each asset point based on the asset transaction behavior time series set;

[0036] Based on the expected turnover fluctuation benchmark, all asset points are classified according to asset numbers within the geographical blocks to which they belong, and the transaction volume stability variation range and extreme value boundaries within each block are recorded to generate a clearing funds flow monitoring baseline.

[0037] The present invention provides a merchant asset liquidation system, comprising:

[0038] The geographic correlation analysis module integrates and compares geospatial data based on asset inventory coordinates, transportation networks, commercial districts, and census data. It calculates the correlation between each merchant's asset location and transportation networks, commercial districts, and population distribution, and generates geographic correlation results.

[0039] The asset feature set construction module extracts multi-dimensional features of merchant assets based on geographic correlation results and combines them with financial account parameters. It analyzes and summarizes factors such as transportation proximity, population density, and income level to generate an asset feature set.

[0040] The spatial value assessment module, based on the asset feature set, quantifies and correlates factors such as traffic proximity, calculates the basic weight of real estate within the region, and evaluates and obtains the comprehensive spatial impact results. Through the comprehensive spatial impact results, it analyzes the impact of regional competition and complementary business distribution on asset management, and generates asset spatial value factors;

[0041] The asset valuation and regional demand analysis module calculates the basic valuation of merchant assets based on the asset space value factor and asset list type, assesses the intensity of demand for target movable assets in different regions, and combines regional liquidation premium valuation to determine buyer demand in potential regions and generate regional liquidation demand valuation;

[0042] The financial settlement monitoring module, based on regional settlement demand valuation and balance sheet requirements, forecasts the income and expenditure of merchants' financial accounts, calculates the statistical mean and standard deviation of financial accounts, obtains the distribution range of transaction amount and frequency, calculates the expected turnover fluctuation benchmark, and establishes a settlement fund flow monitoring baseline.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are:

[0044] In the present invention, geographic correlation analysis is performed based on the coordinates of the asset list in combination with transportation networks, commercial zoning and census data, broadening the data dimension on the basis of traditional asset management, realizing the quantification of asset value by geographical location, and making asset evaluation more in line with the actual business environment; at the same time, through the quantitative correlation of factors such as transportation proximity in the asset feature set, the market influence of real estate in the region is reflected, and an asset space value factor is formed, so that asset valuation can more accurately reflect market supply and demand characteristics and regional competition situation, and improve asset pricing rationality and market adaptability; in addition, by combining asset-liability requirements with buyer potential area identification, a regionalized asset liquidation strategy basis is generated, which can be used to conduct personalized liquidation premium valuation according to different regional market needs, clarify the value flow direction of assets, reduce blindness and arbitrariness in the asset flow process, and enhance the maximization of asset disposal returns; in terms of funds monitoring, the distribution range of transaction amount and frequency is set through statistical analysis of financial accounts, and a liquidation funds flow monitoring baseline is established. The risk of capital fluctuations that may arise in the asset liquidation process is dynamically predicted and managed, making capital flow control more refined and improving the risk prevention and control capabilities in the asset liquidation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the steps of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] See also Figure 1 The present invention provides a technical solution, a merchant asset liquidation method based on data analysis, comprising the following steps:

[0048] Obtain geographic relevance based on asset inventory coordinates and transportation network, commercial zoning, and census data; combine geographic relevance with financial account parameters to establish a multi-dimensional asset feature set;

[0049] Based on the quantitative correlation of transportation proximity in the multi-dimensional asset characteristics, the basic weight of real estate value in the region is calculated to obtain the spatial comprehensive impact results. Based on the spatial comprehensive impact results and the analysis of the impact of the distribution of competing and complementary business formats on asset management, the asset space value factor is established;

[0050] Calculate the asset base valuation based on the asset space value factor and asset list type, determine the intensity of demand for the target movable assets in different regions, and obtain the location liquidation premium valuation. Based on the location liquidation premium valuation and combined with balance sheet requirements and identification of potential buyer areas, generate the basis for regional liquidation strategy;

[0051] Based on the income and expenditure forecast based on the regionalized clearing strategy, the statistical mean and standard deviation of the financial account are set, the distribution range of transaction amount and frequency is obtained, the expected turnover fluctuation benchmark is calculated, and the clearing funds flow monitoring baseline is established.

[0052] The steps to obtain geographic relevance are:

[0053] Extract the longitude and latitude information of each asset point in the asset list coordinates, call the transportation network node coordinates, the commercial district boundary point set and the census unit center coordinates, calculate the direction vector and distance value of each asset point to the nearest point in the three types of target points, and form the asset list spatial direction-distance basis set;

[0054] Based on the spatial direction-distance basis set of the asset list, the spatial fusion response value of each asset point is calculated based on the three-dimensional vectors of transportation accessibility, commercial proximity, and population coverage. The calculation formula is:

[0055]

[0056] Among them, Q j is the spatial fusion response value of the j-th asset point, dtn j is the straight-line distance from the asset point to the nearest transportation network node, can j is the tangent angle between the asset point and the nearest commercial area boundary point, μ j is the registered population in the census unit where the asset point is located, κ j is the permanent population of the street where the asset point is located, wpl j The length of the walking path from the asset point to the nearest commercial area center;

[0057] Based on the spatial fusion response value, all asset points with spatial fusion response values lower than the average are screened, and coordinate mappings with transportation network nodes, commercial district boundaries, and census centers are established to generate geographic correlation.

[0058] Specifically, the longitude and latitude information of each asset point in the asset list coordinates is extracted. First, the asset list number and the corresponding longitude and latitude data are read line by line in the internal coordinate library. Then, the transportation network node coordinate information stored in the same database is matched, and the route identification corresponding to each transportation network node and the coordinate position are sorted accordingly. For the commercial district boundary point set, the street range, the extended boundary of the business district, and the specific coordinate values of each boundary point are retrieved from the regional division data. At the same time, the center coordinates of the census unit are extracted and a number mapping is established between the census unit and its center point. Then, with the longitude and latitude of each asset point as the center, the spatial distance between it and the transportation network node coordinates, the commercial district boundary point coordinates, and the census unit center coordinates is calculated. During the distance comparison process, the positioning accuracy is tested, for example, verifying that the deviation between the longitude and latitude and the actual geographic location is within 0.0001 degrees, and if the difference exceeds 0. Positioning error correction is performed at 0.001 degrees. Next, a direction vector is constructed. The direction component is obtained using plane geometry methods, based on the difference coordinates between the asset point and the nearest transportation network node in the plane coordinate system. The same process is performed on the coordinates of commercial district boundary points and census unit centers. The resulting direction vector is then combined with the distance to form a direction-distance comparison record. For example, in an urban area, when the distance between an asset point and a transportation network node is 0.85 kilometers, its direction vector can be recorded as a set of values deviating to the southeast. The corresponding direction and distance are then calculated for the commercial district boundary and the census unit center. After the summary is completed, each item is checked item by item. If the direction component or distance component of the direction vector is empty, the asset point is manually inspected and the coordinates are recollected. After the direction vector and distance data of the three types of target points are extracted for all asset points, a unified spatial direction-distance basic set of the asset inventory is formed.

[0059] formula: The formula is beneficial in that it can couple traffic accessibility, commercial proximity and population coverage factors. and The combination of the above terms comprehensively reflects the distance and population difference intensity in the numerator, while the walking path length and radical terms are introduced in the denominator. The balanced regulation of the results by strengthening the geometric relationship between transportation and commercial areas can more accurately identify the degree of integration response under the spatial pattern in the overall system.

[0060] Parameter description (1): dtn jIt represents the straight-line distance from the asset point to the nearest transportation network node. Its value is obtained by measuring the difference between the longitude and latitude of the asset point and the longitude and latitude of the transportation network node. The distance calculation function of the geographic information system is used in the measurement process. For example, the longitude 116.412 and latitude 39.906 of the asset point are read from the city coordinate database, and the longitude 116.421 and latitude 39.904 of the transportation network node are read. The spherical distance formula is used. in and λ1,λ2 represent the values of the longitude and latitude of the asset point and the transportation network node converted into radians. If the result is 1.2 kilometers, let dtn j =1.2.

[0061] Parameter description (2): can j It is used to represent the tangent angle between the asset point and the nearest commercial district boundary point. This value is calculated by geometric methods. It is necessary to first perform dot product and modulus operations on the vector of the asset point relative to the commercial district boundary point and the reference base vector. in represents the vector from the commercial area boundary point to the asset point, The east direction can be defined as (1, 0) or other fixed directions according to the coordinates of the administrative divisions. can After conversion to numerical value, we can get can j If the angle is detected to be 45 degrees, it is recorded as approximately 0.7854 radians.

[0062] Parameter description (3): μ j It represents the number of registered people in the census unit where the asset point is located. The registered population number can be extracted from the local government's annual census data report, and detailed statistics are performed according to the division of streets and communities. Then, the specific data is located by matching the number of the census unit where the asset point is located. If the population count of the unit in the annual census report is 1500, then μ j =1500.

[0063] Parameter description (4): κ j Indicates the number of permanent residents in the street where the asset point is located. Its data comes from the street-level population management system and is consistent with μ j The difference is that the statistical scope here is larger, and it is necessary to summarize the population values of all census units under the same street and deduct duplicate registrations. The acquisition process can be queried through the demographic statistics platform of the street office. If the street name is retrieved and the total population is 1,300, then κ j =1300.

[0064] Parameter description (5): wpl jThe length of the walking path from the asset point to the nearest commercial center can be calculated by mapping the urban pedestrian road network into a weighted graph. First, the node and edge information of the pedestrian road must be generated in the geographic system. The weight of the connection between the nodes represents the length of the road segment. Then, the shortest path algorithm such as the Dijkstra algorithm is used to find the shortest path from the asset point to the commercial center. The length of all edges on the path is accumulated to get wpl. j If the calculated distance is 1.8 km, then wpl j =1.8. When determining this value, sections of highway without pedestrian access are excluded, and only sidewalks or accessible secondary roads are retained to ensure the objectivity and accuracy of the walking route. For example, when the path is measured at the asset point numbered AC102, the cumulative result is 1.8 kilometers, so wpl AC102 =1.8.

[0065] Calculation process: Step 1, calculation

[0066]

[0067] The second step is to calculate

[0068] ln(2.0641)≈0.7244;

[0069] Step 3: Calculate

[0070]

[0071] So the numerator is:

[0072] 0.7244+14.1421=14.8665;

[0073] Step 4: Calculate the denominator

[0074] dtn AC102 ·can AC102 =1.2×0.79=0.948;

[0075]

[0076] Step 5. Divide the numerator by the denominator and take the absolute value:

[0077]

[0078] The results show that for the asset point numbered AC102, the spatial fusion response value based on the coupling of multiple factors such as distance, angle, and demographics is approximately 4.65. When this value is too large, it means that it presents a high comprehensive response in terms of traffic proximity and commercial adjacency. If the same calculation is performed on other asset points in the future, the corresponding Q j The value sequence can be combined to further determine the distribution characteristics of spatial fusion and perform asset point screening or management actions.

[0079] According to the spatial fusion response value, first read the Q after all calculations are completed j The results are statistically analyzed and the Q corresponding to each asset point is calculated. j Compare with the overall average of all asset points and select Q j Asset points below this average are marked as a low-convergence response group. The coordinates of the asset points in the low-convergence response group are then aligned with the coordinates of the transportation network nodes to ensure that the resolved longitudinal and transverse positions satisfy the longitude-latitude correspondence in the same coordinate system. The coordinates of the commercial district boundaries are then mapped and verified. For example, if the coordinate range of a commercial district boundary overlaps the coordinates of an asset point, the asset point's tag information is further recorded and compared with the coordinates of the census centroid. If the distance between the asset point and the census centroid exceeds 2.0 kilometers, this value is recorded for subsequent reference. If the distance is within 2.0 kilometers, it is recorded separately and compared with the entries in the street resident information table. The entire comparison process requires thorough verification of coordinate alignment to ensure that the error range is within 0.0001 degrees. Any duplicate or anomaly in the coordinates is rechecked individually. Once correct, all marked asset points are aggregated according to street number, commercial district number, and census unit number. After completing this multi-mapping process, the correspondence between the low-convergence assets and various features is obtained, and the geographic correlation is generated.

[0080] The steps to obtain a multi-dimensional asset feature set are:

[0081] The mapping results between the asset inventory coordinates recorded in the geographic correlation and the transportation network node coordinates, commercial district boundary coordinates, and census unit centroid coordinates are called. Based on the mapped coordinate relationships, the Euclidean distance between each asset point and the corresponding transportation network node is calculated. The adjacency check of the points within the commercial district boundary envelope is performed, and the overlap between the census unit number and the coordinates of the asset point is marked to generate the asset inventory spatial basic correlation information group.

[0082] Based on the asset list spatial basic association information group, the operating income field, transaction frequency field and expenditure detail field recorded in the financial account parameters are extracted. The primary key correspondence is verified between the asset list number field and the asset list spatial basic association information group, and a one-to-one correspondence is established between the asset point coordinates, geographic mapping results and financial account parameters to generate a multi-dimensional asset feature set.

[0083] Specifically, based on the previously obtained geographic correlation mapping results, the coordinate values of each record in the asset list are read and matched with the sorted transportation network node coordinates, commercial district boundary coordinates and census unit centroid coordinates. After the matching is established, the Euclidean distance between each asset point and the corresponding transportation network node is calculated one by one, and the coordinates of the two points are recorded as (x1, y1) and (x2, y2) respectively. During the calculation process, if the calculated result exceeds 3.5 kilometers, a large distance mark will be added to the record. The 3.5-kilometer threshold is set by the average coverage of the traffic commuting area and the distribution of roads within the city. For example, the distance distribution between the central urban area and the surrounding main roads is statistically analyzed. After excluding extreme values, the median is rounded up and 3.5 kilometers is selected as the identification point based on the road density. Then, the envelope range adjacency judgment of the commercial district boundary is performed, and the commercial district boundary is processed according to polygonal coordinates. The asset point is based on whether it falls within the buffer zone of the adjacent boundary. The buffer range is between 0.2 kilometers and 0.5 kilometers. The asset points marked with adjacency relationships are compared with each other. If the x coordinate is between the minimum x value of the boundary minus 0.2 and the maximum x value of the boundary plus 0.2, and the y coordinate is between the minimum y value of the boundary minus 0.2 and the maximum y value of the boundary plus 0.2, then the asset point is considered to be adjacent to the business district. After the adjacency marking is completed, the coordinates of the asset point and the centroid of the census unit are checked for overlap. A point-by-point comparison is performed and the error range is limited to 0.0001 degrees. If the error difference is less than this limit, the overlap is considered to exist. If it exceeds this limit, the actual distance value is recorded and no overlap is processed. After completing these calculations and markings for all asset points, they are uniformly integrated to obtain the asset inventory spatial basic association information group.

[0084] Based on the obtained asset list spatial basic association information group, the corresponding business income field, transaction frequency field and expenditure detail field in the financial account parameters are retrieved one by one, and the unique number of each asset point is matched with the primary key of the financial account parameter. For the business income field, its monthly total value is compared with the coordinates and adjacency status information of the asset point. The transaction frequency field is counted on a daily basis and associated one by one with the previously established asset geographic mapping results. For example, if the cumulative transaction frequency value of an asset point exceeds 120 times in the same week, it will be marked, and the value of 120 times is determined by the general level of urban commercial activities and the field sampling survey. The expenditure details field is aggregated by reading related entries with the same asset number, and the single amount and expenditure category of each expenditure record are analyzed. If the single amount exceeds 5,000 yuan, an additional mark is added. The 5,000 yuan is confirmed based on the regional average expenditure range and financial data distribution. After the screening and matching is completed, all retrieved data are deduplicated and timestamp compared, and inconsistent entries are excluded. The associated operating income, transaction frequency, expenditure details and other information are combined with the asset list space basic association information to form a comprehensive structure containing asset point coordinates, geographic mapping relationships and financial fields, and finally a multi-dimensional asset feature set is generated.

[0085] The steps to obtain the spatial comprehensive impact results are:

[0086] Extract multi-dimensional asset feature sets, perform geographical zoning attribution judgment on all asset locations, and generate a comprehensive basic parameter set for geographical zoning;

[0087] According to the comprehensive basic parameter set of geographical divisions, the basic weight of real estate value in the region is calculated using the following formula:

[0088]

[0089] Among them, R i is the base weight of real estate value in the ith geographical zone, φ i is the total traffic proximity value in the area, η i is the average population density, τ i is the average income level, ψ i is the maximum income value of the region, υ i is the total value of asset circulation frequency in the region, ξ i is the average building density of the region;

[0090] Based on the basic weights of real estate value, the comprehensive basic parameter sets of geographical divisions are numbered and organized and matched with the basic weights of real estate value to generate the spatial comprehensive impact results.

[0091] Specifically, based on the existing multi-dimensional asset feature set, all asset points are read and the coordinates of their locations are classified. Then, according to the administrative division information of the geographical area, each asset point is compared with the overlapping or adjacent relationship with the boundaries of streets or communities, and the asset points are marked one by one with reference to the topographic map and regional boundary coordinate data. When multiple asset points appear in the same street area, the coordinates of these asset points are classified into the same category and recorded and summarized. After the marking is completed, the regional boundary information needs to be compared with the attribution of each asset point item by item to prevent the coordinates from falling outside the dividing line but being mistakenly classified into a certain geographical division. When such deviations are found, a refined coordinate matching strategy is further used, including comparing the upper and lower limits of the longitude and latitude of the coordinate data with the corresponding coordinates of the street boundary, and ensuring that the error value does not exceed 0.000. The density is sorted according to the street or community area data provided by the demographic department and the asset point data information determined previously. When encountering asset points that are too concentrated in a local area, these coordinates are extracted and checked again in the regional vector layer to see if they are within the same complete polygon range. Each record will generate a corresponding spatial index entry and establish a one-to-one or many-to-one correspondence with the street number. In this process, it is necessary to compare the node coordinates of the regional shape file one by one to ensure that the logical edges of all partitions are correct. After the verification is completed, the geographical ownership of all asset points will be correlated with their financial fields and population information in the previous feature set. Finally, the asset point coordinates and their associated information contained in each geographic partition are stripped out from the data results for the entire region to form a comprehensive basic parameter set for the geographic partition.

[0092] formula: The formula is useful in that it i ,η i , τ i , ψ i 、υ i ,ξ i Six parameters are combined and operated, integrating key factors such as traffic proximity, asset circulation frequency and building density into the same expression. Logarithmic and absolute value operations are introduced in the numerator to highlight the impact of basic transportation and income factors. The coupling term of population density and income difference is added to the denominator to dynamically balance possible income gaps within the region. The coupling between building density and circulation frequency is reflected in the final structure in the form of cube and square, which can present the overall level of real estate value at the geographical zoning level.

[0093] The first parameter φ i The steps to obtain φ are: iIt represents the total traffic proximity value in the i-th geographical zone. In the early data collection, it is necessary to record the actual traffic time from each major transportation hub to all asset points in the zone, and then convert it into a comprehensive coefficient and add it up to get the total traffic proximity value of the zone. In the specific implementation, the traffic proximity value of a single asset point q k It is defined as the traffic speed ratio from a certain traffic node to the asset point. It can be determined by factors such as road speed limit, average traffic volume in the area, and travel distance. It is calculated by the following formula: where v base Indicates the normal average driving speed of the city (in kilometers per minute), d road Indicates the mileage (in kilometers) from the asset point to the traffic node. After the calculation is completed, the q k Sum within the partition and get φ i For example, in the geographical area numbered AC201, the q of its 20 asset points is counted. k The values are in the range of 0.25 to 0.65 and the cumulative value is 8.40, then φ i =8.40.

[0094] The second parameter η i The steps to obtain η are: i It represents the average population density, which can be calculated by dividing the area of the partition by the total number of permanent residents. When the partition area is recorded as A i (unit: square kilometers), the permanent population is recorded as P i , then In the geographical area numbered AC201, the area is 2.6 square kilometers and the permanent population is 5200 people, so η i =5200 / 2.6=2000.

[0095] The third parameter τ i The steps to obtain τ are: i It represents the average income level, which is determined by the monthly income details published by the statistical department. First, the median and average income of households in the sub-district are collected, and a weighted method is used to obtain a unified average income value. It can be calculated based on the specific summary of monthly wages, business income, subsidy income, etc. For example, in the AC201 sub-district, after extracting the monthly wages of households, the average income per capita is about 5,000 yuan. Then τ i =5000.

[0096] The fourth parameter ψ i The steps to obtain ψ are: i Indicates the maximum income value in the area. It is necessary to search the highest income record in the same statistical database and remove outliers. If the highest income is registered as 30,000 yuan, then ψ i =30000.

[0097] The fifth parameter υ i The steps to obtain υ are: i It represents the total value of asset circulation frequency in the area. By counting each registered business or transaction, each record contains the transaction timestamp and transaction amount. All transaction frequencies are accumulated within a statistical period, and repeated checks are performed on the situation where adjacent transactions are too dense. If 120 valid transactions are counted in the AC201 zone in a quarter, then υ i =120.

[0098] The sixth parameter ξ i The steps to obtain ξ are: i It represents the average building density of the region, which is obtained by stratifying the ratio of the building area to the total area of the subdivision. The area of a single building can be obtained by combining actual measurement with real estate registration data, dividing the sum of all building areas by the subdivision area and normalizing to get the average value. When the total building area of AC201 subdivision is 2.08 square kilometers, 2.08 is divided by 2.6 to get 0.8, so ξ i =0.8.

[0099] Calculation process: The first step is to calculate the logarithmic term and the denominator basis term:

[0100] ln(φ i +1)=ln(8.40+1)=ln(9.40)≈2.24;

[0101]

[0102] |τ i -ψ i |=|5000-30000|=25000;

[0103] The second step, the molecular part is:

[0104] ln(φ i +1)·υ i =2.24×120=268.80;

[0105] The denominator is:

[0106]

[0107] So the numerator is divided by the square of the denominator:

[0108]

[0109] Step 3: Calculate

[0110]

[0111] (0.00661) 3 ≈2.89×10 -7 ;

[0112] The fourth step is to take the square root of the sum of the two:

[0113]

[0114] The results show that for zone AC201, after integrating traffic proximity, asset circulation frequency and building density, its real estate value base weight is approximately 0.00997. When this value is close to 0, it means that the area is relatively loose in various indicators. When the value is much greater than 0.1, it means that the area is highly concentrated in the indicator combination.

[0115] After obtaining the basic weights of real estate value, each geographical region has clear comprehensive basic parameters and (R i ) values, sort these partition numbers in the order of street or community coding, record the corresponding relationship between the partition and its number one by one, and then (R i ) values are paired with existing traffic and population statistics of the sub-districts and summarized. When duplicate sub-district numbers are detected during the summary process, the geographical boundaries are reviewed and corrected. When the sub-district information is confirmed to be correct, it can be displayed in a unified index table (R i ) and its related parameters, and then by looking up the association between the previously recorded asset point coordinates and the partition number in the index table, assign the real estate value base weight corresponding to each asset point to the corresponding data entry. After completion, all partition data are browsed as a whole, and the (R i ) values are compared with data such as the average population density and income level differences, and each district is marked with a corresponding mark in sequence and summarized to form a comprehensive spatial impact result.

[0116] The steps to obtain the asset space value factor are:

[0117] Based on the spatial comprehensive impact results, the asset coordinate points contained in each geographical zone are extracted, and the asset type identifiers of the corresponding competitive and complementary business formats are read. The total distribution and spatial coordinate positions of each type of asset point within the zone are counted by type, and a spatial structure set of zone asset types is generated;

[0118] According to the spatial structure set of partitioned asset types, the distance value between each asset point and the nearest competing business format, the total number of complementary business formats and the number of heterogeneous non-complementary business formats in the current block are retrieved to generate the asset point spatial relative attribute set;

[0119] Based on the relative attribute set of the asset point space, the asset space value factor is calculated using the following formula:

[0120]

[0121] Among them, S k is the asset space value factor of the kth asset point, d k is the distance between the asset point and the nearest competitive business format, c k is the number of complementary business formats in the block where the asset point is located, a k is the number of heterogeneous non-complementary business formats in the block where the asset point is located, R k The base weight of the real estate value of the zone to which the asset point belongs.

[0122] Specifically, based on the partition number in the spatial comprehensive impact result and the recorded asset coordinate information under each partition, the coordinate points contained in each geographical partition are retrieved one by one and an initial coordinate list is compiled. During the retrieval process, the error range between the partition boundary and the coordinate point position is checked one by one. When the error value is less than 0.0001 degrees, it is regarded as a valid coordinate point under the same geographical partition. If a coordinate point falls in the overlapping area of the dividing line, the previously established boundary node comparison record is queried, and the coordinates of the overlapping situation are visually marked on the map vector layer to facilitate further confirmation of the actual ownership of the coordinate. Then, the asset type identification of the competing business format and the complementary business format is read, and each asset type is extracted from the commercial directory data in a unified format, where each record contains the mapping relationship between the asset type code and the corresponding coordinate. These records are jointly compared with the asset coordinate list under the aforementioned partition. When the coordinates match and the type code is legal, the coordinates are merged into the category of the corresponding asset type. For example, different asset types such as catering, retail, and entertainment are distinguished by their type codes. When counting the total distribution of each type of asset point in the partition, the coordinates of the same type are counted. If the number of coordinates of the same type of asset exceeds 30, the degree of concentration of this type of asset in this partition is marked. At the same time, the coordinate position details are recorded while collecting the total distribution. These data and type classification information are written into the result entry under the corresponding partition entry and split according to the type code. If it is found that the coordinate distribution is too concentrated or relatively scattered, the specific value is noted in the memo record. After completing all statistics, the partition asset type spatial structure set is generated in the order of partition numbers.

[0123] According to the spatial structure set of the partitioned asset type generated previously, first read the various asset points and their coordinates stored by partition number, then retrieve the distance value between each asset point and the nearest competing business one by one, and compare the competing business type code when checking this distance. Once the target asset type code is found to be the same as the current asset type, it will be regarded as a competing business and the distance calculation will be performed. The calculation method is based on the coordinates and is measured through the plane straight line distance formula or combined with the urban road mileage data. When the distance calculation result is greater than 1.5 kilometers, a long distance mark is added. The specific value of 1.5 kilometers is set with reference to the statistical range of urban walking and short-distance traffic. Then, the distance is calculated for the target asset type code. The statistics of complementary business types will first read all records of complementary asset types in the current block and calculate the total number. For example, in the retail zone, types marked as complementary relationships such as catering and leisure and entertainment will be summarized. When the number reaches 15, the value will be recorded as the total number of complementary business types. The same search will be performed for heterogeneous non-complementary business types, and then these numbers will be associated with the target asset point. If the number exceeds 10, it will be automatically identified as a relatively concentrated state of heterogeneous non-complementary business types. After all the search steps are completed, corresponding records will be generated according to the number of each asset point. The records will include the distance value, the total number of complementary business types, and the summary results of heterogeneous non-complementary business types, and finally a set of relative attribute spaces of the asset points will be generated.

[0124] formula: The benefit of the formula lies in that by combining the competitive distance term and the number of complementary business formats in the numerator and combining it with logarithmic operations, as well as doubly scaling the heterogeneous non-complementary business formats under the basic weight of real estate value in the denominator, it can comprehensively measure the impact of the competitive environment and complementary environment as well as the real estate value weight of the region on the asset value in a single indicator, which is conducive to more accurately portraying the potential value of each asset point under the interaction of different business formats.

[0125] Parameter d k The steps to obtain d are: k It represents the distance between the k-th asset point and the nearest competitive business format. It is necessary to search for similar asset points one by one in the spatial relative attribute set obtained above, record the coordinates of the two one by one, and calculate it by the actual road mileage or plane straight-line distance. For example, in the urban area, the distance between the asset point numbered AC305 and the same type of competitive business format point is 1.2 kilometers after measurement, then d k =1.2. This distance needs to be combined with the actual traffic planning map and the calibration accuracy of the latitude and longitude coordinates. If the error value exceeds 0.01 km, a re-comparison is carried out to obtain the final distance value. In an application example, when the final measured distance between AC305 and the competing business format is 1.2 km, d AC305 =1.2 is included in the calculation.

[0126] Parameter c k The steps to obtain are: ck Indicates the number of complementary business types in the block where the asset point is located. By summarizing the records of asset types coded as complementary types in the previously generated partitioned asset type spatial structure set, if there are 10 retail stores in the same block as complementary assets of the catering business, then let c k =10.

[0127] Parameter a k The steps to obtain are: a. k Indicates the number of heterogeneous non-complementary business types in the block where the asset point is located, which is also retrieved from the aforementioned partitioned asset type spatial structure set. At this time, only the business types marked as non-complementary relationships are filtered out and their number is counted. For example, in the block numbered AC305, 5 technology service points that are not complementary to the catering business are detected, then a k =5.

[0128] Parameter R k The steps to obtain R are: k Indicates the real estate value base weight of the partition to which the asset point belongs. Its value comes from the partition-level real estate value base weight result obtained previously. After the asset point is matched with the partition number, the real estate value weight corresponding to the number is directly read. For example, the block numbered AC305 is in partition MC401. When the real estate value base weight of partition MC401 is consulted, 0.012 is obtained. Then R AC305 =0.012.

[0129] Calculation process: The first step is to calculate the sum of the two terms in the numerator:

[0130]

[0131] ln(c k +1)=ln(10+1)=ln(11)≈2.3979;

[0132] So the numerator parts add up:

[0133] 0.674+2.3979=3.0719;

[0134] Square it:

[0135] (3.0719) 2 ≈9.4366;

[0136] The second step is,

[0137]

[0138] tanh(4.94)≈0.9999;

[0139] The third step is to subtract the numerator from the denominator:

[0140] S k =9.4366-0.9999=8.4367;

[0141] The results show that for the asset point numbered AC305, the asset space value factor calculated based on factors such as the distance to competing formats, the number of complementary formats, and the number of heterogeneous non-complementary formats combined with the basic weight of real estate value is approximately 8.4367. When the value is higher, it means that the asset point has obtained stronger spatial complementarity and competitive balance advantages in this block. If the value is lower than 2, it means that there are fewer complementary formats or the competitive environment is relatively severe. k By comparing the values, we can determine the potential value of different asset points within the block.

[0142] The steps to obtain the location liquidation premium valuation are:

[0143] Based on the asset spatial value factor and asset list type field, the asset spatial value factor and asset type code corresponding to each asset point are extracted. The asset code and asset type code are then grouped and integrated to confirm the point distribution and type distribution density of each type of asset within the corresponding geographical block, and generate an asset type block distribution set.

[0144] Based on the asset type block distribution set, extract the asset space value factor, asset type code, valuation sequence of the last twelve months in the valuation record, and the maximum valuation in the last natural month for each asset point to generate an asset type valuation attribute group;

[0145] Based on the asset type valuation attribute group, the location liquidation premium valuation is calculated using the following formula:

[0146]

[0147] Among them, U m is the location liquidation premium estimate for the mth geographic block, ζ m is the mean value factor of the asset space in the block, χ m is the number of assets of the same asset list type in the block, θ m is the median of the valuation records of the same asset type in the past twelve months, m is the maximum valuation of the same asset type in the most recent natural month, ρ m The cumulative number of occurrences of the circulation attribute tag of this asset type in this block.

[0148] Specifically, when reading based on the asset spatial value factor and the asset inventory type field, the database is first searched for the asset number and corresponding asset spatial value factor of all asset points. These are then matched against the asset type code table. Grouping and organization are completed by mapping the asset code field to the type code field one by one. During this process, it is necessary to ensure that each asset point has a consistent number during mapping and that the type information is not missing. If the type code is found to be empty, it is immediately marked and temporarily removed from the grouping process. The total number of occurrences of the same asset type within the geographic block is then summarized and the coordinate distribution is recorded. The clustering degree of the type is assessed by traversing the asset coordinates in the partition vector layer and counting the number of coordinates of the same type. When performing clustering analysis, areas with excessively concentrated distribution are further marked. For example, if a certain asset type has more than 30 coordinates distributed within a block, it is marked as densely distributed, and if it has less than 5, it is marked as sparsely distributed. The specific values of dense and sparse are set based on the common store clustering degree in actual commercial distribution research. After all statistical processes are completed, an asset type block distribution set is generated based on the type and block number.

[0149] Based on the completed asset type block distribution set, each asset point listed in the set is read and its corresponding asset space value factor, asset type code, and previously stored valuation records are retrieved. For the valuation records, the valuation data series for the past twelve months needs to be located and read out in chronological order. After reading, all values are medianized and confirmed to be within the daily valuation range of each asset type. The maximum valuation in the most recent natural month is also extracted to assess the recent price fluctuation. If the maximum value is significantly greater than the median of the previous twelve months, a price spike is added to the corresponding data record. During the statistical process, valuation records with obvious anomalies are re-checked. For example, if the median is usually between 3,000 and 5,000 yuan, and the maximum valuation read this time reaches 10,000 yuan, it is marked as an abnormal spike. The entire process summarizes the median and maximum values and their occurrence dates, along with relevant circulation information. The circulation information includes the number of transactions or rental-sale conversions, forming an asset type valuation attribute group indexed by the asset point number.

[0150] formula: The benefit of this formula is that it combines the overall status of the asset space value factor within the block with the number of assets of the same type, the difference between the median valuation and the recent highest value, and then balances it with the logarithmic operation of the number of occurrences of the circulation attribute label. This allows the calculation of the liquidation premium to reflect the quantitative distribution of assets within the block while taking into account valuation fluctuations and trading enthusiasm.

[0151] Parameter ζ m The steps to obtain ζ are: mTo represent the mean of the asset space value factor within the block, it is necessary to select points in the same geographical block and with the same asset type from all the previously counted asset points, read their asset space value factors one by one, and then take the arithmetic average. Specifically, the values of these points can be added up and then divided by the corresponding number of asset points. For example, in the geographical block numbered KU301, there are 9 such asset points, and their value factors are 3.2, 4.1, 5.5, etc., which add up to about 36.1. Dividing 36.1 by 9 gives the mean of 4.011, so ζ m =4.011.

[0152] Parameter χ m The steps to obtain χ are: m To represent the number of assets of the same asset inventory type in the block, you need to first determine the block identification number and retrieve the asset type code of the same type, and count all assets that match the type code and belong to this geographic block. For example, in the KU301 geographic block, if there are 15 asset points matching the target type code, then χ m =15.

[0153] Parameter θ m The steps to obtain θ are: m It represents the median of the valuation records of the same asset type in the past twelve months. From the previous asset type valuation attribute group, all valuation data of this type in the past twelve months are taken one by one and sorted from low to high according to the value. If the number of data is even, the average of the two middle values after sorting is taken; if it is odd, the middle value is taken. For example, when the valuation data for twelve months is [2800, 2900, 3100, 3100, 3200, 3300, 3300, 3450, 3500, 3700, 3800, 4200], the total number is 12, and the 6th and 7th items are 3300 respectively. Therefore, the median θ m =3300.

[0154] Parameter λ m The steps to obtain λ are: m Indicates the maximum valuation of the same asset type in the most recent natural month. After reading the valuation sequence for the past month, all valid recorded values are compared and the highest value is selected. For example, in the KU301 geographic block, there are five valuation data for this asset type in the most recent month, namely 3600, 4050, 3900, 4500, and 4350. The highest value is 4500, so λ m =4500.

[0155] Parameter ρ m The steps to obtain ρ are: mIndicates the cumulative number of times the circulation attribute tag of this asset type appears in this block. Through the analysis of transaction behavior or rental and sale records, all circulation tags related to this type of asset are counted. For example, in the KU301 block, the transaction timestamp, transaction frequency or transfer record information is searched one by one. If a total of 20 valid identifiers related to the circulation attribute tag are found, then ρ m =20.

[0156] Calculation process: The first step is to calculate (ζ m ·χ m ) 2 :

[0157] ζ m ·χ m =4.011×15=60.165;

[0158] (60.165) 2 ≈3619.82;

[0159] The second step is to calculate |θ m -λ m |:

[0160] |3300-4500|=1200;

[0161] Add the two and take the square root:

[0162]

[0163] The third step is to calculate the denominator ln(ρ m +2):

[0164] ρ m =20,ρ m +2=22;

[0165] ln(22)≈3.091;

[0166] Step 4: Divide the numerator by the denominator:

[0167]

[0168] The results show that in the KU301 geographical block, the location liquidation premium valuation of the same asset type is approximately 22.47. When the value is greater than 10, it often means that the mean and scale of the asset space value factor of the asset type in the block are relatively considerable and the recent valuation fluctuation is more significant. When the value is less than 5, it means that the number of assets in the block is small or the valuation level is not obvious enough. m Values can be compared side-by-side to identify higher or lower liquidation premium possibilities.

[0169] The steps for obtaining the basis for regionalized liquidation strategy are as follows:

[0170] Based on the location-based liquidation premium valuation and the distribution of the asset list covered by each geographical block, the current book value field of the asset and the matured liability record field in the balance sheet are retrieved to establish a matching relationship between the valuation and financial items, and generate a comparison group of asset valuation and book liabilities;

[0171] Based on the asset valuation and book liability comparison, the coordinates of assets in each geographic block where the book value is lower than the premium valuation are extracted. At the same time, regional customer visit records, the address distribution of completed buyers, and the frequency of consultations are read. Each buyer's information is categorized and organized according to the administrative division field to generate a set of identifiers for areas where high-potential buyers gather.

[0172] Based on the set of identifiers of high-potential buyer clusters, combined with the locations of assets in geographic blocks where the book value is lower than the premium valuation, a cross-attribution matching is performed between the asset locations and the buyer regions. The geographic block numbers and recommended liquidation ranking results are generated according to the matching results to form the basis for the regionalized liquidation strategy.

[0173] Specifically, based on the location liquidation premium valuation and the distribution of the asset list corresponding to each geographical block, read the asset current book value field and the due liability record field previously recorded in the balance sheet, separate the book value and liability time node information from the database and load them together with the location liquidation premium valuation value, compare each asset data under the geographical block one by one, and after identifying the asset code, perform the comparison process of the book value field and the aforementioned premium valuation field. If the book value field value is within the set comparison range, keep the normal mark, and conduct detailed verification for records that exceed the set range or cannot be parsed normally. The comparison range can refer to the statistics of peer data and combine the city The annual book value floating range provided by the municipal financial regulatory department is used to set the value. For example, the book value range for commercial real estate is between 2 million and 10 million yuan. After the comparison is completed, the premium valuation value is combined with the book value and the matured liability record value through the primary key field, and then each matching result is aggregated by geographical area using the block number. If there are multiple assets in the same block corresponding to the same matured liability date, their time fields are sorted and added to the output results during the aggregation process, and the matching relationship between assets and book liabilities is summarized. Finally, the aggregation results are used to distinguish by number and write into the data index for classified output, forming an asset valuation and book liability control group.

[0174] Based on the data entries in the asset valuation and book liability control group, identify the coordinates of asset points in each geographic block whose book value is lower than the premium valuation, mark these coordinates in the geographic layer and record their corresponding asset numbers, then read the regional customer group visit records and the address distribution information of the transaction buyers, parse the source location of each visit in the visit record one by one, extract the street or community code and compare it with the administrative division field. If the two match, mark this visit as consistent with the administrative division. The address distribution information of the transaction buyers is also processed in the same way. After summarizing the visit and buyer distribution records, check the consultation frequency field and compare them. The number of consultations is accumulated. When a buyer has consulted more than 10 times, he / she is recorded as an active buyer. The threshold of this number can be obtained by statistically analyzing the behaviors of different types of buyers in the past year and finally confirmed to be 10 times. If the number of active buyers in a certain administrative area exceeds 5, it is regarded as a high-potential buyer gathering area. When generating the high-potential buyer gathering area identification set, the name of each area, its number and the number of active buyers are recorded together. At the same time, the total number of asset types involved by active buyers is marked. When the total number of types exceeds 3, it will be prompted in the additional mark. After encoding all identified buyer gathering areas, the output of the identification set is completed.

[0175] Based on the high-potential buyer gathering area identification set and the asset point location with book value lower than premium valuation in the same geographical block, the coordinates of each asset point are compared with the boundary of the buyer gathering area. If there is a correspondence between the block where the asset point is located and the administrative division field of the gathering area, a cross-attribution match is recorded. When recording the cross result, the asset point number, block number and the unique identifier of the corresponding buyer gathering area are marked. For asset points with multiple buyer gathering areas in the same block and all matching, the coordinates are re-compared, and the specific number of matches is counted. When an asset point is matched with at least one buyer gathering area, it is entered into the cross-attribution match. The blocks are put into the list for subsequent liquidation sorting. If the number of matches is higher than 2, a cumulative mark is made in the record, and a unified sort is performed in combination with the matching situation of other blocks. The recommended liquidation sorting result of each block is generated after a comprehensive score is scored based on the number of matches, the degree by which the book value is lower than the premium valuation, and the maturity debt time field. When the score is higher than a certain preset threshold, it can be listed in the front position. The threshold is set by combining the liquidation cycle and market demand in the city. For example, a score of 80 points or above is considered to have high priority. All geographical block numbers and recommended liquidation sorting results are issued in the same index table to form the basis for regional liquidation strategy.

[0176] The steps for obtaining the clearing funds flow monitoring baseline are as follows:

[0177] Based on the regionalized liquidation strategy, the transaction amount and transaction frequency details of each asset point are extracted, the time field and asset number are aligned, and abnormal records are removed to generate a time series set of asset transaction behavior;

[0178] Based on the asset transaction behavior time series set, the expected turnover fluctuation benchmark for each asset point is calculated using the following formula:

[0179]

[0180] Among them, B t is the expected turnover fluctuation benchmark for the current asset point, AMT p is the transaction volume of the pth period in five consecutive periods, β p is the transaction frequency of the corresponding period, ctf is the cumulative transaction frequency of the asset point in the past thirty days, dtt is the average daily transaction volume of the geographical block where the asset point is located, and ∈ is the total number of circulation transactions of the asset list type of the asset point in the block;

[0181] Based on the expected turnover fluctuation benchmark, all asset points are classified according to asset numbers within their respective geographical blocks, and the transaction volume stability fluctuation range and extreme value boundaries within each block are recorded to generate a clearing funds flow monitoring baseline.

[0182] Specifically, when reading the transaction amount records and transaction frequency details of the asset points based on the regionalized clearing strategy, the historical transaction amount of each asset point is retrieved one by one and compared with the corresponding transaction date. The transaction amount and transaction frequency fields are sorted in chronological order to ensure the formation of continuous time series data. After the sorting is completed, the fields of all records are checked for completeness to exclude records with overlapping or abnormally large intervals in the time field. For example, records with transaction dates that differ by more than 360 days are marked as expired data and eliminated. The 360-day interval is determined with reference to the financial statistical year cycle. If the number of remaining transaction records after elimination is less than 10, additional short sequence markings are performed, and then compared one by one with the asset number and time field. For records that cannot match the same number or date Entries with obvious differences are eliminated. After the elimination is completed, all remaining data are concatenated in chronological order to form a combined sequence of transaction amount and transaction frequency. The recorded transaction amount is judged within the preset range of 0 to 100 million yuan. When the value is found to be lower than 0 or higher than 100 million yuan, it is regarded as a data anomaly and is checked again. If it is confirmed that the data is indeed an entry error, it is removed. After completion, all normal records are associated with their asset numbers and summarized in a time series table. By increasing the transaction date in this sequence table, the transaction amount and frequency distribution of each period can be obtained, and the trend changes of the previous and next periods are also observed. This process finally summarizes the time series set of all asset points and distinguishes their respective transaction trajectories by asset numbers to form a time series set of asset transaction behavior.

[0183] formula: The usefulness of this formula lies in its weighted aggregation of the product of five consecutive periods' transaction volume and the corresponding frequency, associating it with the cumulative transaction frequency over the past thirty days. The average daily transaction volume and number of transactions in the geographic block are incorporated into the cubic relationship to jointly influence this value. This allows for a comprehensive measure of the volatility of the asset's turnover given its recent trading activity and the overall transaction scale of the block in which it is located.

[0184] The first parameter is amt p The steps to obtain amt are: p The transaction amount for the pth period within five consecutive periods can be located in the transaction records. Five adjacent accounting periods can be located in the transaction records. Each accounting period usually corresponds to a week or a month. The specific division method is determined by the settlement cycle of the financial sector. For example, under monthly settlement, the total transaction amount data is read from the settlement list of the most recent five months. If the monthly settlement values of 12,000, 8,000, 9,500, 10,000, and 7,000 yuan are obtained in the transaction records of a certain asset point, then the corresponding values are amt1 = 12,000, amt2 = 8,000, amt3 = 9,500, amt4 = 10,000, and amt5 = 7,000.

[0185] The second parameter β p The steps to obtain β are: p Indicates the same as AMT p The corresponding transaction frequency is calculated by counting and accumulating the number of transactions occurring at the asset point in the same accounting period. If a total of 50 valid transactions are generated in a month, the transaction frequency for that month is recorded as 50. The number of transactions is read one by one in five consecutive periods, and it is ensured that the number of transactions in each period is between 0 and the upper limit. For example, in a business scenario, more than 10,000 transactions in a single month can be considered abnormal and reviewed. In a practical example, if the corresponding numbers of transactions in the five months are 30, 20, 35, 40, and 25 respectively, then β1 = 30, β2 = 20, β3 = 35, β4 = 40, and β5 = 25.

[0186] The third parameter, ctf, is obtained as follows: ctf represents the cumulative transaction frequency of the asset point in the past thirty days. Ctf is obtained by summing up the number of transactions in the 30 days closest to the current date in the transaction record table. For example, if there were 200 valid transactions in the past 30 days, then ctf = 200.

[0187] The fourth parameter, dtt, is obtained as follows: dtt represents the average daily transaction volume of the geographic block where the asset point is located. The transaction volume of all asset points within the block is counted and summed up daily, and then divided by the actual number of business days in the block. If the total daily transaction volume of a block over a period of time is approximately 500,000 yuan, then dtt = 500,000.

[0188] The steps for obtaining the fifth parameter ∈ are as follows: ∈ represents the total number of circulating transactions of the asset list type of the asset point in the block. The transaction details of the same type of assets in the block are read and the transaction numbers are merged. For example, for catering assets, the total number of catering-type transactions in a quarter is 3000, then ∈ = 3000.

[0189] Calculation process: The first step is to calculate the weighted sum of the numerator:

[0190] amt1·β1+amt2·β2+…+amt5·β5;

[0191] Let: amt1 = 12000, β1 = 30, amt2 = 8000, β2 = 20, amt3 = 9500, β3 = 35, amt4 = 10000, β4 = 40, amt5 = 7000, β5 = 25, then:

[0192]

[0193] The second step is to divide the result by ln(ctf+2). If ctf=200, then ctf+2=202.

[0194] ln(202)≈5.31;

[0195]

[0196] Square it:

[0197] (268726) 2 ≈7.228×10 10 ;

[0198] Step 3: Calculate If dtt=500000,∈=3000, then

[0199] ∈+1=3001;

[0200]

[0201] (166.61) 3 ≈4.63×10 6 ;

[0202] The fourth step is to add the two parts and take the square root:

[0203] 7.228×10 10 +4.63×10 6 =7.228463×10 10 ≈7.228×10 10 ;

[0204]

[0205] The results show that for the current asset point, the expected turnover fluctuation benchmark is about 269,000. When the value is close to zero, it can be considered that the transaction volume and frequency are extremely depressed. When the value exceeds 1,000,000, it means that the transaction volume and frequency have increased significantly with the size of the geographical block. t The value is compared horizontally with other asset points in the same block, providing a reference for subsequent fund flow monitoring.

[0206] Based on the calculated expected turnover fluctuation benchmark, each asset point is aggregated by asset number within its geographic block. The expected turnover fluctuation benchmark value is extracted for each asset point and compared with historical transaction fluctuation records. The fluctuation ranges of all asset points are then displayed in the same block dimension for verification. Asset points with values exceeding 500,000 or below 1,000 are additionally labeled during verification. This high and low range is established based on a survey of the actual distribution of daily retail and large-scale purchasing activities in the city's commercial environment. When the fluctuation values of some asset points statistically deviate by more than 20 times from the general level of other asset points, they are noted in the records as high volatility or extremely low volatility. These deviation records are separately collected for subsequent manual screening. The remaining fluctuation values are sorted in the block table, and the highest and lowest values are extracted as the block extreme value boundaries. When multiple asset points have the same highest value or lowest value, they are further sorted by the transaction date field. Finally, after integrating the fluctuation ranges and extreme values of all blocks, a corresponding fluctuation range is formed for each block. This data is aggregated and output as the clearing funds flow monitoring baseline.

[0207] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A merchant asset liquidation method based on data analysis, characterized in that: The following steps are involved: Obtaining geographic relevance based on asset inventory coordinates and transportation network, commercial zoning, and census data; combining the geographic relevance with financial account parameters to establish a multi-dimensional asset feature set; Based on the quantitative correlation of traffic proximity in the multi-dimensional asset feature set, the basic weight of real estate value in the region is calculated to obtain the spatial comprehensive impact results. Based on the spatial comprehensive impact results and the analysis of the impact of the distribution of competing and complementary business formats on asset management, the asset space value factor is established; Calculate the asset base valuation based on the asset space value factor and asset list type, determine the intensity of demand for the target movable assets in different regions, and obtain a location-based liquidation premium valuation. Based on the location-based liquidation premium valuation and combined with balance sheet requirements and identification of potential buyer regions, generate a regionalized liquidation strategy basis; Based on the income and expenditure forecasts based on the regionalized clearing strategy, the statistical mean and standard deviation of the financial accounts are set, the distribution range of transaction amount and frequency is obtained, the expected turnover fluctuation benchmark is calculated, and the clearing funds flow monitoring baseline is established.

2. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the geographic relevance are: Extract the longitude and latitude information of each asset point in the asset list coordinates, call the transportation network node coordinates, the commercial district boundary point set and the census unit center coordinates, calculate the direction vector and distance value of each asset point to the nearest point in the three types of target points, and form the asset list spatial direction-distance basis set; Based on the spatial direction-distance basis set of the asset list, calculate the spatial fusion response value of each asset point based on the three-dimensional vectors of transportation accessibility, commercial proximity and population coverage; Based on the spatial fusion response values, all asset points with spatial fusion response values lower than the average are screened, and coordinate mappings with transportation network nodes, commercial district boundaries, and census centers are established to generate geographic relevance.

3. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the multi-dimensional asset feature set are: Recalling the mapping results between the asset inventory coordinates recorded in the geographic association and the transportation network node coordinates, the commercial district boundary coordinates, and the census unit centroid coordinates; based on the mapped coordinate relationships, calculating the Euclidean distance between each asset point and the corresponding transportation network node; performing adjacency determination on points within the commercial district boundary envelope; and marking whether the census unit number overlaps with the coordinates of the asset point, thereby generating an asset inventory spatial basic association information group; Based on the asset list spatial basic association information group, the operating income field, transaction frequency field and expenditure details field recorded in the financial account parameters are extracted, and the primary key correspondence verification is performed between the asset list number field and the asset list spatial basic association information group, and a one-to-one correspondence is established between the asset point coordinates, the geographic mapping results and the financial account parameters to generate a multi-dimensional asset feature set.

4. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the spatial comprehensive impact results are: Extract the multi-dimensional asset feature set, perform geographical zoning attribution judgment on all asset locations, and generate a geographical zoning comprehensive basic parameter set; Calculate the basic weight of the value of real estate within the region based on the comprehensive basic parameter set of the geographical division; Based on the real estate value basic weight, the geographical division comprehensive basic parameter set is numbered and sorted and matched with the real estate value basic weight to generate a spatial comprehensive impact result.

5. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the asset space value factor are: Based on the spatial comprehensive impact results, the asset coordinate points contained in each geographical zone are extracted, and the asset type identifiers of the corresponding competitive and complementary business formats are read. The total distribution and spatial coordinate positions of each type of asset point within the zone are statistically analyzed by type, and a spatial structure set of the zone asset type is generated; According to the partitioned asset type spatial structure set, the distance value between each asset point and the nearest competing business format, the total number of complementary business formats and the number of heterogeneous non-complementary business formats in the current block are retrieved to generate a set of relative asset point spatial attributes; Based on the asset point spatial relative attribute set, an asset space value factor is calculated.

6. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the location liquidation premium valuation are as follows: Based on the asset space value factor and the asset list type field, the asset space value factor and asset type code corresponding to each asset point are extracted, and the asset code and asset type code are grouped and integrated to confirm the point distribution and type distribution density of each type of asset within the corresponding geographical block, and generate an asset type block distribution set; Extract the asset space value factor, asset type code, valuation sequence of the last twelve months and the maximum valuation value in the last natural month of each asset point from the asset type block distribution set to generate an asset type valuation attribute group; Based on the asset type valuation attribute group, a location liquidation premium valuation is calculated.

7. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the regionalized liquidation strategy are as follows: Based on the location liquidation premium valuation, combined with the distribution of the asset list covered by each geographical block, the asset current book value field and the matured liability record field in the balance sheet are retrieved to establish a matching relationship between the valuation and financial items, and generate an asset valuation and book liability comparison group; Based on the asset valuation and book liability comparison, the coordinates of asset points in each geographic block where the book value is lower than the premium valuation are extracted. At the same time, regional customer visit records, address distribution of completed buyers, and consultation frequency records are read. Each buyer information is categorized and organized according to the administrative division field to generate a set of high-potential buyer concentration area identifiers. Based on the set of identifiers of high-potential buyer clusters, combined with the locations of assets in the geographic blocks where the book value is lower than the premium valuation, a cross-attribution matching is performed between the asset locations and the buyer regions, and a geographic block number and a recommended liquidation ranking result are generated according to the matching results to generate a basis for a regionalized liquidation strategy.

8. The merchant asset liquidation method based on data analysis according to claim 1, characterized in that: The steps for obtaining the clearing funds flow monitoring baseline are as follows: Based on the regionalized liquidation strategy, the transaction amount and transaction frequency details of each asset point are extracted, the time field and asset number are aligned, and abnormal records are removed to generate a time series set of asset transaction behavior; Calculate the expected turnover fluctuation benchmark for each asset point based on the asset transaction behavior time series set; Based on the expected turnover fluctuation benchmark, all asset points are classified according to asset numbers within the geographical blocks to which they belong, and the transaction volume stability variation range and extreme value boundaries within each block are recorded to generate a clearing funds flow monitoring baseline.

9. The merchant asset liquidation system according to any one of claims 1 to 8, wherein: include: The geographic correlation analysis module integrates and compares geospatial data based on asset inventory coordinates, transportation networks, commercial districts, and census data. It calculates the correlation between each merchant's asset location and transportation networks, commercial districts, and population distribution, and generates geographic correlation results. The asset feature set construction module extracts multi-dimensional features of merchant assets based on geographic correlation results and combines them with financial account parameters. It analyzes and summarizes factors such as transportation proximity, population density, and income level to generate an asset feature set. The spatial value assessment module, based on the asset feature set, quantitatively correlates factors such as transportation proximity, population density, and income level, calculates the basic weight of real estate within the region, and evaluates and obtains the comprehensive spatial impact results. This comprehensive spatial impact result analyzes the impact of regional competition and complementary business distribution on asset management, and generates the asset spatial value factor. The asset valuation and regional demand analysis module calculates the basic valuation of merchant assets based on the asset space value factor and asset list type, assesses the intensity of demand for target movable assets in different regions, and combines regional liquidation premium valuation to determine buyer demand in potential regions and generate regional liquidation demand valuation; The financial settlement monitoring module, based on regional settlement demand valuation and balance sheet requirements, forecasts the income and expenditure of merchants' financial accounts, calculates the statistical mean and standard deviation of financial accounts, obtains the distribution range of transaction amount and frequency, calculates the expected turnover fluctuation benchmark, and establishes a settlement fund flow monitoring baseline.

Images