House rental information dynamic recommendation method combined with real-time supply and demand data

Abstract

The application discloses a house rental information dynamic recommendation method combined with real-time supply and demand data, and particularly relates to the technical field of information recommendation; the method comprises the following steps: collecting original space boundary data to generate house rental boundary data; analyzing the spatial distribution characteristics of supply-side house source data and demand-side user data to determine supply and demand hot regions; analyzing the overlapping relationship between different spatial scales to generate spatial overlapping relationship data; evaluating the cross degree of supply and demand data to determine the influence range of regional boundary overlap; correcting the spatial weight of supply-side house source data and demand-side user data to obtain corrected supply and demand hot regions; combining user historical recommendation feedback data to generate user recommendation preference data; and recommending house rental information to users according to the corrected supply and demand hot regions and the user recommendation preference data. The method effectively reduces the data deviation caused by regional boundary overlap and improves the accuracy and individualization degree of house rental information recommendation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information recommendation, more particularly, the present application relates to a dynamic recommendation method for housing rental information combined with real-time supply and demand data. BACKGROUND

[0002] Currently, in the housing rental information recommendation technology, the housing source and demand data in different spatial regions such as administrative districts, sub-districts or commercial circles are analyzed respectively, and the housing is recommended based on the spatial distribution of housing supply and user demand.

[0003] However, the current housing rental information recommendation technology in processing different spatial scale supply and demand information, due to the scale difference and boundary overlap phenomenon between the administrative district, sub-district and commercial circle space boundary, causes the cross and repeated counting of supply side housing data and demand side user data in different scale space region, thus forming data deviation, causing the matching degree between the housing rental information recommendation result and the user real demand to be reduced. At the same time, due to the neglect of the individualized preference characteristics of users in the historical recommendation information feedback, the accuracy of the housing rental information recommendation is insufficient, which cannot meet the actual housing demand of users' individualization and diversification. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a dynamic recommendation method for housing rental information combined with real-time supply and demand data to solve the problems raised in the above background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] The dynamic recommendation method for housing rental information combined with real-time supply and demand data comprises the following steps:

[0007] S1: Collect the original spatial boundary data of administrative districts, sub-districts and commercial circles, and perform unified spatial coding to generate housing rental boundary data;

[0008] S2: According to the housing rental boundary data, analyze the spatial distribution characteristics of the supply side housing data and the demand side user data from different spatial scales to obtain the supply and demand hot area;

[0009] S3: According to the housing rental boundary data, analyze the overlapping relationship between different spatial scales to obtain spatial overlapping relationship data;

[0010] S4: According to the supply and demand hot area and the spatial overlapping relationship data, evaluate the cross degree of supply and demand data caused by the overlapping of regional boundaries, and determine the influence range of the overlapping of regional boundaries;

[0011] S5: According to the influence range of the region boundary overlap, the spatial weight correction of the supply side housing data and the demand side user data is carried out, and the corrected supply and demand hot spot area is generated;

[0012] S6: According to the influence range of the region boundary overlap, the sensitivity of the user to the recommended result is determined in combination with the user historical recommendation feedback data, and the user recommendation preference data is generated;

[0013] S7: According to the corrected supply and demand hot spot area and the user recommendation preference data, the housing rental information is dynamically recommended to the user.

[0014] In a preferred embodiment, S1, specifically:

[0015] Collecting original spatial boundary data of administrative regions, original spatial boundary data of subdistricts and original spatial boundary data of commercial districts;

[0016] The original spatial boundary data of administrative regions, the original spatial boundary data of subdistricts and the original spatial boundary data of commercial districts are subjected to spatial coordinate calibration and unified spatial position coding processing to obtain spatial boundary data of administrative regions, spatial boundary data of subdistricts and spatial boundary data of commercial districts;

[0017] Based on the unified spatial position coding of the spatial boundary data of administrative regions, the spatial boundary data of subdistricts and the spatial boundary data of commercial districts, data fusion processing is carried out to generate housing rental boundary data.

[0018] In a preferred embodiment, S2, specifically:

[0019] Based on the housing rental boundary data, spatial grids are established according to the administrative region scale, the subdistrict scale and the commercial district scale;

[0020] The supply side housing data is mapped according to the geographical position, the number of housing sources in each grid unit is counted, and a supply density matrix is generated;

[0021] The demand side user data is mapped according to the geographical position, the number of user demands in each grid unit is counted, and a demand density matrix is generated;

[0022] The supply density matrix and the demand density matrix are superimposed in the corresponding grid units, and the supply and demand intensity value is calculated;

[0023] The supply and demand intensity value is compared with the preset supply and demand intensity threshold value, and the supply and demand hot spot unit is determined, and the supply and demand hot spot area is formed.

[0024] In a preferred embodiment, S3, specifically:

[0025] Taking the administrative district spatial boundary data and the subdistrict spatial boundary data as a first pair set, performing spatial overlay operation to obtain an administrative district and subdistrict overlapping region set;

[0026] Taking the subdistrict spatial boundary data and the commercial circle spatial boundary data as a second pair set, performing spatial overlay operation to obtain a subdistrict and commercial circle overlapping region set;

[0027] Merging the overlapping region set according to the spatial coding to form a multi-scale spatial overlapping record table;

[0028] In the multi-scale spatial overlapping record table, generating a spatial overlapping proportion value for each spatial overlapping record;

[0029] Based on the multi-scale spatial overlapping record table and the corresponding spatial overlapping proportion value, forming spatial overlapping relationship data.

[0030] In a preferred embodiment, S4, specifically:

[0031] For each spatial overlapping record in the multi-scale spatial overlapping record table, extracting the supply side housing quantity and the demand side user demand times of the corresponding grid cell from the supply and demand hot spot region according to the spatial coding;

[0032] For each spatial overlapping record, calculating a cross-region supply and demand cross index by using the supply side housing quantity and the demand side user demand times;

[0033] Comparing the cross-region supply and demand cross index with a preset cross threshold value, and selecting the spatial overlapping record with the cross-region supply and demand cross index greater than the preset cross threshold value as a region boundary overlapping influence record;

[0034] Summarizing the spatial coding of all region boundary overlapping influence records to generate a region boundary overlapping influence range.

[0035] In a preferred embodiment, S5, specifically:

[0036] Based on the spatial overlapping proportion value, adjusting the supply side housing data and the demand side user data of each grid cell belonging to the region boundary overlapping influence range by spatial weight to generate adjusted demand side user data and demand side user data;

[0037] Re-mapping the adjusted supply side housing data and the demand side user data into the corresponding spatial grid cell, and re-counting the adjusted housing quantity and the adjusted user demand times of each spatial grid cell;

[0038] Based on the adjusted housing quantity and the adjusted user demand times, generating a corrected supply and demand hot spot region.

[0039] In a preferred embodiment, S6, specifically:

[0040] Extract the geographical location, click times, browsing time and close times of each recommendation record from the user history recommendation feedback data, and associate the region boundary overlap influence range according to the spatial coding;

[0041] For the recommendation records located in the region boundary overlap influence range, the click times, browsing time and close times of each user in the spatial grid unit are counted, and a user feedback index matrix is generated;

[0042] According to the user feedback index matrix, the sensitivity coefficient of each user in the spatial grid unit is calculated;

[0043] The sensitivity coefficient is combined with the spatial coding to generate user recommendation preference data.

[0044] In a preferred embodiment, S6, specifically:

[0045] Based on the corrected supply and demand hot area, the house source located in the supply and demand hot area is selected from the supply side house source data to generate a recommended house source candidate set;

[0046] According to the user recommendation preference data, the house source meeting the user recommendation preference is selected from the recommended house source candidate set to form a personalized recommended house source set;

[0047] The personalized recommended house source set is sorted according to the spatial coding of the spatial grid unit and the sensitivity coefficient in the user recommendation preference data, and the corresponding house rental information is pushed to the user in order from high to low.

[0048] The technical effects and advantages of the house rental information dynamic recommendation method combining real-time supply and demand data:

[0049] The spatial consistency and integrity of data fusion are ensured by spatial coordinate calibration and unified coding of the original spatial boundary data of administrative districts, subdistricts and business circles; the spatial distribution characteristics of supply-side housing data and demand-side user data are analyzed from different spatial scales, so that the supply-demand hot areas can be quickly identified and the recommendation efficiency is improved; the overlapping relationship of boundaries of different spatial scales is quantified, and spatial overlapping relationship data is constructed, which provides accurate basis for correction; the cross degree of supply-demand data caused by the overlapping of regional boundaries is evaluated based on the supply-demand hot areas and the spatial overlapping relationship data, and the affected spatial range is determined; the spatial weight of supply-side housing data and demand-side user data is corrected, which eliminates the repeated counting and deviation caused by the overlapping of boundaries, so that the hot areas are more real and reliable; the sensitivity of users to different spatial positions is quantified by combining the historical recommendation feedback data of users, user recommendation preference data is generated, and the individual matching degree of recommended content is enhanced; the housing information meeting the actual needs of users is dynamically pushed by using the corrected supply-demand hot areas and user recommendation preference data, which effectively improves the hit rate and user satisfaction of recommendation, and reduces the risk of information redundancy and misguidance; the spatial matching accuracy, real-time response efficiency and user individual recommendation level of the housing rental information recommendation system are improved. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 A schematic diagram of the dynamic recommendation method of housing rental information combined with real-time supply-demand data is given. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0052] Embodiment 1:

[0053] Figure 1 The dynamic recommendation method of housing rental information combined with real-time supply-demand data is given, which includes the following steps:

[0054] S1: Collecting the original spatial boundary data of administrative districts, subdistricts and business circles, and performing unified spatial coding to generate housing rental boundary data;

[0055] S2: According to the housing rental boundary data, the spatial distribution characteristics of supply-side housing data and demand-side user data are analyzed from different spatial scales, and the supply-demand hot areas are obtained;

[0056] S3: According to the housing rental boundary data, the overlapping relationship between different spatial scales is analyzed to obtain spatial overlapping relationship data;

[0057] S4: According to the supply and demand hot spot area and the spatial overlapping relationship data, the cross degree of supply and demand data caused by the overlapping of regional boundaries is evaluated, and the influence range of the overlapping of regional boundaries is determined;

[0058] S5: According to the influence range of the overlapping of regional boundaries, the spatial weight correction of the supply side housing data and the demand side user data is carried out, and the corrected supply and demand hot spot area is generated;

[0059] S6: According to the influence range of the overlapping of regional boundaries, the sensitivity of the user to the recommended results is determined in combination with the user historical recommendation feedback data, and the user recommendation preference data is generated;

[0060] S7: According to the corrected supply and demand hot spot area and the user recommendation preference data, the housing rental information is dynamically recommended to the user.

[0061] S1: Collect the original spatial boundary data of administrative district, district and business circle, and carry out unified spatial coding to generate housing rental boundary data, including:

[0062] Collect the original spatial boundary data of administrative district, the original spatial boundary data of district and the original spatial boundary data of business circle;

[0063] The original spatial boundary data of administrative district refers to the administrative region division data, which is usually recorded in the form of geographical coordinate point sequence to record the position of administrative district boundary. The original spatial boundary data of administrative district can be represented as a series of continuous longitude and latitude coordinate points forming a closed polygon. Through the closed polygon surrounded by continuous geographical coordinate points, the range of administrative region can be represented; The original spatial boundary data of district refers to the spatial range data of the divided urban area or specific district, which is also recorded in the form of geographical coordinate point set to form the closed boundary of district spatial range; The original spatial boundary data of business circle represents the area where residents' commercial activities are frequent in the city, which is composed of a complete closed polygon space area through a continuous set of geographical coordinate points. The original spatial boundary data of administrative district, the original spatial boundary data of district and the original spatial boundary data of business circle gradually decrease in spatial scale, among which the administrative district spatial boundary data covers the widest range, the district spatial boundary data covers the second widest range, and the business circle spatial boundary data covers the smallest range.

[0064] The original spatial boundary data of administrative district, the original spatial boundary data of district and the original spatial boundary data of business circle are subjected to spatial coordinate calibration and unified spatial position coding processing to obtain the spatial boundary data of administrative district, the spatial boundary data of district and the spatial boundary data of business circle;

[0065] The collected original spatial boundary data of administrative regions, original spatial boundary data of subdistricts and original spatial boundary data of commercial districts are subjected to spatial coordinate calibration processing. Spatial coordinate calibration refers to unified spatial coordinate system conversion of the original spatial boundary data collected from different data sources to meet the requirements of the same standard spatial coordinate system. For example, first, a national unified standard spatial coordinate system is selected as the unified coordinate reference; each geographic location coordinate point in the original spatial boundary data of administrative regions, original spatial boundary data of subdistricts and original spatial boundary data of commercial districts is converted from the respective original coordinate system to the national unified standard coordinate system; the conversion process includes geographic location coordinate conversion, projection conversion and coordinate offset correction processing. After the spatial coordinate calibration is completed, the calibrated spatial boundary data of administrative regions, spatial boundary data of subdistricts and spatial boundary data of commercial districts are subjected to unified spatial position coding processing. Spatial position coding processing refers to coding the unified coordinate spatial boundary data according to unified spatial position coding rules. Spatial position coding rules refer to assigning a unique and accurate position relationship expression coding identifier to each geographic location coordinate point according to the coordinate system. For example, each coordinate point in the administrative region spatial boundary data is subjected to unified spatial position coding, i.e., each spatial position of the administrative region spatial boundary data is assigned a unique spatial position coding; similarly, each coordinate point in the subdistrict spatial boundary data and commercial district spatial boundary data is subjected to unified spatial position coding, and each spatial position in the subdistrict and commercial district is assigned a corresponding spatial position coding. Through the above processing, the spatial boundary data of administrative regions, spatial boundary data of subdistricts and spatial boundary data of commercial districts achieve spatial coordinate and position coding unification.

[0066] Based on the unified spatial position coded spatial boundary data of administrative regions, spatial boundary data of subdistricts and spatial boundary data of commercial districts, data fusion processing is performed to generate housing rental boundary data;

[0067] After the unified spatial position coding processing is completed, the administrative district spatial boundary data, the subdistrict spatial boundary data and the commercial circle spatial boundary data after unified spatial position coding are obtained respectively; based on the administrative district spatial boundary data, the subdistrict spatial boundary data and the commercial circle spatial boundary data after unified spatial position coding, data fusion processing is performed to obtain the housing rental boundary data. The data fusion processing refers to integrating the spatial boundary data of three scales under the unified spatial position coding system, so that the administrative district spatial boundary data, the subdistrict spatial boundary data and the commercial circle spatial boundary data are unified in a complete data system; the data fusion specifically comprises the following steps: first, a unified spatial position coding index table is established, the unified spatial position codes of the administrative districts, the subdistricts and the commercial circles are respectively registered in the unified spatial position coding index table, and a corresponding relationship between the codes is established; each spatial position in the administrative district spatial boundary data, the subdistrict spatial boundary data and the commercial circle spatial boundary data is associated and matched through the unified spatial position coding index table; a multi-scale spatial boundary data set is formed through the associated matching process of the spatial boundary data, the multi-scale spatial boundary data set representing the spatial position and boundary relationship existing between the administrative districts, the subdistricts and the commercial circles of three scales; spatial data merging processing is performed on the multi-scale spatial boundary data set, so that the overlapping parts between the administrative district spatial boundary, the subdistrict spatial boundary and the commercial circle spatial boundary are standardized and integrated, and finally the housing rental boundary data in a unified standard format is generated.

[0068] S2: According to the housing rental boundary data, the spatial distribution characteristics of the supply-side house source data and the demand-side user data are analyzed from different spatial scales to obtain a supply-demand hotspot area, comprising:

[0069] Based on the housing rental boundary data, a spatial grid is established according to the administrative district scale, the subdistrict scale and the commercial circle scale;

[0070] The administrative district scale spatial grid refers to taking administrative district spatial boundary data as a spatial constraint range, dividing the spatial range of the administrative district into a plurality of equal-area or equal-length-width grid regions according to a rule, and the boundary of each grid region is represented by a plurality of geographic coordinate points. For example, a plurality of square or rectangular grids can be divided in the spatial range of the administrative district at equal intervals of longitude and latitude, each grid is referred to as a spatial grid unit, each spatial grid unit is mutually non-overlapping and closely arranged, and all spatial grid units collectively cover the entire spatial range of the administrative district. The district scale spatial grid refers to taking district spatial boundary data as a spatial constraint range, dividing the spatial range of the district into a plurality of grid units according to the same rule, and the area or size of each grid unit is the same as or scaled from the spatial grid unit of the administrative district scale. For example, the spatial range of the district is divided into a plurality of regular grid units, each grid unit is also represented by a set of continuous geographic position coordinate points, and all grid units collectively cover the spatial range of the district. The commercial circle scale spatial grid refers to taking commercial circle spatial boundary data as a constraint, and dividing into a plurality of spatial grid units according to the above method, the area of the commercial circle spatial grid unit is smaller, and each grid unit covers the spatial range of the commercial circle. The establishment process of the administrative district scale spatial grid, the district scale spatial grid and the commercial circle scale spatial grid ensures that the positional relationship between the grid units is clear and has a unified spatial position code.

[0071] The supply side housing data is mapped according to the geographic position, the number of housing sources in each grid unit is counted, and a supply density matrix is generated.

[0072] After the spatial grid is established, the supply side housing data in the housing rental market is respectively mapped into the administrative district scale spatial grid, the district scale spatial grid and the commercial circle scale spatial grid according to the actual geographic position of the housing. The supply side housing data refers to the spatial geographic position information and the housing attribute information of each set of houses provided by the lessor, which is usually represented by longitude and latitude coordinates. For example, the rental house is located in the district, and the house position can be uniquely determined by longitude and latitude coordinates. The geographic position of the house is matched with the spatial grid unit one by one to determine the spatial grid unit to which the rental house belongs. After the spatial position matching of all housing data is completed, the number of successfully matched housing sources in each spatial grid unit is counted, and the counting method of the number of housing sources is that the successfully matched housing sources in the spatial grid unit are counted one by one, and the total number of housing sources in the grid unit is calculated. The above housing number counting operation is performed on all spatial grid units one by one, and after the counting is completed, the overall distribution state of the number of housing sources in the spatial grid unit is formed, which is defined as the supply density matrix. The supply density matrix takes the spatial grid unit as a unit, records the number of housing sources in the spatial grid unit, is saved in the form of a matrix, and expresses the distribution of housing data in space.

[0073] Map demand-side user data by geographic location, count the number of user demands in each grid unit, and generate a demand density matrix.

[0074] Demand-side user data is mapped to spatial grid cells at the administrative, district, and business district scales based on the geographic location information of user demands. Demand-side user data refers to the data generated when users submit rental requests for specific properties or areas, including the user's geographic location or intended location coordinates, number of visits, etc. For example, when a user requests to rent a property near a specific business district, the user's demand information includes the desired location of the property, identified by latitude and longitude coordinates. User demand location information is matched one by one with spatial grid cells to determine the corresponding spatial grid cell, and the number of times a user submits a demand within each spatial grid cell is counted. The counting method is to sum the number of user demand records within each spatial grid cell to obtain the total number of user demand counts within that spatial grid cell. This demand count counting process is performed on all spatial grid cells to complete the comprehensive statistics of demand-side user data within the spatial grid cells, ultimately forming the overall spatial distribution of user demand, defined as the demand density matrix. Each spatial grid cell in the demand density matrix records the number of user demand counts, which can intuitively reflect the density of user demand in different spatial locations.

[0075] The supply density matrix and the demand density matrix are superimposed with corresponding grid cells to calculate the supply and demand intensity value;

[0076] The supply density matrix and demand density matrix are superimposed one-to-one to each spatial grid cell, simultaneously considering the number of housing units and the frequency of user demand within the same spatial grid cell. The sum of the number of housing units and the frequency of user demand within each spatial grid cell is then calculated to obtain the comprehensive supply and demand index value for each spatial grid cell, defined as the supply and demand intensity value. The supply and demand intensity value reflects the overall activity level of housing supply and rental demand within the spatial grid cell. This superimposed calculation is performed on all spatial grid cells to complete the calculation of the supply and demand intensity value for each spatial grid cell, which is then saved as a supply and demand intensity value distribution matrix.

[0077] The supply and demand intensity values ​​are compared with preset supply and demand intensity thresholds to identify supply and demand hotspot units, and these are then aggregated to form supply and demand hotspot areas.

[0078] The supply-demand intensity values of the spatial grid cells are compared with preset supply-demand intensity threshold values respectively, the preset supply-demand intensity threshold values are standard reference values set according to actual application experience; when the supply-demand intensity value of the spatial grid cell is greater than or equal to the supply-demand intensity threshold value, it is determined as a supply-demand hotspot cell; when the supply-demand intensity value of the spatial grid cell is less than the preset supply-demand intensity threshold value, it is not determined as a supply-demand hotspot cell; all the spatial grid cells determined as the supply-demand hotspot cells are collected according to the spatial positions to form a supply-demand hotspot area; the supply-demand hotspot area completely saves the positions of the active spatial grid cells.

[0079] S3: According to the housing rental boundary data, the overlapping relationship between different spatial scales is analyzed to obtain spatial overlapping relationship data, including:

[0080] Taking the administrative district spatial boundary data and the subdistrict spatial boundary data as a first pairing set, spatial overlay operation is performed to obtain an overlapping area set of the administrative district and the subdistrict;

[0081] The administrative region spatial boundary data refers to the spatial range of an administrative region. The administrative region spatial range data is composed of a set of continuous geographic coordinate points. The geographic coordinate points are connected in sequence to form a closed region of a spatial polygon. For example, the spatial boundary data of an administrative region represents a set of all geographic position coordinate points within the range of the administrative region. The geographic position coordinate points are connected in sequence to form the overall boundary of the administrative region. The district spatial boundary data refers to the range data of a delimited district spatial region. The district spatial boundary data is also represented by a set of continuous geographic position coordinate points, which constitute a complete spatial closed boundary within the district range. For example, all geographic position coordinate points within the district range are connected in sequence to form a closed district spatial boundary. Spatial overlay operation is performed on the administrative region spatial boundary data and the district spatial boundary data as input data. The spatial overlay operation refers to spatial intersection calculation between the administrative region range represented by the administrative region spatial boundary data and the district range represented by the district spatial boundary data, to determine the position region of intersection and save it as a new spatial region. The spatial overlay operation specifically includes: performing position relationship calculation on each spatial position coordinate point in the administrative region spatial boundary data and each spatial position coordinate point in the district spatial boundary data. The position relationship calculation includes judging whether each spatial position coordinate point in the administrative region is located within the district spatial range. If the spatial position coordinate point in the administrative region is located within the district spatial range represented by the district spatial boundary data, the spatial position coordinate point is recorded in the overlap region set of the administrative region and the district. If the spatial position coordinate point in the administrative region is not located within the district spatial range represented by the district spatial boundary data, the spatial position coordinate point is not recorded in the overlap region set of the administrative region and the district. The position relationship of all geographic position coordinate points in the administrative region spatial range is judged according to the above method. Finally, all spatial position coordinate points of intersection and overlap in the administrative region and the district spatial range are obtained. The spatial position coordinate points of intersection and overlap are summarized to form a new spatial boundary region, which is defined as the overlap region set of the administrative region and the district. The overlap region set of the administrative region and the district records all intersection region information of the spatial relationship between the administrative region and the district.

[0082] The district spatial boundary data and the commercial circle spatial boundary data are taken as a second pair set for spatial overlay operation to obtain an overlap region set of the district and the commercial circle.

[0083] The second paired set is established by taking the district spatial boundary data and the commercial circle spatial boundary data as input data. The district spatial boundary data represents the spatial range of the planned district, and the commercial circle spatial boundary data represents the spatial range of the region where the urban residents frequently conduct commercial activities. Both the district spatial boundary data and the commercial circle spatial boundary data are constituted by a closed polygon region of a continuous geographical position coordinate point set, and the range of the district is larger than the range of the commercial circle in the spatial scale. The spatial superposition operation is consistent with the spatial superposition operation of the administrative region spatial boundary data and the district spatial boundary data, and specifically, it is determined whether each geographical position coordinate point in the district spatial boundary data falls within the spatial range represented by the commercial circle spatial boundary data. If the spatial position coordinate point within the district is located within the spatial range represented by the commercial circle spatial boundary data, the spatial position coordinate point is recorded in the overlapping region set of the district and the commercial circle. If it is not within the spatial range of the commercial circle, it is not recorded in the overlapping region set of the district and the commercial circle. The judgment of all coordinate points within the spatial range of the district is sequentially completed, and all coordinate point information of the overlapping position of the district and the commercial circle is collected to form the overlapping region set of the district and the commercial circle. The overlapping region set of the district and the commercial circle completely records the spatial position relationship between the district scale and the commercial circle scale.

[0084] The overlapping region sets are merged according to the spatial coding to form a multi-scale spatial overlapping record table.

[0085] Based on the overlapping region set of the administrative region and the district and the overlapping region set of the district and the commercial circle, a unified spatial position coding rule is used to give a unified spatial position coding to the geographical position coordinate point recorded in each overlapping region set, and the overlapping region sets are merged. The spatial position coding information is marked on each spatial position coordinate point in each overlapping region set. According to the uniqueness of the spatial position coding, the coordinate point data records with the same spatial position coding in the overlapping region set of the administrative region and the district and the overlapping region set of the district and the commercial circle are matched to establish the corresponding relationship between the spatial position codings. Then, the coordinate points with the same spatial position coding are merged in the spatial region boundary, that is, the spatial position data records in the unified standardized format are constituted. After the merging processing is completed, the overall data record formed is saved as a multi-scale spatial overlapping record table. The multi-scale spatial overlapping record table records the spatial intersection relationship among the administrative region, the district and the commercial circle. Each record contains the unified spatial position coding, the spatial range and the scale information of the overlapping region, and describes the spatial position relationship and the spatial scale intersection state among the three scales.

[0086] In the multi-scale spatial overlapping record table, a spatial overlapping proportion value is generated for each spatial overlapping record.

[0087] The spatial overlap ratio value is the ratio between the spatial area of the overlapping region and the spatial area of the smaller scale region, which represents the proportion of the cross-region between the two scales in the smaller scale spatial region. First, determine the corresponding spatial area for each spatial overlap record; perform ratio operation between the spatial overlap area and the spatial area of the smaller scale region. The ratio operation means dividing the spatial overlap area by the spatial area of the smaller scale region, and the result is the spatial overlap ratio value; perform the above calculation operation on each record in the multi-scale spatial overlap record table to obtain the spatial overlap ratio value corresponding to each spatial overlap record, which is recorded in the multi-scale spatial overlap record table as a new field.

[0088] Based on the multi-scale spatial overlap record table and the corresponding spatial overlap ratio value, form the spatial overlap relationship data;

[0089] After completing the calculation and recording of the spatial overlap ratio value, use the multi-scale spatial overlap record table and the newly added spatial overlap ratio value as input data to generate spatial overlap relationship data. The spatial overlap relationship data is the processed data of the multi-scale spatial overlap record table, which represents the spatial overlap relationship and overlap ratio between the administrative district, the district, and the commercial circle scales. First, perform data classification processing on the unified spatial location code, scale information, and spatial overlap ratio value recorded in the multi-scale spatial overlap record table, and sort and summarize them according to the scale from large to small or from small to large; perform standard data format conversion on the sorted and summarized data to generate spatial overlap relationship data; the spatial overlap relationship data saves the information such as the spatial position of the overlapping region between spatial scales, the overlapping area, and the spatial overlap ratio value in a standardized spatial position data format.

[0090] S4: According to the supply and demand hot spot region and the spatial overlap relationship data, evaluate the cross degree of the supply and demand data caused by the region boundary overlap, and determine the influence range of the region boundary overlap, including:

[0091] For each spatial overlap record in the multi-scale spatial overlap record table, extract the supply side housing quantity and demand side user demand times of the corresponding grid cell from the supply and demand hot spot region according to the spatial code;

[0092] The multi-scale spatial overlap record table records the unified spatial position code, the spatial overlap area range, the spatial scale cross relationship and the spatial overlap proportion value of the overlapping area between the administrative scale, the district scale and the business circle scale. Each spatial overlap record has a unique unified spatial position code to represent the spatial position of the spatial overlap area. The supply and demand hot spot area is composed of spatial grid units, and each spatial grid unit is identified by a unified spatial position code. Each spatial grid unit records the supply side housing quantity and the demand side user demand times, wherein the supply side housing quantity represents the total number of all rentable houses in the grid unit, and the demand side user demand times represents the total number of user rental demand in the grid unit. For example, in the business circle scale spatial grid unit, the supply side housing quantity is the sum of all houses located in the spatial grid unit, and the demand side user demand times is the number of user rental demand in the spatial grid unit. The supply side housing quantity and the demand side user demand times of each spatial grid unit are recorded in the supply and demand hot spot area and associated with the unified spatial position code and the spatial grid unit. According to the unified spatial position code, through each spatial overlap record in the multi-scale spatial overlap record table, the supply side housing quantity and the demand side user demand times recorded in the same spatial grid unit of the spatial overlap area are queried and extracted: the unified spatial position code marked by a spatial overlap record in the multi-scale spatial overlap record table is queried in the supply and demand hot spot area to find the spatial grid unit corresponding to the same spatial position code; if the spatial position codes are consistent, the supply side housing quantity and the demand side user demand times corresponding to the spatial grid unit are extracted as the supply and demand data of the spatial overlap record; if no corresponding spatial position code is found, the spatial overlap record is marked to record the data state of the unmatched supply side housing quantity and demand side user demand times; the above query and extraction process is repeated for each record in the multi-scale spatial overlap record table until each spatial overlap record has corresponding supply side housing quantity and demand side user demand times or corresponding unmatched mark. Through the above process, the supply and demand data extraction of the spatial overlap record in the multi-scale spatial overlap record table is completed.

[0093] For each spatial overlap record, the supply and demand cross index is calculated by using the supply side housing quantity and the demand side user demand times.

[0094] According to the number of supply-side housing and the number of demand-side user demand extracted from each spatial overlap record in the multi-scale spatial overlap record table, the cross-regional supply and demand cross index of each spatial overlap record is calculated; the cross-regional supply and demand cross index is an index for quantifying and measuring the interaction degree between the supply of housing and the demand of users in the spatial overlap area, reflecting the cross state of the supply and demand relationship between the two scales. The calculation method of the cross-regional supply and demand cross index is as follows: first, determine the total number of supply-side housing and the total number of demand-side user demand recorded in each spatial overlap record; multiply the number of supply-side housing and the number of demand-side user demand, and then divide the sum of the number of supply-side housing and the number of demand-side user demand, the result is defined as the cross-regional supply and demand cross index; for example, when the number of supply-side housing in a spatial position corresponding to a spatial overlap record is one hundred, and the number of demand-side user demand is fifty, the cross-regional supply and demand cross index is the quotient of the product of the number of housing and the number of demand and the total number of the number of housing and the number of demand; through the above calculation formula, the cross-regional supply and demand cross index of each spatial overlap record in the multi-scale spatial overlap record table is calculated, and the calculation result is recorded in the multi-scale spatial overlap record table as a newly added field; the cross-regional supply and demand cross index records the quantitative result of the interaction state between the supply and demand of the administrative district scale, the district scale and the business circle scale.

[0095] The cross-regional supply and demand cross index is compared with the preset cross threshold value, and the spatial overlap record with a cross-regional supply and demand cross index greater than the preset cross threshold value is selected as the regional boundary overlap influence record.

[0096] The cross-regional supply and demand cross index of each spatial overlap record is compared with the preset cross threshold value; the preset cross threshold value is a standard value set in advance, which is used to judge whether the cross-regional supply and demand cross index meets the standard of affecting the spatial overlap area; the comparison method is as follows: first, take out the cross-regional supply and demand cross index in each spatial overlap record, and compare the cross-regional supply and demand cross index with the preset cross threshold value; if the cross-regional supply and demand cross index is greater than the preset cross threshold value, the spatial overlap record is marked as a regional boundary overlap influence record; if the cross-regional supply and demand cross index is less than or equal to the preset cross threshold value, the spatial overlap record is not marked as a regional boundary overlap influence record; repeat the above comparison operation for all spatial overlap records in the multi-scale spatial overlap record table, and finally determine the range of all regional boundary overlap influence records.

[0097] The spatial codes of all regional boundary overlap influence records are summarized to generate the regional boundary overlap influence range.

[0098] The uniform spatial position codes in the spatial overlap records marked as the region boundary overlap influence records are summarized; the uniform spatial position codes represent spatial position and scale information of each region boundary overlap influence record; the summary method is: first, the uniform spatial position code information corresponding to each record is extracted from the region boundary overlap influence records; the extracted uniform spatial position code information is processed to remove duplicates, and unique spatial position codes are retained; the retained spatial position codes are sorted, summarized, and arranged to form a region boundary overlap influence range; the region boundary overlap influence range saves spatial position codes and scale intersection state information that are all identified as being affected by spatial scale overlap.

[0099] S5: According to the influence range of the region boundary overlap, the spatial weight correction of the supply side housing data and the demand side user data is performed to generate a corrected supply and demand hot area, including:

[0100] Based on the spatial overlap proportion value, the spatial weight adjustment of the supply side housing data and the demand side user data belonging to each grid cell in the region boundary overlap influence range is performed to generate adjusted demand side user data and demand side user data.

[0101] The spatial weight adjustment is specifically: for each spatial grid cell in the region boundary overlap influence range, the corresponding spatial overlap proportion value is extracted, and the spatial overlap proportion value represents the spatial proportion of the overlapping region in the smaller scale region; for the original statistical supply side housing data in each spatial grid cell, the spatial overlap proportion value is first used for weight adjustment, that is, the original supply side housing data in the grid cell is multiplied by the spatial overlap proportion value, and the result is the adjusted supply side housing data; for example, when the original statistical supply side housing data in the spatial grid cell is several sets, and the spatial overlap proportion value is the ratio of the overlapping area to the smaller scale area, the adjusted supply side housing data is the result of multiplying the original statistical supply side housing data in the grid cell by the spatial overlap proportion value; the supply side housing data adjustment calculation of all spatial grid cells in the region boundary overlap influence range is completed according to the above method, and the overall adjusted supply side housing data distribution is obtained.

[0102] Similarly, for each spatial grid cell in the area boundary overlapping influence range, the demand side user data is also subjected to a spatial weight adjustment operation consistent with the supply side housing data, i.e., multiplying the original demand side user data in the grid cell by the spatial overlap proportion value; for example, if the original statistical user demand times in the spatial grid cell is several times, then the adjusted demand side user data is the product of the original demand side user data in the grid cell and the spatial overlap proportion value; by analogy, the spatial weight adjustment operation is sequentially performed on all spatial grid cells in the area boundary overlapping influence range to obtain the distribution state of the adjusted demand side user data in the spatial grid cell.

[0103] After the supply side housing data and the demand side user data are adjusted, the effective correction of the spatial scale overlapping influence is achieved, and the real housing supply and user rental demand distribution characteristics are more accurately reflected.

[0104] The adjusted supply side housing data and the demand side user data are remapped into the corresponding spatial grid cells, and the adjusted housing quantity and the adjusted user demand times in each spatial grid cell are re-counted.

[0105] The adjusted supply side housing data and the demand side user data respectively represent the actual housing quantity and user demand times distribution in the spatial grid cell in the area boundary overlapping influence range after the spatial weight adjustment. After the spatial weight adjustment is completed, the adjusted supply side housing data and the demand side user data need to be remapped into the original corresponding spatial grid cells; the remapping operation is to replay the adjusted supply and demand data to the original corresponding spatial grid cells; for example, before adjustment, several houses are counted in the spatial grid cell, and after the weight adjustment, the number of houses changes, so the adjusted number needs to be re-assigned to the spatial grid cell; the mapping method of the demand side user data is the same, i.e., the new number of user demand times after adjustment is also replayed to the original corresponding spatial grid cell; through the remapping processing, the data in the spatial grid cell can be updated, and the real supply and demand data state after the spatial overlap correction can be reflected.

[0106] After the remapping is completed, the adjusted supply-side housing quantity and the adjusted demand-side user demand times are respectively re-counted for each spatial grid cell; the adjusted housing quantity counting method is: re-counting and summarizing the housing quantity in each spatial grid cell that has been adjusted by spatial weight and remapped to obtain the new adjusted housing quantity of each spatial grid cell; the adjusted user demand times counting method is consistent with the housing quantity counting method, that is, summarizing the user demand times in each spatial grid cell that has been adjusted by weight and remapped to form the adjusted user demand times of each spatial grid cell; after the statistics of all spatial grid cells are completed, the overall statistical result reflects the real supply and demand situation after the spatial weight adjustment, and the error caused by the regional scale overlap is corrected.

[0107] Based on the adjusted housing quantity and the adjusted user demand times, a corrected supply and demand hotspot area is generated;

[0108] The adjusted housing quantity and the adjusted user demand times of each spatial grid cell are re-calculated for the supply and demand intensity value, that is, the adjusted housing quantity and the adjusted user demand times are summed up, and the obtained result is used as the new supply and demand intensity value; the adjusted supply and demand intensity value of each spatial grid cell is compared with the preset supply and demand intensity threshold value again; when the adjusted supply and demand intensity value reaches or exceeds the preset supply and demand intensity threshold value, the spatial grid cell is re-identified as a corrected supply and demand hotspot cell; if the adjusted supply and demand intensity value is lower than the preset supply and demand intensity threshold value, the spatial grid cell is not identified as a corrected supply and demand hotspot cell; all spatial grid cells are judged, and finally the spatial grid cells that are re-identified as the corrected supply and demand hotspot cells are summarized according to the unified spatial position code to form the corrected supply and demand hotspot area; the data of the corrected supply and demand hotspot area eliminates the data error caused by the spatial scale overlap, and can more accurately express the real spatial distribution characteristics of housing supply and user demand.

[0109] S6: According to the influence range of the regional boundary overlap, combined with the user historical recommendation feedback data, the sensitivity of the user to the recommendation result is determined to generate user recommendation preference data, including:

[0110] The geographic location, click times, browsing time and close times of each recommendation record are extracted from the user historical recommendation feedback data, and are associated with the regional boundary overlap influence range according to the spatial code;

[0111] The user historical recommendation feedback data refers to a set of data recorded by the housing rental information dynamic recommendation system in actual application, which reflects the reaction of users to the housing recommendation information. Each recommendation record in the user historical recommendation feedback data usually includes multiple index information such as the geographical location of the recommended housing, the total number of times the user clicks the recommendation information, the length of time the user browses the recommended housing information, and the number of times the user closes the recommendation information. For example, a user is recommended a housing located near a commercial area by the housing rental information dynamic recommendation system, the user clicks the housing recommendation information multiple times, and after browsing for a certain length of time, the user finally closes the recommendation information. The housing rental information dynamic recommendation system records the accurate geographical location coordinates of the housing recommendation information, the number of times the user clicks, the total length of time the user browses the housing recommendation information, and the number of times the user closes the recommendation information. The user historical recommendation feedback data records the recommendation feedback behavior of all users in a similar manner, and each record records the above index information of each recommendation information.

[0112] The area boundary overlap influence range records all spatial grid cells that are identified as being affected by the area boundary scale overlap after spatial scale overlap analysis. The spatial grid cells are uniformly represented by spatial encoding. According to the spatial encoding associated area boundary overlap influence range, first, the geographical location coordinates of each recommendation record are extracted from the user historical recommendation feedback data, and the geographical location coordinates are standardized and recorded according to the unified spatial position encoding rule. The spatial position encoding of the recommendation record is matched and associated with the spatial position encoding of the spatial grid cells in the area boundary overlap influence range, i.e., it is determined whether the spatial position encoding of each recommendation record belongs to the area boundary overlap influence range. When the spatial position encoding of the recommendation record is the same as or completely matches the spatial position encoding of a spatial grid cell in the area boundary overlap influence range, it is determined that the recommendation record belongs to the records in the area boundary overlap influence range. When the spatial position encoding of the recommendation record fails to find a completely matched spatial position encoding of a spatial grid cell in the area boundary overlap influence range, it is determined that the recommendation record does not belong to the records in the area boundary overlap influence range. Through the above matching and association, the housing rental information dynamic recommendation system can extract all the recommendation records belonging to the area boundary overlap influence range from all the user historical recommendation feedback data.

[0113] For the recommendation records located in the area boundary overlap influence range, the number of clicks, the browsing time, and the number of closings of each user in the spatial grid cell are counted to generate a user feedback index matrix.

[0114] After the extraction of the recommendation records in the area boundary overlapping influence range is completed, the user feedback indicators of the recommendation records are counted according to each user, including the number of clicks, the browsing time and the number of closings. First, data processing is performed on each spatial grid cell in the area boundary overlapping influence range; for each spatial grid cell, all the recommendation records recorded in the grid cell are summarized; in each spatial grid cell, the recommendation records are classified and summarized according to the user identity, that is, the number of clicks, the browsing time and the number of closings of the user to the recommendation records are counted respectively for each user in the spatial grid cell; the counting method is as follows: the number of clicks of a single user to all the recommendation information in the spatial grid cell is added to obtain the total number of clicks of the user in the spatial grid cell; the time spent by the user in browsing the house recommendation information in the spatial grid cell each time is added to obtain the total browsing time of the user in the spatial grid cell; the number of times that the user closes all the recommendation information in the spatial grid cell is added to obtain the total number of closings of the user in the spatial grid cell; the above user feedback indicator counting operation is performed on all the spatial grid cells in the area boundary overlapping influence range to form an overall user feedback indicator matrix covering all the users and all the spatial grid cells. The user feedback indicator matrix is saved in the form of a matrix, the rows of the matrix mark different users, and the columns of the matrix mark different spatial grid cells; each element in the matrix records the number of clicks, the browsing time and the number of closings of the corresponding user in the corresponding spatial grid cell. The user feedback indicator matrix expresses the feedback behavior indicators of all the users in the area boundary overlapping influence range.

[0115] According to the user feedback indicator matrix, the sensitivity coefficient of each user in the spatial grid cell is calculated;

[0116] The sensitivity coefficient is an index for quantifying the sensitivity of each user to the recommended information. The calculation method is as follows: first, difference operation is performed on the click number and the close number recorded by each user in each spatial grid unit in the user feedback index matrix. The difference operation is the difference between the click number and the close number, and the difference data of the positive or negative reaction of the user to the recommended information is obtained; then, the obtained difference data is divided by the browsing time recorded by the user in the same spatial grid unit, and the result is the sensitivity coefficient of the user in the spatial grid unit. The sensitivity coefficient reflects the actual interest and attention of the user to the housing recommended information. For example, if the click number of the user in the spatial grid unit is much higher than the close number, and the browsing time is long, the sensitivity coefficient is high, indicating that the user has a positive reaction to the housing recommended information in the spatial grid unit; if the close number is higher than the click number or the browsing time is short, the sensitivity coefficient value is low or even negative, indicating that the user has a negative reaction or lack of interest to the recommended information in the spatial grid unit. According to the above method, the sensitivity coefficient of each user and each spatial grid unit recorded in the user feedback index matrix is calculated, and a sensitivity coefficient data set is formed, which records the actual interest degree of the user to the recommended information in the area boundary overlapping influence range.

[0117] The sensitivity coefficient and the spatial code are combined to generate user recommended preference data.

[0118] The sensitivity coefficient of each user in the spatial grid unit in the area boundary overlapping influence range is combined with the corresponding unified spatial position code to generate user recommended preference data. The user recommended preference data is a comprehensive data set combining spatial position and user interest sensitivity. First, the sensitivity coefficient of each user in each spatial grid unit is extracted from the sensitivity coefficient data set, and the unified spatial position code of the spatial grid unit corresponding to the sensitivity coefficient is extracted; the sensitivity coefficient of each user is combined with the corresponding unified spatial position code one by one to form a new user recommended preference data record; the user recommended preference data record includes user identification information, unified spatial position code and corresponding sensitivity coefficient; after sorting and summarizing all user recommended preference data records, a user recommended preference data set is formed. The user recommended preference data set expresses the sensitivity of each user to the housing recommended information in different spatial positions in the area boundary overlapping influence range, and can be used as a reference basis for dynamic recommendation of housing rental information, to ensure that the housing information recommended to the user is more in line with the actual interest needs and spatial position preferences of the user.

[0119] S7: According to the corrected supply and demand hot area and the user recommended preference data, the housing rental information is dynamically recommended to the user, including:

[0120] Based on the corrected supply and demand hot area, select the house source located in the supply and demand hot area from the supply side house source data to generate a recommended house source candidate set;

[0121] In order to generate the recommended house source candidate set, it is necessary to first determine the spatial range of each spatial grid unit in the corrected supply and demand hot area and all geographic location coordinate data contained. For each rental house information in the supply side house source data, it is judged whether the geographic location coordinate of the rental house is located in the spatial range of the spatial grid unit in the range of the corrected supply and demand hot area; the judgment method is: the geographic location coordinate in the rental house information record is matched with the spatial position coordinate set of each spatial grid unit in the corrected supply and demand hot area; if the geographic location coordinate of the rental house falls within the spatial range of a spatial grid unit, the information of the rental house is recorded as a component of the recommended house source candidate set; if the geographic location coordinate of the rental house does not fall within any spatial grid unit of the corrected supply and demand hot area, this rental house information is not recorded in the recommended house source candidate set; for example, when there are several sets of rental house information on the housing rental market, each set of house information records the accurate geographic location coordinate of itself, and each set of rental house information is accurately matched into the spatial grid unit in the corrected supply and demand hot area through the above spatial position judgment method; all rental house information sets successfully matched into the spatial grid unit in the corrected supply and demand hot area are collected to form a recommended house source candidate set; the recommended house source candidate set records the geographic location coordinate, house attribute, house area, rent level and matched spatial grid unit unified spatial position code of each rental house.

[0122] According to the user recommendation preference data, select the house source meeting the user recommendation preference from the recommended house source candidate set to form a personalized recommended house source set;

[0123] The user recommendation preference data records the sensitivity degree of each user to the house recommendation information of different spatial positions in the region boundary overlapping influence range, and each user recommendation preference data record includes user identity information, unified spatial position code and corresponding sensitivity coefficient, indicating the difference in interest degree and sensitivity degree of different users in different spatial grid units to the house recommendation information.

[0124] To form the personalized recommended housing set, first, the spatial grid unit that each user is interested in needs to be determined based on the user recommendation preference data, i.e., the spatial position with a higher user sensitivity coefficient; then, the spatial position code of the spatial grid unit matched by each piece of rental housing information in the recommended housing candidate set is unified, the spatial position code of the spatial grid unit matched by the housing information in the recommended housing candidate set is matched and compared with the spatial position code of the spatial position that the user is interested in in the user recommendation preference data; when the spatial position code of the housing information matched by the recommended housing candidate set is completely consistent with the spatial position code in the user recommendation preference data record, the housing rental information dynamic recommendation system judges that the housing information belongs to the spatial grid unit that the user is interested in, and then records the rental housing information as the housing that meets the user's recommendation preference, and includes it in the personalized recommended housing set; if the spatial position code of the rental housing information is not completely consistent with any spatial position code in the user recommendation preference data record, the rental housing information is not recorded in the personalized recommended housing set of the user; for example, the user recommendation preference data records that the user has a high sensitivity coefficient to a certain commercial circle spatial grid unit, then the housing rental information dynamic recommendation system selects all rental housing information with the spatial position code completely consistent with the unified spatial position code of the commercial circle spatial grid unit in the recommended housing candidate set, and records the housing information in the personalized recommended housing set of the user; the above process is repeated for each user, so that each user obtains a personalized recommended housing set consistent with his own recommendation preference; the personalized recommended housing set records the user identity, the geographic position coordinates of each housing, the housing attribute information, the housing area, the rent level and the spatial position code.

[0125] The personalized recommended housing set is sorted according to the spatial code of the spatial grid unit and the sensitivity coefficient in the user recommendation preference data, and the corresponding housing rental information is pushed to the user in the order from high to low;

[0126] In order to ensure that the user can obtain the most suitable for their own needs and the highest degree of interest in the housing information, the need for personalized recommendation of housing information in the set of sorting; the sorting of housing information based on the uniform spatial location coding of the spatial grid unit and the user sensitivity coefficient recorded in the user recommendation preference data, specifically: first, the spatial location coding of each set of housing information in the personalized recommendation of each user's housing source set and the spatial location coding in the user recommendation preference data are matched to determine the user sensitivity coefficient corresponding to the housing information; the user sensitivity coefficient recorded by each set of housing information is uniformly sorted, and the sorting processing is based on the user sensitivity coefficient size, from high to low in turn to arrange the order of housing information; when the user sensitivity coefficient is the same, the uniform spatial location coding is sorted again; for example, if there are several sets of housing information in the user's personalized recommendation of housing source set, the user sensitivity coefficient corresponding to each set of housing information is determined, the sensitivity coefficients of these housing information are compared and sorted, the housing information with higher sensitivity coefficient is arranged in the front row, and the housing information with relatively lower sensitivity coefficient is arranged in the back row; through the above sorting method, the housing information sorting list is finally formed in the personalized recommendation of housing source set, and each set of housing information in the list is determined according to the user sensitivity coefficient and the spatial location coding to determine the recommendation order.

[0127] After sorting, the corresponding housing rental information is pushed to each user according to the order from high to low; the housing rental information dynamic recommendation system extracts all attribute contents of each set of housing information in the sorting list in turn, including housing geographical location, area, rent level and spatial grid unit uniform spatial location coding and other information, and pushes the extracted housing information to the user terminal device in the form of recommended information suitable for user reading; the user terminal device displays the housing recommendation information for the user to browse or select; through the above pushing process, it is ensured that the recommended housing information received by the user completely meets the user's own interest and preference degree, and expresses the actual supply and demand situation of the housing rental market.

[0128] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center through a wired (for example, infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0129] Those skilled in the art can appreciate that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0130] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0131] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0132] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0133] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0134] The functions, if realized in the form of software functional modules and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0135] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0136] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A method for dynamically recommending housing rental information based on real-time supply and demand data, characterized in that: Includes the following steps: S1: Collect the original spatial boundary data of administrative districts, areas and business districts, and perform unified spatial coding to generate housing rental boundary data; S2: Based on housing rental boundary data, analyze the spatial distribution characteristics of supply-side housing data and demand-side user data at different spatial scales to identify supply and demand hotspots, specifically: Based on housing rental boundary data, spatial grids are established according to administrative district scale, area scale, and business district scale; The supply-side housing data is mapped by geographical location, and the number of housing units in each grid unit is counted to generate a supply density matrix. Map demand-side user data by geographic location, count the number of user demands in each grid unit, and generate a demand density matrix. The supply density matrix and the demand density matrix are superimposed with corresponding grid cells to calculate the supply and demand intensity value; The supply and demand intensity values ​​are compared with preset supply and demand intensity thresholds to identify supply and demand hotspot units, and these are then aggregated to form supply and demand hotspot areas. S3: Based on the housing rental boundary data, analyze the overlap relationship between different spatial scales to obtain spatial overlap relationship data, specifically: Using the spatial boundary data of administrative regions and the spatial boundary data of districts as the first pairing set, spatial overlay operation is performed to obtain the set of overlapping areas between administrative regions and districts. Using the spatial boundary data of the district and the spatial boundary data of the business district as the second pairing set, spatial overlay operation is performed to obtain the set of overlapping areas between the district and the business district; The sets of overlapping regions are merged according to spatial coding to form a multi-scale spatial overlap record table; In the multi-scale spatial overlap record table, a spatial overlap ratio value is generated for each spatial overlap record; Based on the multi-scale spatial overlap record table and the corresponding spatial overlap ratio value, spatial overlap relationship data is formed; S4: Based on the supply and demand hotspots and spatial overlap data, assess the degree of crossover of supply and demand data caused by regional boundary overlap, and determine the scope of influence of regional boundary overlap, specifically: For each spatial overlap record in the multi-scale spatial overlap record table, extract the number of supply-side housing units and the number of demand-side user requests for the corresponding grid unit from the supply and demand hotspot area according to the spatial code; For each spatially overlapping record, calculate cross-regional supply and demand cross indicators using the number of housing units on the supply side and the number of user demands on the demand side; The cross-regional supply and demand cross-indicators are compared with preset cross-indicators, and spatial overlap records where the cross-regional supply and demand cross-indicators are greater than the preset cross-indicators are selected as regional boundary overlap impact records. Summarize the spatial codes of all regional boundary overlap impact records to generate the regional boundary overlap impact range; S5: Based on the impact range of overlapping regional boundaries, spatial weights of supply-side housing data and demand-side user data are adjusted to generate adjusted supply and demand hotspot areas, specifically: Based on the spatial overlap ratio, the supply-side housing data and demand-side user data of each grid unit within the area of ​​regional boundary overlap are spatially weighted to generate adjusted demand-side user data and demand-side user data. The adjusted supply-side housing data and demand-side user data are remapped to the corresponding spatial grid cells, and the adjusted number of housing units and the adjusted number of user demands are recalculated for each spatial grid cell. Based on the adjusted number of available properties and the adjusted number of user requests, a revised supply and demand hotspot area is generated; S6: Based on the influence range of overlapping regional boundaries and combined with historical user recommendation feedback data, determine the user's sensitivity to the recommendation results and generate user recommendation preference data; S7: Based on the revised supply and demand hotspots and user recommendation preference data, dynamically recommend housing rental information to users.

2. The method for dynamically recommending housing rental information by combining real-time supply and demand data according to claim 1, characterized in that, S1, specifically: Collect original spatial boundary data of administrative regions, original spatial boundary data of districts, and original spatial boundary data of business districts; Spatial coordinate calibration and unified spatial location coding are performed on the original spatial boundary data of administrative regions, areas, and business districts to obtain the spatial boundary data of administrative regions, areas, and business districts. Based on the spatial boundary data of administrative regions, areas, and business districts after unified spatial location coding, data fusion processing is performed to generate housing rental boundary data.

3. The method for dynamically recommending housing rental information by combining real-time supply and demand data according to claim 2, characterized in that, S6, specifically: Extract the geographic location, number of clicks, browsing duration, and number of closes for each recommendation record from the user's historical recommendation feedback data, and associate the overlapping influence range of the region boundaries according to spatial coding; For recommended records located within the overlapping influence range of regional boundaries, the number of clicks, browsing duration, and closing times for each user within the spatial grid cell are counted to generate a user feedback indicator matrix; Based on the user feedback indicator matrix, calculate the sensitivity coefficient of each user within the spatial grid cell; By combining sensitivity coefficients with spatial coding, user recommendation preference data can be generated.

4. The method for dynamically recommending housing rental information by combining real-time supply and demand data according to claim 3, characterized in that, S7, specifically: Based on the revised supply and demand hotspot areas, housing units located within these hotspot areas are selected from the supply-side housing data to generate a candidate set of recommended housing units; Based on user recommendation preference data, properties that meet the user's recommendation preferences are selected from the candidate set of recommended properties to form a personalized recommended property set. The personalized recommended property set is sorted according to the spatial code of the spatial grid unit and the sensitivity coefficient in the user recommendation preference data, and the corresponding housing rental information is pushed to the user in descending order of sorting.

Images