A method, apparatus, electronic device, and storage medium for confirming the address of a bidding project.

Abstract

This invention relates to a method, apparatus, electronic device, and storage medium for confirming the address of a bidding project. The method includes decomposing the bidding project address text to obtain fuzzy modifiers in the decomposition results; thereby identifying the fuzzy modifiers in the bidding project address text; obtaining a set of candidate coordinates from the decomposition results through multiple geocoding service software, improving the coverage and accuracy of address resolution; quantifying the fuzzy modifiers and the set of candidate coordinates based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range, quantifying the spatial range of fuzzy modifiers such as "nearby," and solving the problem of insufficient fuzziness processing; and fusing the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates. By fusing multi-source data, the lack of mathematical optimization is solved, enabling the method to adapt to complex scenarios. Based on the optimal latitude and longitude coordinates, a more accurate bidding project address can be obtained, improving the visualization effect of the project location.

Description

Technical Field

[0001] This invention relates to the field of geographic information system technology, and in particular to a method, apparatus, electronic device, and storage medium for confirming the address of a marker project. Background Technology

[0002] In bidding systems, project addresses are usually provided in text form. The address descriptions are often unstructured, including administrative divisions, street names, house numbers, and vague modifiers (such as "nearby"), making it difficult to directly convert them into precise geographical coordinates (latitude and longitude).

[0003] Existing geocoding technologies primarily rely on address resolution services (such as Baidu Maps API and Gaode Maps API) to convert text addresses into coordinates by matching them against a standard address database. However, existing geocoding technologies suffer from the following problems: 1. Insufficient handling of ambiguity: Traditional methods struggle to quantify the spatial range of vague descriptions such as "nearby" or "approximately," leading to decreased positioning accuracy. 2. Dependence on a single data source: Relying on a single API may fail due to incomplete data coverage, especially in remote areas or non-standard address scenarios. 3. Lack of mathematical optimization: Existing methods are mostly rule-based matching, lacking mathematical model support and unable to adapt to complex scenarios. These problems make it difficult to calculate accurate project location coordinates from unstructured address information, affecting subsequent spatial analysis and visualization.

[0004] Therefore, there is an urgent need to propose a method, device, electronic device, and storage medium for confirming the address of a bidding project, in order to solve the technical problem that existing methods are unable to calculate the accurate project location coordinates from unstructured bidding addresses, resulting in unsatisfactory visualization of project locations. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, apparatus, electronic device and storage medium for confirming the address of a tender project, in order to solve the technical problem that existing methods are unable to calculate the accurate project location coordinates from unstructured tender addresses, resulting in unsatisfactory project location visualization.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a method for confirming the address of a bidding project, comprising: Obtain the bidding project address text, decompose the bidding project address text to obtain the decomposition result; the decomposition result includes fuzzy modifiers; the fuzzy modifiers are words that express spatial uncertainty; The decomposition results are identified in multiple geocoding service software programs to obtain a set of candidate coordinates for each geocoding service software program. The fuzzy modifiers and the candidate coordinate set are quantized based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range; The candidate coordinate set and the fuzzy range are fused to obtain the optimal latitude and longitude coordinates, which are then used as the address of the bidding project.

[0007] In one possible implementation, the decomposition of the tender project address text to obtain the decomposition result includes: Set the decomposition type; the decomposition type includes administrative division, specific address, and the fuzzy modifier; The address text of the tag project is decomposed using regular expressions and word segmentation tools to obtain the decomposition result corresponding to each decomposition type.

[0008] In one possible implementation, the quantization of the fuzzy modifier and the candidate coordinate set based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range includes: The address scenario is determined based on the address text of the bidding project; Based on the address scenario, set the range radius of the fuzzy modifier; The reference coordinates are obtained based on the spatial distribution center point of the candidate coordinate set; The range radius and the reference coordinates are quantized according to the two-dimensional Gaussian probability distribution model to obtain the position probability density; An initial range is obtained based on the range radius and the reference coordinates, and the geographical area boundary in the initial range where the location probability density is greater than a preset threshold is identified as a fuzzy range.

[0009] In one possible implementation, the step of fusing the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates includes: Based on the candidate coordinate set of each geocoding service software, the corresponding confidence level is obtained; Based on the address scenario, set constraint weights; The confidence level, constraint weight, and location probability density of each geocoding service software are optimized based on the optimization formula to obtain the optimal latitude and longitude coordinates.

[0010] In one possible implementation, using the optimal latitude and longitude coordinates as the address of the tagging project includes: A confidence score is obtained based on all confidence scores and the number of candidates for the multiple geocoding service software. When the confidence score meets the conditions, the optimal latitude and longitude coordinates are determined as the address of the bidding project.

[0011] In one possible implementation, the two-dimensional Gaussian probability distribution model is:

[0012] In the formula, For location probability density, As the reference coordinates, The coordinates are in the candidate coordinate set. For the first i The first service j The coordinate values ​​of each candidate coordinate. For scale parameters, , The radius is the range.

[0013] In one possible implementation, the optimization formula is:

[0014]

[0015] In the formula, For optimal latitude and longitude coordinates, As the reference coordinates, For the first i The number of candidates for geocoding service software. The total number of geocoding service software. For the first i The first service j The confidence level of each candidate coordinate. For the first i The first service j The coordinate values ​​of each candidate coordinate. To constrain the weights, For scale parameters, .

[0016] Secondly, the present invention also provides a device for confirming the address of a bidding project, comprising: The text decomposition module is used to obtain the bidding project address text, decompose the bidding project address text, and obtain the decomposition result; the decomposition result includes fuzzy modifiers; the fuzzy modifiers are words that express spatial uncertainty; The coordinate recognition module is used to recognize the decomposition results in multiple geocoding service software to obtain a candidate coordinate set corresponding to each geocoding service software. The range quantization module is used to quantize the fuzzy modifier and the candidate coordinate set based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range; The data fusion module is used to fuse the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates, and use the optimal latitude and longitude coordinates as the address of the bidding project.

[0017] Thirdly, embodiments of the present invention disclose an electronic device, including: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the various steps of the above-described method embodiments for confirming the address of a bidding project.

[0018] Fourthly, embodiments of the present invention disclose a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the various steps of the above-described method for confirming the address of a bidding project.

[0019] The beneficial effects of this invention are as follows: The decomposition of the bidding project address text yields decomposition results, including fuzzy modifiers. This allows for the identification of fuzzy modifiers in the bidding project address text. Furthermore, the decomposition results can be identified using multiple geocoding service software programs to obtain candidate coordinate sets for each software. This approach, by using multiple geocoding service software programs to acquire candidate coordinates, improves the coverage and accuracy of address resolution, solving the problem of dependence on a single data source. Based on a two-dimensional Gaussian probability distribution model, the fuzzy modifiers and candidate coordinate sets are quantified to obtain a fuzzy range, quantifying the spatial range of fuzzy modifiers such as "nearby," thus addressing the problem of insufficient fuzziness processing. Data fusion of the candidate coordinate set and fuzzy range yields optimal latitude and longitude coordinates. Multi-source data fusion solves the problem of insufficient mathematical optimization, enabling the method to adapt to complex scenarios. Based on the optimal latitude and longitude coordinates, a more accurate bidding project address can be obtained, improving the visualization effect of the project location. Attached Figure Description

[0020] Figure 1 A schematic flowchart of an embodiment of the method for confirming the address of a bidding project provided by the present invention; Figure 2 For the present invention Figure 1 A schematic flowchart of an embodiment of step S103; Figure 3 For the present invention Figure 1 A schematic flowchart of an embodiment of step S104; Figure 4 A schematic diagram of an embodiment of the address confirmation device for bidding projects provided by the present invention; Figure 5 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0021] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0022] In existing technologies, project addresses in bidding systems are typically presented as unstructured text, often containing administrative divisions, street names, and vague modifiers. Traditional geocoding techniques rely on a single address resolution service, which cannot effectively handle ambiguous spatial descriptions, leading to insufficient coordinate transformation accuracy. For example, when address text contains vague modifiers such as "nearby" or "approximately," existing methods struggle to quantify its spatial range, and address matching in remote areas is prone to failure due to incomplete data coverage. This situation often results in coordinate offsets or positioning failures when precise location of a project is required for visualization.

[0023] To address the aforementioned issues, the quantification of fuzzy modifiers must first be tackled. Traditional rule-based matching methods cannot adapt to dynamic range changes; therefore, a probabilistic model is introduced for spatial distribution modeling. To address the insufficient coverage of a single data source, data reliability is improved by integrating results from multiple geocoding service software. Finally, a mathematical optimization model needs to be designed to fuse multi-source heterogeneous data, rather than simply using a weighted average. This approach breaks through the single-matching mode of traditional geocoding technology, combining fuzzy semantic parsing with multi-source data fusion.

[0024] like Figure 1 As shown in the figure, a specific embodiment of the present invention discloses a method for confirming the address of a bidding project, comprising: S101. Obtain the bidding project address text, decompose the bidding project address text, and obtain the decomposition result; the decomposition result includes fuzzy modifiers; fuzzy modifiers are words that express spatial uncertainty.

[0025] The project address text in the bidding information refers to unstructured address information containing administrative divisions, road names, and vague spatial descriptions. It can exist in the form of natural language text, such as "near the east side of XX Road, XX City, XX Province". Vague modifiers refer to words that express spatial uncertainty. They can be identified using semantic analysis tools, such as words like "nearby", "around", and "approximately", which are used to describe the spatial distribution range of coordinate points.

[0026] S102. The decomposition results are identified in multiple geocoding service software to obtain the candidate coordinate set corresponding to each geocoding service software.

[0027] Geocoding service software refers to application programming interfaces (APIs) that provide address-to-coordinate conversion functions. Specifically, it can utilize third-party services such as Baidu Maps API and Gaode Maps API, improving address resolution reliability through multi-source data complementarity. For the input address in the decomposition results... A Let the first i The candidate coordinates returned by each service are shown in formula (1): (1) In the formula, and They represent latitude and longitude respectively. No. i The number of candidates for each service.

[0028] Assume there is a total n One service, total candidate coordinate set As shown in formula (2): (2) S103. Based on the two-dimensional Gaussian probability distribution model, the fuzzy modifiers and candidate coordinate sets are quantized to obtain the fuzzy range.

[0029] Among them, the two-dimensional Gaussian probability distribution model refers to the mathematical model that describes the probability density of two-dimensional space. Specifically, it can be expressed by parameterization using the covariance matrix and the mean vector, and is used to convert fuzzy modifiers into a computable spatial probability distribution.

[0030] S104. Perform data fusion on the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates, and use the optimal latitude and longitude coordinates as the address of the billing project.

[0031] Data fusion refers to the process of optimizing and calculating multi-source coordinate data. Specifically, it can use maximum likelihood estimation or Bayesian inference methods to generate optimal latitude and longitude coordinates by combining probability distribution characteristics.

[0032] Specifically, after obtaining address text containing fuzzy modifiers, structured information such as administrative divisions and road names is extracted through semantic decomposition. This structured information is then input into multiple geocoding service software programs to obtain candidate coordinate sets returned by each service. For example, inputting "east side of XX Road" may yield ten candidate coordinates returned by three services. The spatial expansion radius is determined based on the fuzzy modifiers; for example, "nearby" corresponds to a 500-meter range. A two-dimensional Gaussian distribution model is constructed using the mean of the candidate coordinates as a reference point. By calculating the distribution weight of each candidate coordinate in the probability density space and combining it with the confidence level of multi-source data for weighted optimization, the optimal latitude and longitude coordinates located in the region of highest probability density are finally generated and used as the address of the tag project. This process effectively integrates semantic parsing, multi-source data verification, and mathematical optimization, ensuring that the output coordinates conform to both the text description and spatial distribution characteristics.

[0033] Compared to existing technologies, traditional methods rely on a single geocoding service software to directly output coordinates, which cannot handle fuzzy descriptions and lacks data verification mechanisms. This invention, through multi-service collaborative identification, effectively avoids the problems of missing or incorrectly parsed data from a single service. Simultaneously, it employs a probabilistic model to quantify the fuzzy spatial range, converting natural language descriptions into calculable mathematical parameters, thus solving the problem of poor adaptability in traditional rule-based matching methods. Furthermore, the optimized calculation process comprehensively considers the confidence distribution of multi-source data, providing higher accuracy compared to simple coordinate averaging methods.

[0034] Through the above technical solution, this application can accurately parse unstructured address text containing fuzzy modifiers, improve data reliability through collaborative verification using multi-source geocoding service software, quantify the range of spatial uncertainty using a probabilistic model, and finally output optimal coordinates that conform to the actual location distribution characteristics. This effectively solves the accuracy problems of traditional methods in fuzzy address processing and remote area positioning, and provides accurate geographic coordinate data support for the visualization of project locations.

[0035] In some embodiments of the present invention, the text of the bidding project address is decomposed to obtain the decomposition result, including: Set the decomposition type; decomposition types include administrative division, specific address, and vague modifiers; The address text of the bidding project is decomposed using regular expressions and word segmentation tools to obtain the decomposition results corresponding to each decomposition type.

[0036] In a specific embodiment of the present invention, the decomposition type refers to a predefined category of address elements, which can be implemented using three classification methods: administrative division, specific address, and fuzzy modifiers. This classification can structurally process different semantic units in non-standardized address text. Regular expressions refer to matching fixed-format administrative division information in address text using predefined pattern strings. This can be implemented using Python's regular expression operation module, effectively recognizing regular expressions such as "XX province XX city". The word segmentation tool refers to a dictionary-based text segmentation component, which can be implemented using the Jieba word segmentation tool. By loading a custom address dictionary, it segments continuous text, thereby extracting unstructured elements such as specific addresses and fuzzy modifiers.

[0037] Specifically, a structured processing framework for address text is established by setting three decomposition types: administrative divisions, specific addresses, and vague modifiers. Regular expressions are used to precisely match administrative division information with fixed formats, such as matching administrative division levels containing keywords like "province," "city," and "district." Simultaneously, a word segmentation tool is used to intelligently segment the remaining text. By loading an address dictionary containing specialized terms such as street names and landmarks, specific address elements and vague modifiers are accurately identified. The combined application of these two techniques achieves effective synergy between rule matching and semantic segmentation, transforming unstructured address text into a structured data set containing explicit semantic tags.

[0038] Compared to existing technologies, traditional methods typically lack a clearly defined decomposition level and often employ single regular expression matching or simple word segmentation to process address text, resulting in incomplete address element recognition. This invention, by establishing a three-level decomposition system and combining the complementary advantages of regular expressions and word segmentation tools, significantly improves the completeness and accuracy of address element extraction, particularly effectively separating out ambiguous modifiers that affect positioning accuracy.

[0039] Through the above technical solution, this application achieves refined decomposition of the address text, ensuring independent processing of various address elements in the subsequent geocoding process. In particular, by separating fuzzy modifiers, it provides key input parameters for the subsequent establishment of a two-dimensional Gaussian probability distribution model, solving the problem of quantitative modeling difficulties caused by mixed address elements in traditional methods, and improving the accuracy of coordinate positioning from the source of data processing.

[0040] In some embodiments of the present invention, such as Figure 2 As shown, step S103 includes: S201. Determine the address scenario based on the address text of the bidding project.

[0041] The address scenario refers to the administrative level or functional type to which the project address belongs. Specifically, it can be achieved by recognizing the administrative division level in the address text through natural language processing technology, such as distinguishing between provincial, municipal, or community-level scenarios.

[0042] S202. Set the radius of the range of fuzzy modifiers according to the address scenario.

[0043] The range radius refers to the spatial coverage distance corresponding to the fuzzy modifier. Specifically, the radius value can be preset according to different address scenarios. For example, it can be set to 500 meters in urban scenarios and 500 meters in suburban scenarios.

[0044] S203. Obtain the reference coordinates based on the spatial distribution center point of the candidate coordinate set.

[0045] The reference coordinates refer to the spatial distribution center point of the candidate coordinate set. Specifically, they can be calculated using the arithmetic mean or geometric median of the candidate coordinates. The specific area confirmation process can be set according to the actual situation, and this embodiment of the invention does not impose any limitations on it.

[0046] S204. Quantize the range radius and reference coordinates according to the two-dimensional Gaussian probability distribution model to obtain the position probability density.

[0047] Among them, the two-dimensional Gaussian probability distribution model is a mathematical model used to describe the probability density of spatial location. Specifically, it can be achieved by setting a scale parameter to adjust the probability decay rate, which is positively correlated with the radius of the range.

[0048] S205. Obtain the initial range based on the range radius and reference coordinates, and identify the geographical area boundaries in the initial range where the location probability density is greater than a preset threshold as fuzzy ranges.

[0049] In a specific embodiment of the present invention, when vague modifiers such as "nearby" or "surrounding area" appear in the address text, the address scene category is first determined through semantic analysis. For example, for the address "near XXX in XX Village, XX District, XX City", it is identified as belonging to the urban core area scene, and a preset range radius of 500 meters is automatically matched. Then, the coordinate point with the highest frequency of occurrence is selected from the candidate coordinates returned by multiple geocoding service software as the reference coordinate, or the geometric center of the coordinate set is calculated through a clustering algorithm. Subsequently, the range radius is used as the standard deviation parameter to input into the two-dimensional Gaussian probability distribution model, and the probability density peak point is used as the reference coordinate to generate a probability distribution surface that decays exponentially with increasing distance. The two-dimensional Gaussian probability distribution model is shown in formula (3): (3) In the formula, For location probability density, As the reference coordinates, The coordinates are in the candidate coordinate set. For the first i The first service j The coordinate values ​​of each candidate coordinate. For scale parameters, , The radius is the range.

[0050] Draw a circle with the reference coordinates as the center and the radius as the range radius to obtain the initial range. Then, the geographical area boundary with the position probability density greater than the preset threshold in the initial range is output as the fuzzy range, which is presented as an irregular ellipse centered on the reference point.

[0051] Compared to existing technologies, traditional geocoding techniques only use a fixed-distance buffer for handling ambiguous modifiers, such as a uniform 500-meter range, which cannot distinguish the spatial characteristics of different scenarios. This solution, however, dynamically adjusts the range radius by classifying address scenarios and establishes a probabilistic model based on the distribution characteristics of candidate coordinate sets. This ensures that the quantification process of ambiguous ranges considers both semantic features and spatial data distribution patterns, effectively solving the problem of insufficient adaptability of the single buffer method in different scenarios such as mountainous areas and urban areas.

[0052] Through the above technical solution, this application can dynamically generate a probabilistic fuzzy range that matches the scene based on the actual semantic environment of the address text and the spatial distribution characteristics of the candidate coordinates, significantly improving the spatial quantification accuracy of fuzzy modifiers. This technical solution combines semantic logic and spatial statistics in the address parsing process, ensuring that the final generated fuzzy range accurately reflects the uncertainty in the address description and adapts to the spatial expansion characteristics of different geographical environments.

[0053] In some embodiments of the present invention, such as Figure 3 As shown, step S104 includes: S301. Based on the candidate coordinate set of each geocoding service software, obtain the corresponding confidence level.

[0054] Among them, confidence level refers to the reliability of the candidate coordinate set returned by geocoding service software. Specifically, it can be calculated through historical matching accuracy or the accuracy index provided by the service provider, and is used to measure the credibility of the results returned by different services.

[0055] S302. Set constraint weights according to the address scenario.

[0056] Among them, constraint weight refers to dynamically adjusting the importance of different geocoding service software in the optimization process according to the address scenario. Specifically, it can be implemented by using a preset scenario-weight mapping table. For example, high-precision map services are given higher weight in urban central areas, while open map services are given higher weight in rural areas.

[0057] S303. Based on the optimization formula, the confidence level, constraint weight, and location probability density of each geocoding service software are optimized to obtain the optimal latitude and longitude coordinates.

[0058] The optimization formula is a mathematical expression that weights confidence, weights, and probability density. It can be implemented using a multivariate linear combination or probability product, and is used to generate the optimal solution by integrating multi-source data. The goal is to find the optimal latitude and longitude coordinates. This minimizes the weighted sum of squared distances while considering the probability constraints of the fuzzy range. The optimization problem is defined as shown in formula (4): (4) In the formula, For optimal latitude and longitude coordinates, As the reference coordinates, For the first i The number of candidates for geocoding service software. The total number of geocoding service software. For the first i The first service j The confidence level of each candidate coordinate. For the first i The first service j The coordinate values ​​of each candidate coordinate. To constrain the weights, For scale parameters, , The radius is the range.

[0059] Let the objective function be as shown in formula (5): (5) right and Calculate the partial derivatives and set them to zero. As shown in formulas (6) and (7): (6) (7) The optimized formulas are shown in formulas (8) and (9): (8) (9) Specifically, during the data fusion process, the confidence score is first assigned to the candidate coordinates returned by each geocoding service software. For example, coordinates returned by the Baidu Maps API can be assigned a confidence score of 0.9 based on its officially published positioning accuracy level. Then, preset constraint weight parameters are loaded according to the current address scenario. For example, when the address text is identified to contain "industrial park," the weight allocation scheme corresponding to the industrial park scene is automatically applied. Finally, the confidence score, weight, and location probability density calculated by the two-dimensional Gaussian model are substituted into the optimization formula, and the final coordinates are determined through weighted calculation. For example, for a certain industrial park address, the calculation will combine the candidate coordinate confidence score of 0.85 from Amap, the preset weight of 0.7, and its Gaussian probability value of 0.92 within the fuzzy range to generate the optimal latitude and longitude.

[0060] Compared to existing technologies, current geocoding methods rely solely on coordinate matching results from a single data source, failing to consider the impact of confidence differences among service providers and scene characteristics on weight allocation. Our proposed method, however, establishes a mathematical optimization model to dynamically integrate results from multiple geocoding service software. By introducing a confidence assessment mechanism and scene-based weight adjustment during data processing, it effectively solves the positioning error problem caused by the bias of a single data source in traditional methods.

[0061] Through the above technical solution, this application can achieve adaptive fusion of multi-source geographic data. In the presence of fuzzy modifiers and multiple candidate coordinates, the optimal positioning result is generated through quantitative evaluation and scenario-based parameter configuration, which significantly improves the accuracy and reliability of the address coordinates of the bidding project and provides accurate coordinate input for subsequent geographic information visualization.

[0062] In some embodiments of the present invention, step S104 includes: The confidence score is obtained based on all confidence levels and the number of candidates from multiple geocoding service software.

[0063] Among them, confidence level refers to the degree of trust that each geocoding service software has in the candidate coordinates. Specifically, it can be calculated using the matching score or historical accuracy returned by the service, and is used to measure the credibility of the coordinates. Candidate number refers to the total number of candidate coordinates returned by each geocoding service software. Specifically, it can be calculated by counting the number of coordinate entries in the service response results, reflecting the coverage and diversity of the data source. Confidence score refers to the evaluation index that combines confidence level and candidate number. Specifically, it can be calculated by weighted summation or multiplication, for example, by combining confidence level and candidate number in proportion, to quantify the overall credibility of the coordinate results. The calculation of confidence score is shown in formula (10): (10) When the confidence score meets the conditions, the optimal latitude and longitude coordinates are determined as the address of the billing project.

[0064] Among them, the condition refers to the preset threshold or logical judgment rule, such as setting the coordinates with a score higher than 0.7 or ranked in the top 10% as valid results, which is used to filter geographical coordinates that meet the reliability standards.

[0065] Specifically, after obtaining the optimal latitude and longitude coordinates through data fusion, their effectiveness needs further verification. This is done by counting the number of candidate coordinates from all geocoding service software and combining the confidence scores of each candidate coordinate to calculate a comprehensive confidence score. For example, if one geocoding service software returns 5 candidate coordinates with an average confidence score of 0.8, and another service returns 3 candidate coordinates with an average confidence score of 0.9, both can be normalized and weighted to calculate the total score. When the score reaches a preset threshold, such as a total score exceeding 0.75, the optimal latitude and longitude coordinates are deemed valid and used as the final address of the tagging project. If the threshold is not reached, a re-resolution or manual intervention process is triggered to prevent erroneous coordinates from being adopted.

[0066] Compared to existing technologies, traditional methods rely solely on the confidence level of a single geocoding service software or a simple majority voting mechanism, failing to consider the impact of differences in the number of results returned by different services on reliability. For example, when a service returns a large number of low-quality candidate coordinates, its numerical advantage may lead to misjudgment. Our proposed solution, however, employs a dual verification mechanism of confidence level and candidate quantity, dynamically adjusting evaluation weights to effectively avoid positioning errors caused by uneven data source coverage or noise interference.

[0067] Through the above technical solution, this application can establish a dynamic verification standard based on the statistical characteristics of multi-source data, solve the problem of misjudgment caused by a single evaluation index in traditional methods, and ensure that the final coordinates meet the requirements in terms of confidence and data coverage completeness, thereby improving the accuracy of the bidding project address and the robustness of the system.

[0068] The address text of the bidding project in this embodiment of the invention can be "XX City, XX District, XX Village, XXX (nearby)", and the decomposition results can be "XX City, XX District, XX Village", "XXX", and "nearby". Baidu Maps: (39.9691, 116.4892) (confidence level 0.95). Gaode Maps: (39.9690, 116.4895) (confidence level 0.90). Fuzzy range: set r =500 m, =166.7 m, reference coordinates (39.9691, 116.4892). The values ​​are 0.95 and 0.90, respectively. λ =0.1. Substituting into the formula, we get... =39.9691, =39.9691. Output: The optimal latitude and longitude coordinates are (39.9691, 116.4892), with a confidence score of 0.90. If the condition is greater than 50%, then the confidence score meets the condition, and the optimal latitude and longitude coordinates are determined to be (39.9691, 116.4892) as the project coordinates.

[0069] This invention addresses the problem of parsing unstructured addresses through address text preprocessing, decomposing addresses into processable components. Multi-source geocoding improves the coverage and accuracy of address resolution by acquiring candidate coordinates from multiple services, resolving the issue of reliance on a single data source. Fuzzy range modeling quantifies the spatial range of vague modifiers such as "nearby" by constructing a two-dimensional Gaussian probability distribution model, addressing the problem of insufficient handling of fuzziness. The multi-source data fusion and optimization steps introduce a mathematical optimization model, using weighted least squares and Bayesian inference to calculate the optimal latitude and longitude coordinates by fusing multi-source data. This addresses the lack of mathematical optimization, enabling the method to adapt to complex scenarios.

[0070] To better implement the bidding project address confirmation method in the embodiments of the present invention, the embodiments of the present invention also provide a bidding project address confirmation device, such as... Figure 4 As shown, the bidding project address confirmation device 400 includes: The text decomposition module 401 is used to obtain the bidding project address text, decompose the bidding project address text, and obtain the decomposition result; the decomposition result includes fuzzy modifiers; fuzzy modifiers are words that express spatial uncertainty; The coordinate recognition module 402 is used to recognize the decomposition results in multiple geocoding service software to obtain a set of candidate coordinates for each geocoding service software. The range quantization module 403 is used to quantize the fuzzy modifiers and candidate coordinate sets based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range; The data fusion module 404 is used to fuse the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates, and use the optimal latitude and longitude coordinates as the address of the billing project.

[0071] The bidding project address confirmation device 400 provided in the above embodiments can realize the technical solutions described in the above bidding project address confirmation method embodiments. The specific implementation principles of each module or unit can be found in the corresponding content in the above bidding project address confirmation method embodiments, and will not be repeated here.

[0072] like Figure 5 As shown, the present invention also provides an electronic device 500. The electronic device 500 includes a processor 501, a memory 502, and a display 503. Figure 5 Only some components of the electronic device 500 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0073] In some embodiments, memory 502 may be an internal storage unit of electronic device 500, such as a hard disk or memory of electronic device 500. In other embodiments, memory 502 may also be an external storage device of electronic device 500, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 500.

[0074] Furthermore, the memory 502 may include both internal storage units of the electronic device 500 and external storage devices. The memory 502 is used to store application software and various types of data installed on the electronic device 500.

[0075] In some embodiments, processor 501 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in memory 502 or process data, such as the bidding project address confirmation method in this invention.

[0076] In some embodiments, display 503 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 503 is used to display information from electronic device 500 and to display a visual user interface. Components 501-503 of electronic device 500 communicate with each other via a system bus.

[0077] In some embodiments of the present invention, when the processor 501 executes the flag item address confirmation program in the memory 502, the following steps can be implemented: Obtain the project address text of the bidding information, decompose the project address text to obtain the decomposition result; the decomposition result includes fuzzy modifiers; fuzzy modifiers are words that express spatial uncertainty; The decomposition results are identified in multiple geocoding service software programs to obtain a set of candidate coordinates for each geocoding service software program. The fuzzy range is obtained by quantizing the fuzzy modifiers and candidate coordinate sets based on a two-dimensional Gaussian probability distribution model. Data fusion is performed on the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates, which are then used as the address of the billing project.

[0078] It should be understood that when the processor 501 executes the flag item address confirmation program in the memory 502, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0079] Furthermore, the embodiments of the present invention do not specifically limit the type of electronic device 500 mentioned. Electronic device 500 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, electronic device 500 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0080] Accordingly, this application also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions of the tag address confirmation method provided in the above-described method embodiments.

[0081] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0082] The foregoing has provided a detailed description of the method, apparatus, electronic device, and storage medium for confirming the address of a bidding project provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for confirming the address of a bidding project, characterized in that, include: Obtain the bidding project address text, decompose the bidding project address text, and obtain the decomposition result; The decomposition results include fuzzy modifiers; The fuzzy modifiers are words that express spatial uncertainty; The decomposition results are identified in multiple geocoding service software programs to obtain a set of candidate coordinates for each geocoding service software program. The fuzzy modifiers and the candidate coordinate set are quantized based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range; The candidate coordinate set and the fuzzy range are fused to obtain the optimal latitude and longitude coordinates, and the optimal latitude and longitude coordinates are used as the address of the bidding project. The quantization of the fuzzy modifier and the candidate coordinate set based on the two-dimensional Gaussian probability distribution model yields the fuzzy range, including: The address scenario is determined based on the address text of the bidding project; Based on the address scenario, set the range radius of the fuzzy modifier; The reference coordinates are obtained based on the spatial distribution center point of the candidate coordinate set; The range radius and the reference coordinates are quantized according to the two-dimensional Gaussian probability distribution model to obtain the position probability density; An initial range is obtained based on the range radius and the reference coordinates, and the geographical area boundary in the initial range where the location probability density is greater than a preset threshold is identified as a fuzzy range.

2. The method for confirming the address of a bidding project according to claim 1, characterized in that, The process of decomposing the address text of the bidding project to obtain the decomposition result includes: Set the decomposition type; the decomposition type includes administrative division, specific address, and the fuzzy modifier; The address text of the tag project is decomposed using regular expressions and word segmentation tools to obtain the decomposition result corresponding to each decomposition type.

3. The method for confirming the address of a bidding project according to claim 1, characterized in that, The step of fusing the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates includes: Based on the candidate coordinate set of each geocoding service software, the corresponding confidence level is obtained; Based on the address scenario, set constraint weights; The confidence level, constraint weight, and location probability density of each geocoding service software are optimized based on the optimization formula to obtain the optimal latitude and longitude coordinates.

4. The method for confirming the address of a bidding project according to claim 3, characterized in that, The step of using the optimal latitude and longitude coordinates as the address of the bidding project includes: A confidence score is obtained based on all confidence scores and the number of candidates for the multiple geocoding service software. When the confidence score meets the conditions, the optimal latitude and longitude coordinates are determined as the address of the bidding project.

5. The method for confirming the address of a bidding project according to claim 1, characterized in that, The two-dimensional Gaussian probability distribution model is as follows: In the formula, For location probability density, As the reference coordinates, The coordinates are in the candidate coordinate set. For the first i The first service j The coordinate values ​​of each candidate coordinate. For scale parameters, , The radius is the range.

6. The method for confirming the address of a bidding project according to claim 3, characterized in that, The optimization formula is: In the formula, For optimal latitude and longitude coordinates, As the reference coordinates, For the first i The number of candidates for geocoding service software. The total number of geocoding service software. For the first i The first service j The confidence level of each candidate coordinate. For the first i The first service j The coordinate values ​​of each candidate coordinate. To constrain the weights, For scale parameters, , The radius is the range.

7. A device for confirming the address of a bidding project, characterized in that, include: The text decomposition module is used to obtain the bidding project address text, decompose the bidding project address text, and obtain the decomposition result. The decomposition results include fuzzy modifiers; The fuzzy modifiers are words that express spatial uncertainty; The coordinate recognition module is used to recognize the decomposition results in multiple geocoding service software to obtain a candidate coordinate set corresponding to each geocoding service software. The range quantization module is used to quantize the fuzzy modifier and the candidate coordinate set based on a two-dimensional Gaussian probability distribution model to obtain the fuzzy range; The data fusion module is used to fuse the candidate coordinate set and the fuzzy range to obtain the optimal latitude and longitude coordinates, and use the optimal latitude and longitude coordinates as the address of the bidding project; The quantization of the fuzzy modifier and the candidate coordinate set based on the two-dimensional Gaussian probability distribution model yields the fuzzy range, including: The address scenario is determined based on the address text of the bidding project; Based on the address scenario, set the range radius of the fuzzy modifier; The reference coordinates are obtained based on the spatial distribution center point of the candidate coordinate set; The range radius and the reference coordinates are quantized according to the two-dimensional Gaussian probability distribution model to obtain the position probability density; An initial range is obtained based on the range radius and the reference coordinates, and the geographical area boundary in the initial range where the location probability density is greater than a preset threshold is identified as a fuzzy range.

8. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for confirming the address of a bidding project as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the method for confirming the address of a bidding project as described in any one of claims 1-6.

Images