WO2018177316A1 - 信息识别方法、计算设备及存储介质 - Google Patents

信息识别方法、计算设备及存储介质 Download PDF

Info

Publication number
WO2018177316A1
WO2018177316A1 PCT/CN2018/080822 CN2018080822W WO2018177316A1 WO 2018177316 A1 WO2018177316 A1 WO 2018177316A1 CN 2018080822 W CN2018080822 W CN 2018080822W WO 2018177316 A1 WO2018177316 A1 WO 2018177316A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
interest
attribute information
data
point
Prior art date
Application number
PCT/CN2018/080822
Other languages
English (en)
French (fr)
Inventor
吴坤
沈沁
孟凡超
Original Assignee
腾讯科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 腾讯科技(深圳)有限公司 filed Critical 腾讯科技(深圳)有限公司
Publication of WO2018177316A1 publication Critical patent/WO2018177316A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9537Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29Geographical information databases

Definitions

  • the present application relates to the field of Internet technologies, and in particular, to an information identification method, a computing device, and a storage medium.
  • Points of interest may include landmarks, attractions, government departments, or commercial establishments (such as gas stations, department stores, hotels, or hospitals).
  • the server may obtain the address information of the point of interest through the third-party platform, and mark the corresponding point of interest on the corresponding position in the digital map based on the address information, so that the user can search for the point of interest and its coordinates in the digital map. Since the address information of the point of interest is obtained through the third-party platform, the credibility of the address information is not verified, and the server directly marks the point of interest on the digital map according to the address information of the point of interest obtained from the third-party platform. The corresponding location results in a lower accuracy of the points of interest displayed in the digital map.
  • the technical problem to be solved by the embodiments of the present application is to provide an information recognition scheme, which can accurately identify the credibility of the attribute information of the point of interest and improve the accuracy of the points of interest marked in the digital map.
  • an information identification method is provided, which is applied to a computing device, and the method includes: acquiring attribute information of a point of interest, where the attribute information includes identification information or address information of the point of interest; Coordinate data associated with the attribute information, the coordinate data is used to indicate association information of at least one dimension; the attribute information is compared with the coordinate data to obtain a comparison result, and the comparison result is used to describe the Whether the attribute information matches the coordinate data; when it is determined that the attribute information matches the coordinate data according to the comparison result, determining that the credibility of the attribute information is greater than a first credibility threshold; When the comparison result determines that the attribute information does not match the coordinate data, it is determined that the credibility of the attribute information is less than the second credibility threshold.
  • an information identification method is provided, which is applied to a computing device, the method comprising: acquiring identification information corresponding to a point of interest of a credibility to be verified; and determining, according to the identifier information, determining the point of interest corresponding to the point of interest a coordinate position; obtaining peripheral data of the coordinate position; parsing the identification information to acquire associated data corresponding to the identification information; performing feature extraction operation on the associated data and the surrounding data to obtain Corresponding feature data; determining, based on the feature data, whether the coordinate position of the point of interest is authentic using a machine learning model.
  • a computing device including: a processor and a memory; the memory stores computer readable instructions, and the processor is configured to: acquire attribute information of a point of interest, where the attribute information includes Demyimating information or address information of the point of interest; acquiring coordinate data associated with the attribute information by an index algorithm, the coordinate data is used to indicate association information of at least one dimension; and performing the attribute information with the coordinate data Comparing, obtaining a comparison result, the comparison result is used to describe whether the attribute information matches the coordinate data; and when determining that the attribute information matches the coordinate data according to the comparison result, determining the attribute information The credibility is greater than the first credibility threshold; when it is determined that the attribute information does not match the coordinate data according to the comparison result, determining that the credibility of the attribute information is less than the second credibility threshold.
  • a computing device comprising: a processor and a memory; the memory storing computer readable instructions, wherein the processor is configured to: obtain an identifier corresponding to a point of interest of the credibility to be verified Determining a coordinate position corresponding to the point of interest according to the identification information; acquiring peripheral data of the coordinate position; parsing the identification information to obtain associated data corresponding to the identification information; The association data and the peripheral data perform a feature extraction operation to acquire corresponding feature data; and based on the feature data, determine, by the machine learning model, whether the coordinate position of the point of interest is authentic.
  • a non-volatile storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform An instruction according to the information identification method of the present application.
  • FIG. 1A shows a schematic diagram of an application scenario according to some embodiments of the present application
  • FIG. 1B is a schematic structural diagram of an information recognition system provided in an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of an information identification method provided in an embodiment of the present application.
  • FIG. 3 is a schematic diagram of an interface provided in an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of an information identification apparatus provided in an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of a computing device provided in an embodiment of the present application.
  • FIG. 6 shows a schematic diagram of an information identification method 600 in accordance with some embodiments of the present application.
  • the embodiment of the present application provides an information identification method, which acquires attribute information of a point of interest through the Internet, and acquires coordinate data associated with the attribute information by using an index algorithm.
  • the coordinate data is used to indicate association information of at least one dimension.
  • the information recognition method can compare the attribute information with the coordinate data to obtain a comparison result.
  • the comparison result includes the attribute information matching the coordinate data, or the attribute information does not match the coordinate data.
  • the information recognition method may determine that the reliability of the attribute information is greater than the first reliability threshold.
  • the information identification method may determine that the credibility of the attribute information is less than the second credibility threshold, thereby accurately identifying the credibility of the attribute information of the interest point, thereby increasing the number The accuracy of the points of interest marked in the map.
  • the above information identification method can be run on a personal computer, a smart phone (such as an Android mobile phone, an iOS mobile phone, etc.), a tablet computer, a palmtop computer, a mobile Internet device (MID, Mobile Internet Devices), or a wearable smart device.
  • the above information identification method can also be run in a server.
  • the server may be a stand-alone service device in the Internet, or a cluster service device composed of a plurality of independent service devices in the Internet, and the server may include an instant messaging server or a SNS (Social Networking Services) server.
  • SNS Social Networking Services
  • the terminal device and the server are collectively referred to as a computing device hereinafter.
  • FIG. 1A shows a schematic diagram of an application scenario in accordance with some embodiments of the present application.
  • the application scenario may include the terminal device 110 and the server 120.
  • the terminal device 110 can be implemented as the above-described terminal device.
  • the server 120 can be implemented as the above server, and details are not described herein again.
  • the terminal device 110 can communicate with the server 120 over the network 130.
  • the information identification method according to the present application may be performed in the terminal device 110.
  • the information identification method according to the present application may be performed in the server 120.
  • FIG. 1B is a schematic structural diagram of an information recognition system according to an embodiment of the present application.
  • the information recognition system may include a resource layer, an index layer, a recall layer, and a policy layer.
  • the system architecture shown in FIG. 1B can be applied, for example, in the terminal device 110 or the server 120.
  • the resource layer is used to store attribute information of a point of interest and coordinate data associated with the attribute information.
  • the attribute information may include identification information or address information of the point of interest.
  • the identification information may be the name of a point of interest, such as "Tencent Building” and the like.
  • the address information may be location information of the point of interest, such as "No. 66 North Fourth Ring Road” and the like.
  • the attribute information may include, for example, Point of Interest (POI) data as shown in FIG. 1B.
  • the resource layer may obtain attribute information of the point of interest in the digital map.
  • the resource layer may also obtain the attribute information of the point of interest through the Internet, for example, the attribute information of the point of interest is obtained through a browser search or a third-party platform, which is not limited in this application.
  • the index layer is configured to receive attribute information of the point of interest sent by the resource layer, and obtain coordinate data associated with the attribute information in the digital map according to an index algorithm such as a text index, a point data index, a line data index, or a face data index.
  • the coordinate data may include target attribute information, road network data, or a target area.
  • the target attribute information may include target identification information or target address information.
  • Road network data can include road information.
  • the target area may include the zoning surface data, the artificial surface data, the water surface data, the township surface data, or the mining surface data as shown in FIG. 1B.
  • the recall layer is used to acquire the coordinate data retrieved by the index layer, and store the above coordinate data in the resource layer.
  • the recall layer in the embodiment of the present application implements a global recall of text and online clustering on the index chain recall collection.
  • the policy layer is used to obtain attribute information of a point of interest in the resource layer, and coordinate data associated with the attribute information, and perform natural language processing (NLP), text parsing, and feature extraction on the attribute information and the coordinate data. And credibility recognition and so on.
  • NLP natural language processing
  • the specific manner in which the policy layer performs NLP on the attribute information or the coordinate data may include word segmentation, normalization, name role labeling, or address segmentation.
  • the word segmentation may perform data segmentation on attribute information or coordinate data based on proper nouns, category words or business nouns to obtain one or more unit attribute information. Normalization can be a case conversion, a simplified and traditional conversion, or a Chinese character number conversion for attribute information or coordinate data.
  • the specific manner in which the policy layer performs text parsing on the attribute information or the coordinate data may include name resolution and address resolution.
  • the name resolution may be, for example, parsing the identification information in the attribute information to obtain a profile, a road, an entity, a house number, a landmark, or an intersection.
  • the address resolution may be, for example, parsing the address information in the attribute information to obtain a city, a zoning, a township, a contour, a road, an entity, a house number, a landmark, or an intersection.
  • the feature extraction may specifically extract a location feature, a text feature, an environment feature, or an attribute feature.
  • the credibility identification can determine whether the extracted location points of the city, the zoning, the township, the outline, or the road are consistent with the city, the zoning, the township, the outline, or the road included in the coordinate data. In one embodiment, the credibility identification may determine the distribution of text features such as entities, house numbers, landmarks, or intersections that are extracted from the attribute information in the global POI. In one embodiment, the credibility identification may determine environmental characteristics such as perimeter density or regional address diversity of the POI. In one embodiment, the credibility identification can determine whether the attribute characteristics such as the entity are unique. For example, if there is only one Tiananmen in the country, Tiananmen Square is unique, and if the Carrefour supermarket is a chain store, Carrefour is not unique. The credibility identification identifies the credibility of the attribute information of the point of interest by the above judgment, and the attribute information can be corrected using the preset rule for the data of the boundary part of the model classification.
  • FIG. 2 is a schematic flowchart diagram of an information identification method disclosed in an embodiment of the present application.
  • the method shown in FIG. 2 can be performed, for example, in a computing device such as the terminal device 110 or the server 120.
  • the information identification method may at least include the following steps:
  • the attribute information includes identification information or address information of the point of interest.
  • the computing device can obtain attribute information of the point of interest through the Internet, and the attribute information includes identification information or address information of the point of interest.
  • the computing device may obtain attribute information of the point of interest in the digital map or obtain attribute information of the point of interest through the browser.
  • the attribute information may include identification information or address information of the point of interest, and the like.
  • the identification information may be the name of the point of interest, and the like.
  • the computing device can obtain coordinate data associated with the attribute information through an indexing algorithm.
  • the coordinate data may be used to indicate association information of at least one dimension, for example, the coordinate data may include target attribute information, road network data or a target area, and the like.
  • the computing device may perform data segmentation on the attribute information of the point of interest to obtain one or more unit attribute information, and find a target in the information database that has a similarity with the unit attribute information that is greater than the first ratio threshold.
  • Attribute information wherein the target attribute information may include target identification information or target address information.
  • the computing device may tag the points of interest in the digital map based on the address information, and obtain road network data connected to the points of interest in the digital map, the road network data including road information.
  • the computing device may mark a point of interest in the digital map based on the address information, and obtain a target area in the digital map that is less than the first distance threshold from the point of interest.
  • the computing device can compare the attribute information with the coordinate data to obtain a comparison result.
  • the comparison result may indicate that the attribute information matches the coordinate data, or that the attribute information and the coordinate data do not match.
  • the comparison result may include a position feature comparison result, a text feature comparison result, an environmental feature comparison result, or an attribute feature comparison result.
  • the computing device may acquire a set of points of interest including a point of interest, where the similarity between the identification information of each two points of interest included in the set of points of interest is greater than a second ratio threshold, and the determined set of points of interest is included The sum of the number of all points of interest is less than the number threshold.
  • the computing device may determine that the attribute information does not match the coordinate data.
  • the computing device may determine that the attribute feature comparison result is 0; when not in the information database When the target attribute information with the similarity between the unit attribute information and the first ratio threshold is found, the computing device may determine that the attribute feature comparison result is 1.
  • the computing device may perform data segmentation on the address information of the point of interest to obtain one or more unit address information.
  • the computing device may determine that the attribute information and the coordinate data are not match.
  • the computing device may determine that the location feature comparison result is 0; when the cell address information matches the road information, the computing device may determine that the location feature comparison result is 1.
  • the computing device may perform an analysis process on the target area, acquire the flux of the target area, and obtain a sum of the number of target points of interest that are less than the second distance threshold from the target area in the digital map. The similarity between the identification information of the target interest point and the identification information of the interest point is greater than the second proportional threshold.
  • the computing device may determine that the attribute information does not match the coordinate data.
  • the computing device may determine that the environmental feature comparison result is 0; when the liquidity of the target area matches the sum of the target interest points, The computing device can determine that the environmental feature comparison result is one.
  • the computing device may perform data segmentation on the identification information of the point of interest to obtain one or more unit identification information.
  • the computing device may determine the attribute information and The coordinate data does not match.
  • the computing device may determine that the text feature comparison result is 0; when the unit identification information matches the attribute information of the target area, the computing device may determine the text feature. The comparison result is 1.
  • the computing device can retrieve the point line surface data, and recall the coordinate data around the point of interest, and analyze and identify from multiple dimensions. For example, the computing device can analyze the attribute information of the point of interest to obtain a road, a city, a door, a zoning, an entity, a town, a building, a contour, or an intersection. The computing device can also perform feature calculations such as reference to road distance, contour matching, number of gate supports, gate support distance, number of physical supports, or physical support distance. The computing device is identified by rules, for example, by sub-rules such as location features, text features, attribute features, or environment features, and then the rules are combined to obtain the credibility of the attribute information.
  • rules for example, by sub-rules such as location features, text features, attribute features, or environment features, and then the rules are combined to obtain the credibility of the attribute information.
  • the computing device may determine that the attribute information has a high degree of credibility, and further determine that the credibility of the attribute information is greater than the first credibility threshold.
  • the first confidence threshold may be 80% or 90%, and the like.
  • the computing device may identify the credibility of the attribute information based on the location feature comparison result, the text feature comparison result, the attribute feature comparison result, or the environment feature comparison result.
  • the computing device may determine that the attribute information has low credibility, and further determine that the credibility of the attribute information is less than the second credibility threshold, for example, the second The confidence threshold can be 60% or 50%, and the like. Specifically, the computing device may identify the credibility of the attribute information based on the location feature comparison result, the text feature comparison result, the attribute feature comparison result, or the environment feature comparison result.
  • the computing device may delete the points of interest noted in the digital map based on the address information.
  • the computing device may mark the point of interest in the digital map.
  • the third confidence threshold may be 40% or 50%, and the like.
  • the first credibility threshold may be greater than the second credibility threshold, and the second credibility threshold may be greater than the third credibility threshold.
  • the attribute information of the interest point is obtained through the Internet, the coordinate data associated with the attribute information is obtained by the index algorithm, and the attribute information is compared with the coordinate data to obtain a comparison result, and the comparison result includes the attribute information and the coordinate data. If the matching, or the attribute information does not match the coordinate data, and the attribute information is matched with the coordinate data according to the comparison result, the credibility of the identification attribute information is greater than the first credibility threshold, and the attribute information does not match the coordinate data according to the comparison result.
  • the credibility of the identification attribute information is less than the second credibility threshold, and the credibility of the attribute information of the point of interest can be accurately identified, and the accuracy of the points of interest marked in the digital map is improved.
  • FIG. 4 is a schematic structural diagram of an information identification apparatus provided in an embodiment of the present application.
  • the information identifying apparatus in this embodiment may include at least an attribute information acquiring module 401, a coordinate data acquiring module 402, a comparing module 403, and a credibility identifying module 404, where:
  • the attribute information obtaining module 401 is configured to acquire attribute information of a point of interest through the Internet, where the attribute information includes identification information or address information of the point of interest.
  • the coordinate data obtaining module 402 is configured to acquire coordinate data associated with the attribute information by using an index algorithm, where the coordinate data is used to indicate association information of at least one dimension.
  • the comparison module 403 is configured to compare the attribute information with the coordinate data to obtain a comparison result, where the comparison result includes the attribute information matching the coordinate data, or the attribute information does not match the coordinate data.
  • the credibility identification module 404 is configured to: when the attribute information is matched with the coordinate data according to the comparison result, identify that the credibility of the attribute information is greater than a first credibility threshold.
  • the credibility identification module 404 is further configured to: when determining that the attribute information does not match the coordinate data according to the comparison result, identify that the credibility of the attribute information is less than a second credibility threshold.
  • the coordinate data obtaining module 402 is specifically configured to:
  • the comparing module 403 is specifically configured to:
  • the similarity between the identification information of each two points of interest included in the set of points of interest is greater than a second ratio threshold.
  • the target attribute information that is similar to the first ratio threshold is found in the information database, it is determined that the attribute information does not match the coordinate data.
  • the coordinate data obtaining module 402 is specifically configured to:
  • the point of interest is marked in the digital map based on the address information.
  • the comparing module 403 is specifically configured to:
  • the coordinate data obtaining module 402 is specifically configured to:
  • the point of interest is marked in the digital map based on the address information.
  • a target area having a distance from the point of interest that is less than a first distance threshold is acquired in the digital map.
  • the comparing module 403 is specifically configured to:
  • the comparing module 403 is specifically configured to:
  • the attribute information acquiring module 401 acquires the attribute information of the point of interest through the Internet
  • the coordinate data acquiring module 402 acquires the coordinate data associated with the attribute information by using an indexing algorithm
  • the comparing module 403 compares the attribute information with the The coordinate data is compared to obtain a comparison result.
  • the comparison result includes the attribute information matching the coordinate data, or the attribute information does not match the coordinate data
  • the credibility identification module 404 determines the attribute information when the attribute information matches the coordinate data according to the comparison result.
  • the credibility is greater than the first credibility threshold; when the attribute information does not match the coordinate data according to the comparison result, the credibility of the identifier information is less than the second credibility threshold, and the attribute information of the point of interest can be accurately identified. Reliability to improve the accuracy of points of interest marked in digital maps.
  • FIG. 5 is a schematic structural diagram of a computing device according to an embodiment of the present disclosure.
  • the computing device provided by the embodiment of the present application may be used to implement the information identifying method shown in FIG. 2, and is only shown for convenience of description.
  • FIG. 2 For a part related to the embodiment of the present application, and the specific technical details are not disclosed, please refer to the embodiment of the present application shown in FIG. 2 .
  • the computing device includes at least one processor 501, such as a CPU, at least one input device 503, at least one output device 504, a memory 505, and at least one communication bus 502.
  • the communication bus 502 is used to implement connection communication between these components.
  • the input device 503 may specifically be a network interface or the like for acquiring attribute information of a point of interest.
  • the output device 504 may be a network interface or the like for outputting a digital map marked with a point of interest.
  • the memory 505 may include a high speed RAM memory, and may also include a non-unstable memory, such as at least one disk memory, specifically for storing attribute information of a point of interest, coordinate data associated with the attribute information, and the like.
  • the memory 505 can optionally include at least one storage device located remotely from the aforementioned processor 501.
  • a set of program codes is stored in the memory 505, and may include, for example, the information identifying means shown in FIG.
  • the processor 501, the input device 503, and the output device 504 call the program code stored in the memory 505 for performing the following operations:
  • the input device 503 acquires attribute information of a point of interest through the Internet, and the attribute information includes identification information or address information of the point of interest.
  • the processor 501 acquires coordinate data associated with the attribute information by an indexing algorithm, the coordinate data being used to indicate association information of at least one dimension.
  • the processor 501 compares the attribute information with the coordinate data to obtain a comparison result, the comparison result including the attribute information matching the coordinate data, or the attribute information does not match the coordinate data.
  • the processor 501 determines that the attribute information is matched with the coordinate data according to the comparison result, and the reliability of the attribute information is greater than the first reliability threshold.
  • the processor 501 determines that the attribute information does not match the coordinate data according to the comparison result, the processor 501 identifies that the reliability of the attribute information is less than the second reliability threshold.
  • the processor 501 identifies the credibility of the attribute information according to the comparison result.
  • the following operations may also be performed:
  • the processor 501 deletes the point of interest marked in the digital map based on the address information.
  • the processor 501 acquires coordinate data associated with the attribute information by using an index algorithm, which may be specifically:
  • the processor 501 performs data segmentation on the attribute information of the interest point to obtain one or more unit attribute information.
  • the processor 501 searches the information database for target attribute information having a degree of similarity with the unit attribute information that is greater than a first ratio threshold, the target attribute information including target identification information or target address information.
  • the processor 501 compares the attribute information with the coordinate data to obtain a comparison result, which may be specifically:
  • the processor 501 acquires a set of points of interest including the point of interest, and the similarity between the identification information of each two points of interest included in the set of points of interest is greater than a second ratio threshold.
  • the processor 501 determines that the sum of the number of all points of interest included in the set of points of interest is less than a quantity threshold.
  • the processor 501 determines that the attribute information does not match the coordinate data.
  • the processor 501 acquires coordinate data associated with the attribute information by using an index algorithm, which may be specifically:
  • the processor 501 labels the points of interest in the digital map based on the address information.
  • the processor 501 acquires road network data connected to the point of interest in the digital map, and the road network data includes road information.
  • the processor 501 compares the attribute information with the coordinate data to obtain a comparison result, which may be specifically:
  • the processor 501 performs data segmentation on the address information of the point of interest to obtain one or more unit address information.
  • the processor 501 determines that the attribute information does not match the coordinate data.
  • the processor 501 acquires coordinate data associated with the attribute information by using an index algorithm, which may be specifically:
  • the processor 501 labels the points of interest in the digital map based on the address information.
  • the processor 501 acquires, in the digital map, a target area having a distance from the point of interest that is less than a first distance threshold.
  • the processor 501 compares the attribute information with the coordinate data to obtain a comparison result, which may be specifically:
  • the processor 501 performs an analysis process on the target area to acquire a throughput of the target area.
  • the processor 501 acquires, in the digital map, a sum of the number of target points of interest that is less than a second distance threshold between the target area, and the identifier information of the target point of interest and the identification information of the point of interest The similarity is greater than the second proportional threshold.
  • the processor 501 determines that the attribute information does not match the coordinate data.
  • the processor 501 compares the attribute information with the coordinate data to obtain a comparison result, which may be specifically:
  • the processor 501 performs data segmentation on the identification information of the point of interest to obtain one or more unit identification information.
  • the processor 501 determines that the attribute information does not match the coordinate data.
  • the terminal introduced in the embodiment of the present application may be used to implement some or all of the processes in the method embodiment introduced in conjunction with FIG. 2 in this application.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).
  • FIG. 6 shows a schematic diagram of an information identification method 600 in accordance with some embodiments of the present application.
  • the information recognition method 600 can be performed, for example, in the computing device shown in FIG. 5, but is not limited thereto.
  • step S601 identification information corresponding to the point of interest of the credibility to be verified is acquired.
  • the identification information acquired in step S601 is an address or a name.
  • step S602 a coordinate position corresponding to the point of interest is determined based on the identification information.
  • step S602 may determine the coordinate position by means of a geographic location service, but is not limited thereto.
  • step S603 peripheral data of the coordinate position is acquired.
  • step S603 may acquire a set of points of interest distributed around the coordinate position.
  • step S603 may also acquire an attribute item corresponding to the coordinate position.
  • the attribute item corresponding to the coordinate position may include, for example, at least one of a region outline, a road, an entity, a house number, a landmark, and an intersection.
  • step S604 the identification information is parsed.
  • step S604 can perform semantic analysis on the identification information to obtain a semantic analysis result.
  • the semantic analysis method is, for example, a natural language processing method, but is not limited thereto.
  • step S604 can determine the associated data based on the results of the speech analysis.
  • the associated data may include dotted line data and the like related to the identification information.
  • step S604 may acquire an attribute item corresponding to the identification information according to the result of the voice analysis.
  • the attribute item may include, for example, at least one of an area outline, a road, an entity, a house number, a landmark, and an intersection to which the identification information relates.
  • step S604 can acquire an attribute item referenced by the text content of the identification information.
  • step S605 a feature extraction operation is performed on the associated data and the surrounding data to acquire corresponding feature data.
  • step S605 may determine at least one of a location feature, a text feature, an environment feature, and an attribute feature corresponding to the identification information.
  • the location feature is used to describe whether the attribute item corresponding to the coordinate position is consistent with the identifier information.
  • the location features may include: reference road distance, contour matching, gate support number, gate support distance, entity support number, or entity support distance.
  • the text feature is configured to describe at least one of a house number, a landmark, and an intersection referenced by the identification information to distribute features in the set of points of interest.
  • the environmental feature is used to describe at least one of a peripheral density and a regional address diversity of the coordinate position.
  • the attribute feature is used to describe whether the entity referenced by the identification information is unique and at least one of the radiation ranges.
  • step S606 based on the feature data, the machine learning model is used to determine whether the coordinate position of the point of interest is authentic. For example, based on the feature data, step S606 can utilize the gradient promotion decision tree to determine if the coordinate location is authentic. In addition, the step S606 can also determine whether the coordinate position is trusted by using other machine learning manners, which is not limited in this application. In addition, for a more specific implementation of the method 600, please refer to the descriptions of FIG. 1B and FIG. 3 above, and details are not described herein again. In summary, the method 600 can improve the accuracy of determining the points of interest marked in the digital map based on the surrounding information of the identification information and the coordinate position and using the machine learning method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Remote Sensing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请实施例公开了信息识别方法、计算设备及存储介质。其中,一种信息识别方法应用于计算设备中,包括:获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息;通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息;将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果用于描述所述属性信息与所述坐标数据是否匹配;当根据所述比较结果确定所述属性信息与所述坐标数据匹配时,确定所述属性信息的可信度大于第一可信度阈值;当根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,确定所述属性信息的可信度小于第二可信度阈值。

Description

信息识别方法、计算设备及存储介质
本申请要求于2017年03月29日提交中国专利局、申请号为201710198353.0、申请名称为“一种信息识别方法及装置”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及互联网技术领域,尤其涉及信息识别方法、计算设备及存储介质。
背景技术
目前,数字地图可以提供导航以及搜索兴趣点等功能。兴趣点可以包括地标、景点、政府部门或者商业机构(例如加油站、百货公司、酒店或者医院)等。示例性的,服务器可以通过第三方平台获取兴趣点的地址信息,并基于该地址信息将对应的兴趣点标注在数字地图中的相应位置,以便用户可以在数字地图中搜索兴趣点及其坐标。由于兴趣点的地址信息是通过第三方平台获取到的,其地址信息的可信度无从考证,服务器直接根据从第三方平台获取到的兴趣点的地址信息,将该兴趣点标注在数字地图的相应位置,导致数字地图中显示的兴趣点的准确性较低。
发明内容
本申请实施例所要解决的技术问题在于,提供一种信息识别方案,可精确识别兴趣点的属性信息的可信度,提高数字地图中标注的兴趣点的准确性。
根据本申请一方面,提供一种信息识别方法,应用于计算设备,所述方法包括:获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息;通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息;将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果用于描述所述属性信息与所述坐标数据是否匹配;当根据所述比较结果确定所述属性信息与所述坐标数据匹配时,确定所述属性信息的可信度大于第一可信度阈值;当根据所述比较结果确定所述属性 信息与所述坐标数据不匹配时,确定所述属性信息的可信度小于第二可信度阈值。
根据本申请一方面,提供一种信息识别方法,应用于计算设备,所述方法包括:获取与待验证可信度的兴趣点对应的标识信息;根据所述标识信息确定与所述兴趣点对应的坐标位置;获取所述坐标位置的周边数据;对所述标识信息进行解析,以获取与所述标识信息对应的关联数据;对所述关联数据和所述周边数据进行特征提取操作,以获取相应的特征数据;基于所述特征数据,利用机器学习模型确定所述兴趣点的所述坐标位置是否可信。
根据本申请一方面,提供一种计算设备,包括:处理器和存储器;所述存储器中存储有计算机可读指令,可以使所述处理器:获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息;通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息;将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果用于描述所述属性信息与所述坐标数据是否匹配;当根据所述比较结果确定所述属性信息与所述坐标数据匹配时,确定所述属性信息的可信度大于第一可信度阈值;当根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,确定所述属性信息的可信度小于第二可信度阈值。
根据本申请一方面,提供一种计算设备,包括:处理器和存储器;所述存储器中存储有计算机可读指令,可以使所述处理器:获取与待验证可信度的兴趣点对应的标识信息;根据所述标识信息确定与所述兴趣点对应的坐标位置;获取所述坐标位置的周边数据;对所述标识信息进行解析,以获取与所述标识信息对应的关联数据;对所述关联数据和所述周边数据进行特征提取操作,以获取相应的特征数据;基于所述特征数据,利用机器学习模型确定所述兴趣点的所述坐标位置是否可信。
根据本申请一方面,提供一种非易失性存储介质,存储有一个或多个程序,所述一个或多个程序包括指令,所述指令当由计算设备执行时,使得所述计算设备执行根据本申请的信息识别方法的指令。
附图简要说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1A示出了根据本申请一些实施例的应用场景的示意图;
图1B是本申请实施例中提供的一种信息识别系统的架构示意图;
图2是本申请实施例中提供的一种信息识别方法的流程示意图;
图3是本申请实施例中提供的一种界面示意图;
图4是本申请实施例中提供的一种信息识别装置的结构示意图;
图5是本申请实施例中提供的一种计算设备的结构示意图;
图6示出了根据本申请一些实施例的信息识别方法600的示意图。
实施本申请的方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请实施例提供了一种信息识别方法,通过互联网获取兴趣点的属性信息,通过索引算法获取与该属性信息相关联的坐标数据。其中,坐标数据用于指示至少一个维度的关联信息。信息识别方法可以将属性信息与坐标数据进行比较,得到比较结果。比较结果包括属性信息与坐标数据匹配,或者属性信息与坐标数据不匹配。在根据比较结果确定属性信息与坐标数据匹配时,信息识别方法可以确定属性信息的可信度大于第一可信度阈值。在根据比较结果确定属性信息与坐标数据不匹配时,信息识别方法可以确定属性信息的可信度小于第二可信度阈值,从而可以精确识别兴趣点的属性信息的可信度,进而提高数字地图中标注的兴趣点的准确性。
在一个实施例中,上述信息识别方法可以运行在个人电脑、智能手机(如Android手机、iOS手机等)、平板电脑、掌上电脑、移动互联网设备(MID,Mobile Internet Devices)或穿戴式智能设备等终端设备中。在一个实施例中,上 述信息识别方法也可以运行在服务器中。服务器可以是互联网中的一个独立服务设备,或者是由互联网中的多个独立服务设备共同构成的集群服务设备,服务器可以包括即时通信服务器或者SNS(Social Networking Services,社会性网络服务)服务器等,例如数字地图服务器或者导航服务器等。为了简化描述,下文中终端设备和服务器统称为计算设备。图1A示出了根据本申请一些实施例的应用场景的示意图。如图1A所示,应用场景可以包括终端设备110和服务器120。这里,终端设备110可以实现为上述终端设备。服务器120可以实现为上述服务器,这里不再赘述。终端设备110可以通过网络130与服务器120通信。在一个实施例中,根据本申请的信息识别方法可以在终端设备110中执行。在另一个实施例中,根据本申请的信息识别方法可以在服务器120中执行。
图1B示出了本申请实施例的信息识别系统的架构示意图。该信息识别系统可以包括资源层、索引层、召回层以及策略层。这里,图1B所示的系统架构例如可以应用在终端设备110或服务器120中。
资源层用于存储兴趣点的属性信息以及与该属性信息相关联的坐标数据。其中,属性信息可以包括兴趣点的标识信息或者地址信息。标识信息可以为兴趣点的名称,例如“腾讯大厦”等。地址信息可以为该兴趣点的位置信息,例如“北四环西路66号”等。属性信息例如可以包括图1B所示的兴趣点(Point of Interest,POI)数据。具体的,资源层可以在数字地图中获取兴趣点的属性信息。在一个实施例中,资源层还可以通过互联网获取兴趣点的属性信息,例如通过浏览器搜索或者第三方平台等方式获取兴趣点的属性信息,本申请对此不做限定。
索引层用于接收资源层发送的兴趣点的属性信息,并根据文本索引、点数据索引、线数据索引或者面数据索引等索引算法,在数字地图中获取与该属性信息相关联的坐标数据。坐标数据可以包括目标属性信息、路网数据或者目标区域等。目标属性信息可以包括目标标识信息或者目标地址信息。路网数据可以包括道路信息。目标区域可以包括如图1B所示的区划面数据、人工面数据、水域面数据、乡镇面数据或者挖掘面数据等。
召回层用于获取索引层检索到的坐标数据,并将上述坐标数据存储到资源层中。本申请实施例中的召回层实现了对文本的全局召回,以及索引链召回集 合上的在线聚类。
策略层用于获取资源层中的兴趣点的属性信息,以及与该属性信息关联的坐标数据,并对该属性信息和坐标数据进行自然语言处理(Natural Language Processing,NLP)、文本解析、特征提取以及可信度识别等。其中,策略层对属性信息或者坐标数据进行NLP的具体方式可以包括分词、归一化、名称角色标注或者地址切分等。分词可以为基于专有名词、类别词或者业务名词等对属性信息或者坐标数据进行数据切分,得到一个或多个单元属性信息。归一化可以为对属性信息或者坐标数据进行大小写转换、简繁体转换或者汉字数字转换等。策略层对属性信息或者坐标数据进行文本解析的具体方式可以包括名称解析和地址解析。名称解析例如可以是对属性信息中的标识信息进行解析,从而得到轮廓、道路、实体、门牌号、地标或者交叉路口等。地址解析例如可以是对属性信息中的地址信息进行解析,从而得到城市、区划、乡镇、轮廓、道路、实体、门牌号、地标或者交叉路口等。特征提取具体可以提取位置特征、文本特征、环境特征或者属性特征等。可信度识别可以判断提取到的兴趣点位于的城市、区划、乡镇、轮廓或者道路等位置特征,与坐标数据包含的城市、区划、乡镇、轮廓或者道路等是否一致。在一个实施例中,可信度识别可以确定对属性信息进行文本解析提取到的实体、门牌号、地标或者交叉路口等文本特征在全局POI中的分布。在一个实施例中,可信度识别可以确定POI的周边密度或者区域地址多样性等环境特征。在一个实施例中,可信度识别可以判断实体等属性特征是否唯一,例如全国仅有一个天安门,则天安门是唯一的,又如家乐福超市是连锁店,则家乐福不是唯一的。可信度识别通过上述判断识别兴趣点的属性信息的可信度,对于模型分类边界部分的数据,可使用预置规则对该属性信息进行修正。
请参见图2,图2是本申请实施例公开的一种信息识别方法的流程示意图。图2所示的方法例如可以在终端设备110或服务器120等计算设备中执行。如图2所示,该信息识别方法至少可以包括以下步骤:
S201,通过互联网获取兴趣点的属性信息。属性信息包括兴趣点的标识信息或者地址信息。
计算设备可以通过互联网获取兴趣点的属性信息,属性信息包括兴趣点的 标识信息或者地址信息。例如,计算设备可以在数字地图中获取兴趣点的属性信息,或者通过浏览器获取兴趣点的属性信息。其中,属性信息可以包括兴趣点的标识信息或者地址信息等。示例性的,标识信息可以为兴趣点的名称等。
S202,通过索引算法获取与属性信息相关联的坐标数据。
计算设备可以通过索引算法获取与属性信息相关联的坐标数据。其中,坐标数据可以用于指示至少一个维度的关联信息,例如坐标数据可以包括目标属性信息、路网数据或者目标区域等。
在一个实施例中,计算设备可以对兴趣点的属性信息进行数据切分,得到一个或多个单元属性信息,在信息数据库中查找与单元属性信息之间的相似度大于第一比例阈值的目标属性信息,其中目标属性信息可以包括目标标识信息或者目标地址信息。
在一个实施例中,计算设备可以基于地址信息在数字地图中标注兴趣点,在数字地图中获取与兴趣点相连接的路网数据,路网数据包括道路信息。
在一个实施例中,计算设备可以基于地址信息在数字地图中标注兴趣点,在数字地图中获取与兴趣点之间的距离小于第一距离阈值的目标区域。
S203,将属性信息与坐标数据进行比较,得到比较结果,比较结果包括属性信息与坐标数据匹配,或者属性信息与坐标数据不匹配。
计算设备可以将属性信息与坐标数据进行比较,得到比较结果。例如,比较结果可以表示属性信息和坐标数据匹配,或者表示属性信息和坐标数据不匹配。又如,比较结果可以包括位置特征比较结果、文本特征比较结果、环境特征比较结果或者属性特征比较结果等。
在一个实施例中,计算设备可以获取包含兴趣点的兴趣点集合,兴趣点集合所包含的每两个兴趣点的标识信息之间的相似度大于第二比例阈值,确定兴趣点集合所包含的所有兴趣点的数量总和小于数量阈值。当在信息数据库中查找到与单元属性信息之间的相似度大于第一比例阈值的目标属性信息时,计算设备可以确定属性信息与坐标数据不匹配。在一个实施例中,当在信息数据库中查找到与单元属性信息之间的相似度大于第一比例阈值的目标属性信息时,计算设备可以确定属性特征比较结果为0;当在信息数据库中未查找到与单元属性信息之间的相似度大于第一比例阈值的目标属性信息时,计算设备可以确定 属性特征比较结果为1。
在一个实施例中,计算设备可以对兴趣点的地址信息进行数据切分,得到一个或多个单元地址信息,当单元地址信息与道路信息不匹配时,计算设备可以确定属性信息与坐标数据不匹配。在一个实施例中,当单元地址信息与道路信息不匹配时,计算设备可以确定位置特征比较结果为0;当单元地址信息与道路信息匹配时,计算设备可以确定位置特征比较结果为1。
在一个实施例中,计算设备可以对目标区域进行分析处理,获取目标区域的流通量,在数字地图中获取与目标区域之间的距离小于第二距离阈值的目标兴趣点的数量总和。目标兴趣点的标识信息与兴趣点的标识信息之间的相似度大于第二比例阈值。当目标区域的流通量与目标兴趣点的数量总和不匹配时,计算设备可以确定属性信息与坐标数据不匹配。在一个实施例中,当目标区域的流通量与目标兴趣点的数量总和不匹配时,计算设备可以确定环境特征比较结果为0;当目标区域的流通量与目标兴趣点的数量总和匹配时,计算设备可以确定环境特征比较结果为1。
在一个实施例中,计算设备可以对兴趣点的标识信息进行数据切分,得到一个或多个单元标识信息,当单元标识信息与目标区域的属性信息不匹配时,计算设备可以确定属性信息与坐标数据不匹配。在一个实施例中,当单元标识信息与目标区域的属性信息不匹配时,计算设备可以确定文本特征比较结果为0;当单元标识信息与目标区域的属性信息匹配时,计算设备可以确定文本特征比较结果为1。
以图3所示的界面示意图为例,计算设备获取到兴趣点的属性信息之后,可以检索点线面数据,并召回兴趣点周边的坐标数据,从多个维度进行分析识别。例如计算设备可以对兴趣点的属性信息进行分析,从而得到道路、城市、门址、区划、实体、乡镇、楼栋、轮廓或者交叉路口等。计算设备还可以进行特征计算,例如引用道路距离,轮廓是否匹配、门址支持数量、门址支持距离、实体支持数量或者实体支持距离等。计算设备通过规则进行识别,例如可以通过位置特征、文本特征、属性特征或者环境特征等子规则进行识别,进而进行规则组合,得到属性信息的可信度。
S204,根据比较结果确定属性信息与坐标数据匹配时,识别属性信息的可 信度大于第一可信度阈值。
在一个实施例中,当比较结果为属性信息与坐标信息匹配时,计算设备可以确定该属性信息的可信度较高,进而确定该属性信息的可信度大于第一可信度阈值,示例性的,第一可信度阈值可以为80%或者90%等。具体的,计算设备可以基于位置特征比较结果、文本特征比较结果、属性特征比较结果或者环境特征比较结果,识别属性信息的可信度。
S205,根据比较结果确定属性信息与坐标数据不匹配时,识别属性信息的可信度小于第二可信度阈值。
当比较结果为属性信息与坐标信息不匹配时,计算设备可以确定该属性信息的可信度较低,进而确定该属性信息的可信度小于第二可信度阈值,示例性的,第二可信度阈值可以为60%或者50%等。具体的,计算设备可以基于位置特征比较结果、文本特征比较结果、属性特征比较结果或者环境特征比较结果,识别属性信息的可信度。
在一个实施例中,当属性信息的可信度小于第三可信度阈值时,计算设备可以将基于地址信息在数字地图中标注的兴趣点进行删除。当属性信息的可信度大于或者等于第三可信度阈值时,计算设备可以在数字地图中标注该兴趣点。示例性的,第三可信度阈值可以为40%或者50%等。
在一个实施例中,第一可信度阈值可以大于第二可信度阈值,第二可信度阈值可以大于第三可信度阈值。
本申请实施例中,通过互联网获取兴趣点的属性信息,通过索引算法获取与该属性信息相关联的坐标数据,将属性信息与坐标数据进行比较,得到比较结果,比较结果包括属性信息与坐标数据匹配,或者属性信息与坐标数据不匹配,根据比较结果确定属性信息与坐标数据匹配时,识别属性信息的可信度大于第一可信度阈值,根据比较结果确定属性信息与坐标数据不匹配时,识别属性信息的可信度小于第二可信度阈值,可精确识别兴趣点的属性信息的可信度,提高数字地图中标注的兴趣点的准确性。
请参见图4,图4是本申请实施例中提供的一种信息识别装置的结构示意图。如图4所示,本实施例中的信息识别装置至少可以包括属性信息获取模块401、坐标数据获取模块402、比较模块403以及可信度识别模块404,其中:
属性信息获取模块401,用于通过互联网获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息。
坐标数据获取模块402,用于通过索引算法获取与所述属性信息相关联的坐标数据,坐标数据用于指示至少一个维度的关联信息。
比较模块403,用于将所述属性信息与所述坐标数据进行比较,得到比较结果,比较结果包括属性信息与坐标数据匹配,或者属性信息与坐标数据不匹配。
可信度识别模块404,用于根据所述比较结果确定所述属性信息与所述坐标数据匹配时,识别所述属性信息的可信度大于第一可信度阈值。
所述可信度识别模块404,还用于根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,识别所述属性信息的可信度小于第二可信度阈值。
在一个实施例中,所述坐标数据获取模块402,具体用于:
对所述兴趣点的属性信息进行数据切分,得到一个或多个单元属性信息。
在信息数据库中查找与所述单元属性信息之间的相似度大于第一比例阈值的目标属性信息,所述目标属性信息包括目标标识信息或者目标地址信息。
在一个实施例中,所述比较模块403,具体用于:
获取包含所述兴趣点的兴趣点集合,所述兴趣点集合所包含的每两个兴趣点的标识信息之间的相似度大于第二比例阈值。
确定所述兴趣点集合所包含的所有兴趣点的数量总和小于数量阈值。
当在所述信息数据库中查找到与所述单元属性信息之间的相似度大于所述第一比例阈值的目标属性信息时,确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,所述坐标数据获取模块402,具体用于:
基于所述地址信息在所述数字地图中标注所述兴趣点。
在所述数字地图中获取与所述兴趣点相连接的路网数据,所述路网数据包括道路信息。
在一个实施例中,所述比较模块403,具体用于:
对所述兴趣点的地址信息进行数据切分,得到一个或多个单元地址信息。
当所述单元地址信息与所述道路信息不匹配时,确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,所述坐标数据获取模块402,具体用于:
基于所述地址信息在所述数字地图中标注所述兴趣点。
在所述数字地图中获取与所述兴趣点之间的距离小于第一距离阈值的目标区域。
在一个实施例中,所述比较模块403,具体用于:
对所述目标区域进行分析处理,获取所述目标区域的流通量。
在所述数字地图中获取与所述目标区域之间的距离小于第二距离阈值的目标兴趣点的数量总和,所述目标兴趣点的标识信息与所述兴趣点的标识信息之间的相似度大于第二比例阈值。
当所述目标区域的流通量与所述目标兴趣点的数量总和不匹配时,确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,所述比较模块403,具体用于:
对所述兴趣点的标识信息进行数据切分,得到一个或多个单元标识信息。
当所述单元标识信息与所述目标区域的属性信息不匹配时,确定所述属性信息与所述坐标数据不匹配。
本申请实施例中,属性信息获取模块401通过互联网获取兴趣点的属性信息,坐标数据获取模块402通过索引算法获取与所述属性信息相关联的坐标数据,比较模块403将所述属性信息与所述坐标数据进行比较,得到比较结果,比较结果包括属性信息与坐标数据匹配,或者属性信息与坐标数据不匹配,可信度识别模块404根据比较结果确定属性信息与坐标数据匹配时,识别属性信息的可信度大于第一可信度阈值;根据比较结果确定属性信息与坐标数据不匹配时,识别属性信息的可信度小于第二可信度阈值,可精确识别兴趣点的属性信息的可信度,提高数字地图中标注的兴趣点的准确性。
请参见图5,图5为本申请实施例提供的一种计算设备的结构示意图,本申请实施例提供的计算设备可以用于实施上述图2所示的信息识别方法,为了便于说明,仅示出了与本申请实施例相关的部分,具体技术细节未揭示的,请参照图2所示的本申请实施例。
如图5所示,该计算设备包括:至少一个处理器501,例如CPU,至少一个输入装置503,至少一个输出装置504,存储器505,至少一个通信总线502。其中,通信总线502用于实现这些组件之间的连接通信。其中,输入装置503 具体可以为网络接口等,用于获取兴趣点的属性信息。其中,输出装置504具体可以为网络接口等,用于输出标注了兴趣点的数字地图。其中,存储器505可能包含高速RAM存储器,也可能还包括非不稳定的存储器,例如至少一个磁盘存储器,具体用于存储兴趣点的属性信息,以及与属性信息关联的坐标数据等。存储器505可选的可以包含至少一个位于远离前述处理器501的存储装置。存储器505中存储一组程序代码,例如可以包括图4所示的信息识别装置。处理器501、输入装置503以及输出装置504调用存储器505中存储的程序代码,用于执行以下操作:
输入装置503通过互联网获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息。
处理器501通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息。
处理器501将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果包括所述属性信息与所述坐标数据匹配,或者所述属性信息与所述坐标数据不匹配。
处理器501根据所述比较结果确定所述属性信息与所述坐标数据匹配时,识别所述属性信息的可信度大于第一可信度阈值。
处理器501根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,识别所述属性信息的可信度小于第二可信度阈值。
在一个实施例中,处理器501根据所述比较结果识别所述属性信息的可信度之后,还可以执行以下操作:
当所述属性信息的可信度小于第三可信度阈值时,处理器501将基于所述地址信息在数字地图中标注的所述兴趣点进行删除。
在一个实施例中,处理器501通过索引算法获取与所述属性信息相关联的坐标数据,具体可以为:
处理器501对所述兴趣点的属性信息进行数据切分,得到一个或多个单元属性信息。
处理器501在信息数据库中查找与所述单元属性信息之间的相似度大于第一比例阈值的目标属性信息,所述目标属性信息包括目标标识信息或者目标地 址信息。
在一个实施例中,处理器501将所述属性信息与所述坐标数据进行比较,得到比较结果,具体可以为:
处理器501获取包含所述兴趣点的兴趣点集合,所述兴趣点集合所包含的每两个兴趣点的标识信息之间的相似度大于第二比例阈值。
处理器501确定所述兴趣点集合所包含的所有兴趣点的数量总和小于数量阈值。
当在所述信息数据库中查找到与所述单元属性信息之间的相似度大于所述第一比例阈值的目标属性信息时,处理器501确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,处理器501通过索引算法获取与所述属性信息相关联的坐标数据,具体可以为:
处理器501基于所述地址信息在所述数字地图中标注所述兴趣点。
处理器501在所述数字地图中获取与所述兴趣点相连接的路网数据,所述路网数据包括道路信息。
在一个实施例中,处理器501将所述属性信息与所述坐标数据进行比较,得到比较结果,具体可以为:
处理器501对所述兴趣点的地址信息进行数据切分,得到一个或多个单元地址信息。
当所述单元地址信息与所述道路信息不匹配时,处理器501确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,处理器501通过索引算法获取与所述属性信息相关联的坐标数据,具体可以为:
处理器501基于所述地址信息在所述数字地图中标注所述兴趣点。
处理器501在所述数字地图中获取与所述兴趣点之间的距离小于第一距离阈值的目标区域。
在一个实施例中,处理器501将所述属性信息与所述坐标数据进行比较,得到比较结果,具体可以为:
处理器501对所述目标区域进行分析处理,获取所述目标区域的流通量。
处理器501在所述数字地图中获取与所述目标区域之间的距离小于第二距离阈值的目标兴趣点的数量总和,所述目标兴趣点的标识信息与所述兴趣点的标识信息之间的相似度大于第二比例阈值。
当所述目标区域的流通量与所述目标兴趣点的数量总和不匹配时,处理器501确定所述属性信息与所述坐标数据不匹配。
在一个实施例中,处理器501将所述属性信息与所述坐标数据进行比较,得到比较结果,具体可以为:
处理器501对所述兴趣点的标识信息进行数据切分,得到一个或多个单元标识信息。
当所述单元标识信息与所述目标区域的属性信息不匹配时,处理器501确定所述属性信息与所述坐标数据不匹配。
具体的,本申请实施例中介绍的终端可以用以实施本申请结合图2介绍的方法实施例中的部分或全部流程。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory,ROM)或随机存储记忆体(Random Access Memory,RAM)等。
图6示出了根据本申请一些实施例的信息识别方法600的示意图。信息识别方法600例如可以在图5所示的计算设备中执行,但不限于此。
如图6所示,在步骤S601中,获取与待验证可信度的兴趣点对应的标识信息。在一个实施例中,步骤S601所获取的标识信息为地址或者名称。
在步骤S602中,根据标识信息确定与兴趣点对应的坐标位置。在一个实施例中,步骤S602可以通过地理位置服务方式确定坐标位置,但不限于此。
在步骤S603中,获取坐标位置的周边数据。在一个实施例中,步骤S603可以获取坐标位置周边所分布的兴趣点集合。另外,步骤S603还可以获取坐标位置所对应的属性项。这里,坐标位置所对应的属性项例如可以包括区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。
在步骤S604中,对标识信息进行解析。在一个实施例中,步骤S604可以 对标识信息进行语义分析,而得到语义分析结果。这里,语义分析方式例如为自然语言处理方式,但不限于此。这样,步骤S604可以根据语音分析结果确定关联数据。换言之,关联数据可以包括与标识信息有关的点线面数据等。例如,步骤S604可以根据语音分析结果,获取与标识信息对应的属性项。属性项例如可以包括标识信息所涉及的区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。换言之,步骤S604可以获取标识信息的文字内容所引用的属性项。
在步骤S605中,对关联数据和周边数据进行特征提取操作,以获取相应的特征数据。在一个实施例中,步骤S605可以确定标识信息对应的位置特征、文本特征、环境特征和属性特征中至少一个。其中,位置特征用于描述所述坐标位置对应的所述属性项与所述标识信息是否一致。例如,位置特征可以包括:引用道路距离,轮廓是否匹配、门址支持数量、门址支持距离、实体支持数量或者实体支持距离等。文本特征用于描述所述标识信息所引用的门牌号、地标和交叉路口中至少一项在所述兴趣点集合中分布特征。环境特征用于描述坐标位置的周边密度和区域地址多样性中至少一个。属性特征用于描述所述标识信息所引用的实体是否唯一和辐射范围中至少一个。
在步骤S606中,基于特征数据,利用机器学习模型确定所述兴趣点的所述坐标位置是否可信。例如,基于特征数据,步骤S606可以利用梯度提升决策树确定所述坐标位置是否可信。另外,步骤S606还可以采用其他机器学习方式确定坐标位置是否可信,本申请对此不做限制。另外说明的是,方法600更具体的实施方式请参见上文中图1B和图3的描述,这里不再赘述。综上,方法600基于将标识信息与坐标位置的周边数据,并利用机器学习方式,可以提高判断数字地图中标注的兴趣点的准确性。
以上所揭露的仅为本申请可选实施例而已,当然不能以此来限定本申请之权利范围,因此依本申请权利要求所作的等同变化,仍属本申请所涵盖的范围。

Claims (29)

  1. 一种信息识别方法,应用于计算设备,所述方法包括:
    获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息;
    通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息;
    将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果用于描述所述属性信息与所述坐标数据是否匹配;
    当根据所述比较结果确定所述属性信息与所述坐标数据匹配时,确定所述属性信息的可信度大于第一可信度阈值;
    当根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,确定所述属性信息的可信度小于第二可信度阈值。
  2. 如权利要求1所述的方法,其特征在于,所述通过索引算法获取与所述属性信息相关联的坐标数据,包括:
    对所述兴趣点的属性信息进行数据切分,得到一个或多个单元属性信息;
    在信息数据库中查找与所述单元属性信息之间的相似度大于第一比例阈值的目标属性信息并将其作为所述坐标数据,所述目标属性信息包括目标标识信息或者目标地址信息。
  3. 如权利要求2所述的方法,其特征在于,所述将所述属性信息与所述坐标数据进行比较,得到比较结果,包括:
    当在所述信息数据库中查找到与所述单元属性信息之间的相似度大于所述第一比例阈值的目标属性信息时,确定所述属性信息与所述坐标数据不匹配。
  4. 如权利要求1所述的方法,其特征在于,所述通过索引算法获取与所述属性信息相关联的坐标数据,包括:
    基于所述地址信息在所述数字地图中标注所述兴趣点;
    在所述数字地图中获取与所述兴趣点相连接的路网数据,所述路网数据包括道路信息。
  5. 如权利要求4所述的方法,其特征在于,所述将所述属性信息与所述坐标数据进行比较,得到比较结果,包括:
    对所述兴趣点的地址信息进行数据切分,得到一个或多个单元地址信息;
    当所述单元地址信息与所述道路信息不匹配时,确定所述属性信息与所述坐标数据不匹配。
  6. 如权利要求1所述的方法,其特征在于,所述通过索引算法获取与所述属性信息相关联的坐标数据,包括:
    基于所述地址信息在所述数字地图中标注所述兴趣点;
    在所述数字地图中获取与所述兴趣点之间的距离小于第一距离阈值的目标区域。
  7. 如权利要求6所述的方法,其特征在于,所述将所述属性信息与所述坐标数据进行比较,得到比较结果,包括:
    对所述目标区域进行分析处理,获取所述目标区域的流通量;
    在所述数字地图中获取与所述目标区域之间的距离小于第二距离阈值的目标兴趣点的数量总和,所述目标兴趣点的标识信息与所述兴趣点的标识信息之间的相似度大于第二比例阈值;
    当所述目标区域的流通量与所述目标兴趣点的数量总和不匹配时,确定所述属性信息与所述坐标数据不匹配。
  8. 一种信息识别方法,应用于计算设备,所述方法包括:
    获取与待验证可信度的兴趣点对应的标识信息;
    根据所述标识信息确定与所述兴趣点对应的坐标位置;
    获取所述坐标位置的周边数据;
    对所述标识信息进行解析,以获取与所述标识信息对应的关联数据;
    对所述关联数据和所述周边数据进行特征提取操作,以获取相应的特征数据;以及
    基于所述特征数据,利用机器学习模型确定所述兴趣点的所述坐标位置是否可信。
  9. 如权利要求8所述的方法,其中,所述获取与待验证可信度的兴趣点对应的标识信息,包括:获取所述兴趣点的名称或者地址。
  10. 如权利要求8所述的方法,其中,所述对所述标识信息进行解析,以获取与所述标识信息对应的关联数据,包括:
    对所述标识信息进行语义分析,而得到语义分析结果;
    根据所述语音分析结果确定所述关联数据。
  11. 如权利要求10所述的方法,其中,所述根据所述语音分析结果确定所述关联数据,包括:
    根据所述语音分析结果,获取与所述标识信息对应的属性项,所述属性项包括标识信息所涉及的区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。
  12. 如权利要求8所述的方法,其中,所述获取所述坐标位置的周边数据,包括:
    获取所述坐标位置周边所分布的兴趣点集合;
    获取所述坐标位置所对应的属性项,所述坐标位置所对应的属性项包括区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。
  13. 如权利要求12所述的方法,其中,所述对所述关联数据和所述周边数据进行特征提取操作,以获取相应的特征数据,包括:
    确定所述标识信息对应的位置特征、文本特征、环境特征和属性特征中至少一个,其中,所述位置特征用于描述所述坐标位置对应的所述属性项与所述标识信息是否一致,所述文本特征用于描述所述标识信息所引用的门牌号、地标和交叉路口中至少一项在所述兴趣点集合中分布特征,所述环境特征用于描述所述坐标位置的周边密度和区域地址多样性中至少一个,所述属性特征用于描述所述标识信息所引用的实体是否唯一和辐射范围中至少一个。
  14. 如权利要求8所述的方法,其中,所述基于所述特征数据,利用机器学习模型确定所述兴趣点的坐标位置是否可信,包括:基于所述特征数据,利用梯度提升决策树确定所述坐标位置是否可信。
  15. 一种计算设备,包括:处理器和存储器;所述存储器中存储有计算机可读指令,可以使所述处理器:
    获取兴趣点的属性信息,所述属性信息包括所述兴趣点的标识信息或者地址信息;
    通过索引算法获取与所述属性信息相关联的坐标数据,所述坐标数据用于指示至少一个维度的关联信息;
    将所述属性信息与所述坐标数据进行比较,得到比较结果,所述比较结果用于描述所述属性信息与所述坐标数据是否匹配;
    当根据所述比较结果确定所述属性信息与所述坐标数据匹配时,确定所述属性信息的可信度大于第一可信度阈值;
    当根据所述比较结果确定所述属性信息与所述坐标数据不匹配时,确定所述属性信息的可信度小于第二可信度阈值。
  16. 如权利要求15所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    对所述兴趣点的属性信息进行数据切分,得到一个或多个单元属性信息;
    在信息数据库中查找与所述单元属性信息之间的相似度大于第一比例阈值的目标属性信息并将其作为所述坐标数据,所述目标属性信息包括目标标识信息或者目标地址信息。
  17. 如权利要求16所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    当在所述信息数据库中查找到与所述单元属性信息之间的相似度大于所述第一比例阈值的目标属性信息时,确定所述属性信息与所述坐标数据不匹配。
  18. 如权利要求15所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    基于所述地址信息在所述数字地图中标注所述兴趣点;
    在所述数字地图中获取与所述兴趣点相连接的路网数据,所述路网数据包括道路信息。
  19. 如权利要18所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    对所述兴趣点的地址信息进行数据切分,得到一个或多个单元地址信息;
    当所述单元地址信息与所述道路信息不匹配时,确定所述属性信息与所述坐标数据不匹配。
  20. 如权利要求15所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    基于所述地址信息在所述数字地图中标注所述兴趣点;
    在所述数字地图中获取与所述兴趣点之间的距离小于第一距离阈值的目标区域。
  21. 如权利要求20所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    对所述目标区域进行分析处理,获取所述目标区域的流通量;
    在所述数字地图中获取与所述目标区域之间的距离小于第二距离阈值的目标兴趣点的数量总和,所述目标兴趣点的标识信息与所述兴趣点的标识信息之间的相似度大于第二比例阈值;
    当所述目标区域的流通量与所述目标兴趣点的数量总和不匹配时,确定所述属性信息与所述坐标数据不匹配。
  22. 一种计算设备,包括:处理器和存储器;所述存储器中存储有计算机可读指令,可以使所述处理器:
    获取与待验证可信度的兴趣点对应的标识信息;
    根据所述标识信息确定与所述兴趣点对应的坐标位置;
    获取所述坐标位置的周边数据;
    对所述标识信息进行解析,以获取与所述标识信息对应的关联数据;
    对所述关联数据和所述周边数据进行特征提取操作,以获取相应的特征数据;以及
    基于所述特征数据,利用机器学习模型确定所述兴趣点的所述坐标位置是否可信。
  23. 如权利要求22所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:获取所述兴趣点的名称或者地址。
  24. 如权利要求22所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    对所述标识信息进行语义分析,而得到语义分析结果;
    根据所述语音分析结果确定所述关联数据。
  25. 如权利要求24所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    根据所述语音分析结果,获取与所述标识信息对应的属性项,所述属性项 包括标识信息所涉及的区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。
  26. 如权利要求22所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    获取所述坐标位置周边所分布的兴趣点集合;
    获取所述坐标位置所对应的属性项,所述坐标位置所对应的属性项包括区域轮廓、道路、实体、门牌号、地标和交叉路口中至少一项。
  27. 如权利要求26所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:
    确定所述标识信息对应的位置特征、文本特征、环境特征和属性特征中至少一个,其中,所述位置特征用于描述所述坐标位置对应的所述属性项与所述标识信息是否一致,所述文本特征用于描述所述标识信息所引用的门牌号、地标和交叉路口中至少一项在所述兴趣点集合中分布特征,所述环境特征用于描述所述坐标位置的周边密度和区域地址多样性中至少一个,所述属性特征用于描述所述标识信息所引用的实体是否唯一和辐射范围中至少一个。
  28. 如权利要求22所述的计算设备,其中,所述处理器进一步执行所述计算机可读指令,用于:基于所述特征数据,利用梯度提升决策树确定所述坐标位置是否可信。
  29. 一种非易失性存储介质,存储有一个或多个程序,所述一个或多个程序包括指令,所述指令当由计算设备执行时,使得所述计算设备执行权利要求1-14中任一项所述方法的指令。
PCT/CN2018/080822 2017-03-29 2018-03-28 信息识别方法、计算设备及存储介质 WO2018177316A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710198353.0A CN108304423B (zh) 2017-03-29 2017-03-29 一种信息识别方法及装置
CN201710198353.0 2017-03-29

Publications (1)

Publication Number Publication Date
WO2018177316A1 true WO2018177316A1 (zh) 2018-10-04

Family

ID=62872097

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/080822 WO2018177316A1 (zh) 2017-03-29 2018-03-28 信息识别方法、计算设备及存储介质

Country Status (2)

Country Link
CN (1) CN108304423B (zh)
WO (1) WO2018177316A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110457420A (zh) * 2019-08-13 2019-11-15 腾讯云计算(北京)有限责任公司 兴趣点位置识别方法、装置、设备及存储介质
CN111209354A (zh) * 2018-11-22 2020-05-29 北京搜狗科技发展有限公司 一种地图兴趣点判重的方法、装置及电子设备
CN111324679A (zh) * 2018-12-14 2020-06-23 阿里巴巴集团控股有限公司 地址信息的处理方法、装置和系统
CN112381162A (zh) * 2020-11-19 2021-02-19 北京百度网讯科技有限公司 信息点识别方法、装置及电子设备
CN112948517A (zh) * 2021-02-26 2021-06-11 北京百度网讯科技有限公司 区域位置标定方法、装置及电子设备
CN113723405A (zh) * 2021-08-31 2021-11-30 北京百度网讯科技有限公司 区域轮廓的确定方法、装置和电子设备
CN114896363A (zh) * 2022-04-19 2022-08-12 北京月新时代科技股份有限公司 一种数据管理方法、装置、设备及介质
CN117112587A (zh) * 2023-10-19 2023-11-24 腾讯科技(深圳)有限公司 地图数据处理方法、装置、介质及设备

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109344209A (zh) * 2018-08-20 2019-02-15 中国平安人寿保险股份有限公司 一种基于大数据的地址测试方法及终端设备
CN109558874B (zh) * 2018-12-11 2024-05-31 上海集成电路研发中心有限公司 一种基于图像识别的定位方法及装置
CN110390279A (zh) * 2019-07-08 2019-10-29 丰图科技(深圳)有限公司 坐标识别方法、装置、设备、及计算机可读存储介质
CN110413904A (zh) * 2019-07-25 2019-11-05 北京百度网讯科技有限公司 一种兴趣点地址数据处理方法、装置、服务器和介质
CN113282690B (zh) * 2020-02-19 2024-04-02 百度在线网络技术(北京)有限公司 兴趣点召回的排序方法、装置、设备和存储介质
CN111797183A (zh) * 2020-05-29 2020-10-20 汉海信息技术(上海)有限公司 挖掘信息点的道路属性的方法、装置及电子设备
CN111767478B (zh) * 2020-06-22 2023-08-15 北京百度网讯科技有限公司 一种关联关系构建方法、装置、设备及存储介质
CN112836472A (zh) * 2021-02-18 2021-05-25 中国城市规划设计研究院 一种地址批注方法、装置、设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015119371A1 (ko) * 2014-02-05 2015-08-13 에스케이플래닛 주식회사 Poi 그룹화를 이용한 poi 정보 제공 장치 및 방법
CN104866542A (zh) * 2015-05-05 2015-08-26 腾讯科技(深圳)有限公司 一种poi数据验证方法和装置
CN106126719A (zh) * 2016-06-30 2016-11-16 百度在线网络技术(北京)有限公司 信息处理方法及装置
US9529857B1 (en) * 2014-02-03 2016-12-27 Google Inc. Disambiguation of place geometry

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102867004B (zh) * 2011-07-06 2016-06-29 高德软件有限公司 一种地址匹配的方法及设备
CN102841920B (zh) * 2012-06-30 2017-05-10 北京百度网讯科技有限公司 一种页面信息提取方法及装置
CN105468632B (zh) * 2014-09-05 2019-08-09 高德软件有限公司 一种地理编码方法及装置
CN104572902B (zh) * 2014-12-26 2018-01-23 北京中交兴路车联网科技有限公司 一种信息点匹配的方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9529857B1 (en) * 2014-02-03 2016-12-27 Google Inc. Disambiguation of place geometry
WO2015119371A1 (ko) * 2014-02-05 2015-08-13 에스케이플래닛 주식회사 Poi 그룹화를 이용한 poi 정보 제공 장치 및 방법
CN104866542A (zh) * 2015-05-05 2015-08-26 腾讯科技(深圳)有限公司 一种poi数据验证方法和装置
CN106126719A (zh) * 2016-06-30 2016-11-16 百度在线网络技术(北京)有限公司 信息处理方法及装置

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111209354A (zh) * 2018-11-22 2020-05-29 北京搜狗科技发展有限公司 一种地图兴趣点判重的方法、装置及电子设备
CN111324679B (zh) * 2018-12-14 2023-04-11 阿里巴巴集团控股有限公司 地址信息的处理方法、装置和系统
CN111324679A (zh) * 2018-12-14 2020-06-23 阿里巴巴集团控股有限公司 地址信息的处理方法、装置和系统
CN110457420B (zh) * 2019-08-13 2024-04-16 腾讯云计算(北京)有限责任公司 兴趣点位置识别方法、装置、设备及存储介质
CN110457420A (zh) * 2019-08-13 2019-11-15 腾讯云计算(北京)有限责任公司 兴趣点位置识别方法、装置、设备及存储介质
CN112381162A (zh) * 2020-11-19 2021-02-19 北京百度网讯科技有限公司 信息点识别方法、装置及电子设备
CN112381162B (zh) * 2020-11-19 2024-05-07 北京百度网讯科技有限公司 信息点识别方法、装置及电子设备
CN112948517B (zh) * 2021-02-26 2023-06-23 北京百度网讯科技有限公司 区域位置标定方法、装置及电子设备
CN112948517A (zh) * 2021-02-26 2021-06-11 北京百度网讯科技有限公司 区域位置标定方法、装置及电子设备
CN113723405A (zh) * 2021-08-31 2021-11-30 北京百度网讯科技有限公司 区域轮廓的确定方法、装置和电子设备
CN114896363B (zh) * 2022-04-19 2023-03-28 北京月新时代科技股份有限公司 一种数据管理方法、装置、设备及介质
CN114896363A (zh) * 2022-04-19 2022-08-12 北京月新时代科技股份有限公司 一种数据管理方法、装置、设备及介质
CN117112587A (zh) * 2023-10-19 2023-11-24 腾讯科技(深圳)有限公司 地图数据处理方法、装置、介质及设备
CN117112587B (zh) * 2023-10-19 2024-06-18 腾讯科技(深圳)有限公司 地图数据处理方法、装置、介质及设备

Also Published As

Publication number Publication date
CN108304423B (zh) 2021-09-28
CN108304423A (zh) 2018-07-20

Similar Documents

Publication Publication Date Title
WO2018177316A1 (zh) 信息识别方法、计算设备及存储介质
CN107656913B (zh) 地图兴趣点地址提取方法、装置、服务器和存储介质
US11698261B2 (en) Method, apparatus, computer device and storage medium for determining POI alias
EP3153978B1 (en) Address search method and device
CN110472066B (zh) 一种城市地理语义知识图谱的构建方法
US9551586B2 (en) System and method for providing contextual information for a location
CN108388559B (zh) 地理空间应用下的命名实体识别方法及系统、计算机程序
US10789078B2 (en) Method and system for inputting information
US8811656B2 (en) Selecting representative images for establishments
CN110020433B (zh) 一种基于企业关联关系的工商高管人名消歧方法
WO2022198854A1 (zh) 多模态poi特征的提取方法和装置
CN108628811B (zh) 地址文本的匹配方法和装置
US9251395B1 (en) Providing resources to users in a social network system
CN106033416A (zh) 一种字符串处理方法及装置
US20160371840A1 (en) Determination of a geographical location of a user
US20150186455A1 (en) Systems and methods for automatic electronic message annotation
CN111488468B (zh) 地理信息知识点抽取方法、装置、存储介质及计算机设备
TW201933879A (zh) 內容推薦方法及裝置
CN110019617B (zh) 地址标识的确定方法和装置、存储介质、电子装置
WO2022100154A1 (zh) 基于人工智能的地址标准化方法、装置、设备和存储介质
WO2019227581A1 (zh) 兴趣点识别方法、装置、终端设备及存储介质
CN116414823A (zh) 一种基于分词模型的地址定位方法和装置
CN110990651B (zh) 地址数据处理方法、装置、电子设备及计算机可读介质
CN110647595B (zh) 一种新增兴趣点的确定方法、装置、设备和介质
US11347820B2 (en) Facilitating identification of an intended country associated with a query

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18778122

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18778122

Country of ref document: EP

Kind code of ref document: A1