Method for determining navigation address, data processing apparatus, and storage medium
By identifying the POI address in the navigation command, using address dataset and map matching technology, and combining similarity filtering to determine the navigation address, the problem of inaccurate address recognition in the navigation system is solved, and navigation accuracy and user experience are improved.
Patent Information
- Application Number
- PCT/CN2024/095612
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-31
- Filing Date
- 2024-05-27
- Publication Date
- 2024-12-05
AI Technical Summary
There are accuracy problems when identifying navigation addresses in existing navigation systems. The address entered by the user may be inconsistent with the actual destination, resulting in inaccurate navigation.
By identifying the POI address in the navigation command, searching similar addresses and address types based on the address dataset, determining navigation addresses based on similarity matching, improving the accuracy of address types, and determining navigation addresses in the map, using thresholds to filter out high similarity address.
Improves navigation accuracy and user experience, ensures that the navigation address is closer to the user's intentions, and reduces the execution of invalid addresses.
Smart Images

Figure CN2024095612_05122024_PF_FP_ABST
Abstract
Description
Method for determining navigation address, data processing device, and storage medium
[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on May 31, 2023, with application number 202310632874.8 and application name “A method for determining a navigation address and related devices”, the entire contents of which are incorporated by reference into this application. Technical Field
[0002] The present application relates to the field of information technology, and in particular to a method for determining a navigation address, a data processing device, and a storage medium. Background Art
[0003] With advances in industrial technology and improvements in living standards, automobiles, as a means of transportation, are playing an increasingly important role in our daily lives. Along with the widespread use of automobiles, in-vehicle navigation systems, as a common automotive electronic product, have also experienced rapid growth. Because navigation systems can plan and display driving routes, people are increasingly relying on them while driving.
[0004] Navigation is the process of guiding and monitoring a device or user from one place to another, and navigation address recognition is a key aspect of navigation technology. Currently, when performing navigation, the industry typically identifies the entity to be navigated to and then enters that entity into a map for navigation. Because navigation commands are user-entered, and locations on a map can be named in various styles, it is very likely that the navigation address identified on the map, while identical or similar to the entity in the command, is not actually the user's intended destination, significantly reducing navigation accuracy.
[0005] Summary of the Invention
[0006] The embodiments of the present application provide a method for determining a navigation address, a data processing device, and a storage medium, which can improve the accuracy of navigation and thereby enhance the user experience.
[0007] In a first aspect, an embodiment of the present application provides a method for determining a navigation address, the method comprising:
[0008] Identify the first point of interest (POI) address in the navigation command;
[0009] According to the first POI address, retrieve N similar POI addresses and address types of the N similar POI addresses from an address dataset, wherein each of the N similar POI addresses is similar to the first POI address, and N is an integer greater than or equal to 1;
[0010] Determining a first address type according to the address types of the N similar POI addresses; and
[0011] A navigation address is determined in a map according to the first POI address and the first address type.
[0012] For example, the value of N can be an integer such as 1, 2, 3, or 4.
[0013] The embodiment of the present application can identify a first POI address in a navigation command and determine a first address type that the first POI address may correspond to based on an address dataset. When determining the navigation address corresponding to the POI address on a map, the first address type can be combined to obtain a more likely navigation address, that is, addresses of less likely address types can be filtered out, making the determined navigation address more accurate, thereby improving navigation accuracy and enhancing the user experience.
[0014] In some implementations of the first aspect, identifying the first POI address in the navigation command includes:
[0015] Recognizing the navigation command input by the user through a navigation entity recognition model to obtain entities in the navigation command; and
[0016] The first POI address is determined according to the entity in the navigation command.
[0017] This implementation can perform entity recognition on the command input by the user and re-analyze the entity to obtain the POI address.
[0018] In some implementations of the first aspect, the navigation entity recognition model is trained based on part or all of the POI addresses in the address dataset.
[0019] In this way, the recognition capability of the navigation entity recognition model can be improved, the accuracy of navigation can be improved, and the user experience can be improved.
[0020] In some implementations of the first aspect, the address dataset includes multiple POI addresses and address types corresponding to the multiple POI addresses; the POI addresses in the address dataset are determined based on at least one of the following: multiple navigation intentions or multiple real addresses.
[0021] In some implementations of the first aspect, the plurality of navigation intentions may be destinations of navigation processes accumulated on the navigation line, and the plurality of real addresses may be common addresses nationwide in address data disclosed online.
[0022] In this way, the company's data resources can be effectively utilized to build a dedicated address data set, improve the accuracy of the address type corresponding to the address data, and enhance the user experience.
[0023] In some implementations of the first aspect, determining the first address type according to the address types of the N similar POI addresses includes:
[0024] Calculating a first similarity between each of the N similar POI addresses and the first POI address; and
[0025] The first address type is determined according to the first similarity and the address type of each similar POI address of the N similar POI addresses.
[0026] In this embodiment, the N similar POI addresses can be obtained by fuzzy retrieval based on the first POI address. When performing the fuzzy retrieval, the fuzzy retrieval of similarity can be performed based on semantics, text, voice, etc. For the N similar POI addresses obtained by the fuzzy retrieval, the similarity can be calculated again to obtain the similarity with the first POI address, and the first address type can be determined based on the address type of the similar POI addresses with high similarity. In other words, in the process of determining the first address type, the similar POI addresses are matched twice for similarity, so that the first address type is closer to the address type of the first POI address. In this way, the address type can be determined more accurately and quickly, the accuracy of navigation can be improved, and the user experience can be enhanced.
[0027] In some implementations of the first aspect, determining the first address type based on the first similarity and address type of each of the N similar POI addresses includes: determining the address type of a similar POI address among the N similar POI addresses whose first similarity is greater than or equal to a first threshold as the first address type.
[0028] In some implementations of the first aspect, determining the first address type based on the first similarity and address type of each of the N similar POI addresses includes: determining the address type of a similar POI address with the greatest first similarity among the N similar POI addresses as the first address type.
[0029] In some implementations of the first aspect, determining the first address type according to the address types of the N similar POI addresses includes: determining the address type of the similar POI address with the largest number of occurrences among the N similar POI addresses as the first address type.
[0030] In some embodiments of the first aspect, the first threshold is used to indicate a condition that must be met for the similarity between the similar POI address and the first POI address. The first threshold is predefined or pre-specified. For example, the first threshold may be a value greater than 60%, such as 75%, 80%, or the like.
[0031] By setting the first threshold, the address type of similar POI addresses that are closer to the first POI address can be screened out from the determined similar POI addresses, thereby determining the address type more accurately and quickly.
[0032] In some implementations of the first aspect, determining a navigation address in a map according to the first POI address and the first address type includes:
[0033] Determining, in a map, M map POI addresses according to the first POI address and the first address type, wherein the M map POI addresses have corresponding real addresses in the map, and M is an integer greater than or equal to 1;
[0034] determining a target address according to a second similarity between each of the M map POI addresses and the first POI address; and
[0035] If the second similarity between the target address and the first POI address is greater than a second threshold, the target address is used as the navigation address.
[0036] The value of M can be an integer such as 1, 2, 3, 4, etc. and M≥1.
[0037] In this embodiment, the M map POI addresses can be obtained by searching the map based on the first POI address and the first address type, and similarity searches can be performed based on semantics, text, etc. For the M map POI addresses obtained by the search, similarity calculations can be performed again to obtain their similarity with the first POI address, and the navigation address can be determined based on the map POI addresses with high similarity. In other words, in the process of determining the navigation address, the map POI addresses are matched for similarity, so that the determined navigation address is closer to the first POI address. In this way, the navigation address can be determined more accurately and quickly, which improves the accuracy of navigation and thus improves the user experience.
[0038] In some embodiments of the first aspect, the second threshold is used to indicate a condition that must be met for the degree of similarity between the map POI address and the first POI address. The second threshold is predefined or pre-specified. For example, the second threshold may be a value greater than 75%, such as 80%, 90%, or the like.
[0039] By setting the second threshold, a map POI address that is closer to the first POI address can be screened out from the determined map POI addresses, thereby determining the navigation address more accurately and quickly.
[0040] In some implementations of the first aspect, determining the target address based on the second similarity between each of the M map POI addresses and the first POI address includes: determining the map POI address with the greatest second similarity among the M map POI addresses as the target address.
[0041] In some implementations of the first aspect, the navigation address is used to generate a navigation route to the navigation address.
[0042] In this way, navigation commands can be executed more accurately and intelligently.
[0043] In some embodiments of the first aspect, the method further comprises:
[0044] If the second similarity between the target address and the first POI address is less than the second threshold, a prompt message is output, wherein the prompt message is used to instruct the user to input a POI address or to indicate that the first POI address is an invalid address.
[0045] In this way, the execution of invalid addresses can be effectively avoided, improving the user experience.
[0046] In a second aspect, an embodiment of the present application provides a data processing device, the data processing device comprising a recognition unit and a processing unit. The recognition unit is configured to recognize a first POI address in a navigation command.
[0047] The processing unit is configured to:
[0048] According to the first POI address, retrieve N similar POI addresses and address types of the N similar POI addresses from an address dataset, wherein the N similar POI addresses are similar to the first POI address, and N is an integer greater than or equal to 1;
[0049] Determining a first address type according to the address types of the N similar POI addresses; and
[0050] A navigation address is determined in a map according to the first POI address and the first address type.
[0051] In some implementations of the second aspect, the recognition unit is further configured to recognize the navigation command input by the user through a navigation entity recognition model to obtain an entity in the navigation command.
[0052] In some implementations of the second aspect, the processing unit is further configured to determine the first POI address according to the entity in the navigation command.
[0053] In some implementations of the second aspect, the navigation entity recognition model is trained based on part or all of the POI addresses in the address dataset.
[0054] In some implementations of the second aspect, the address dataset includes a plurality of POI addresses and address types corresponding to the plurality of POI addresses; and
[0055] The POI addresses in the address data set are determined according to at least one of the following: multiple navigation intentions or multiple real addresses.
[0056] In some implementations of the second aspect, in determining the first address type according to the address types of the N similar POI addresses, the processing unit is configured to:
[0057] Calculating a first similarity between each of the N similar POI addresses and the first POI address; and
[0058] The first address type is determined according to the first similarity and the address type of each similar POI address of the N similar POI addresses.
[0059] In some embodiments of the second aspect, in determining the first address type based on the first similarity and address type of each of the N similar POI addresses, the processing unit is configured to: determine the address type of a similar POI address among the N similar POI addresses whose first similarity is greater than or equal to a first threshold as the first address type.
[0060] In some embodiments of the second aspect, in determining the first address type based on the first similarity and address type of each of the N similar POI addresses, the processing unit is configured to: determine the address type of a similar POI address with the greatest first similarity among the N similar POI addresses as the first address type.
[0061] In some implementations of the second aspect, in determining the first address type based on the address types of the N similar POI addresses, the processing unit is configured to: determine the address type of the similar POI address whose address type appears the most times among the N similar POI addresses as the first address type.
[0062] In some implementations of the second aspect, in determining a navigation address in a map according to the first POI address and the first address type, the processing unit is configured to:
[0063] Determining, in a map, M map POI addresses according to the first POI address and the first address type, wherein the M map POI addresses have corresponding real addresses in the map, and M is an integer greater than or equal to 1;
[0064] determining a target address according to a second similarity between each of the M map POI addresses and the first POI address; and
[0065] If the second similarity between the target address and the first POI address is greater than a second threshold, the target address is used as the navigation address.
[0066] In some embodiments of the second aspect, in determining the target address based on the second similarity between each of the M map POI addresses and the first POI address, the processing unit is configured to: determine the map POI address with the greatest second similarity among the M map POI addresses as the target address.
[0067] In some implementations of the second aspect, the navigation address is used to generate a navigation route to the navigation address.
[0068] In some embodiments of the second aspect, the processing unit is further configured to output a prompt message if the second similarity between the target address and the first POI address is less than the second threshold, wherein the prompt message is configured to instruct the user to enter a POI address or to indicate that the first POI address is an invalid address.
[0069] In a third aspect, an embodiment of the present application provides a data processing device, which includes a processor and a memory; the processor executes instructions stored in the memory so that the data processing device implements the method described in any one of the first aspects above.
[0070] In some embodiments of the first aspect, the data processing device further includes a communication interface, wherein the communication interface is used to receive and / or send data, and / or the communication interface is used to provide input and / or output for the processor.
[0071] It should be noted that the above embodiments are described using a processor (or general-purpose processor) that executes a method by calling a computer instruction. In specific implementations, the processor may also be a dedicated processor, in which case the computer instructions are pre-loaded into the processor. In some embodiments of the first aspect, the processor may include both a dedicated processor and a general-purpose processor.
[0072] In some implementations of the first aspect, the processor and the memory may also be integrated into one device, that is, the processor and the memory may also be integrated together.
[0073] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions. When the instructions are executed by a data processing device, the data processing device implements the method described in any one of the first aspects above.
[0074] In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the instructions are executed by a data processing device, the data processing device implements the method described in any one of the first aspects above.
[0075] In some embodiments of the first aspect, the computer program product may be a software installation package or an image file. When the aforementioned method is required, the computer program product may be obtained and executed on a computing device.
[0076] The beneficial effects of the technical solutions provided in the second to fifth aspects of this application can refer to the beneficial effects of the technical solution in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0077] The following is a brief introduction to the drawings used in describing the embodiments.
[0078] FIG1 is a schematic diagram of the architecture of a system for determining a navigation address provided in an embodiment of the present application.
[0079] FIG2 is a flow chart of a method for determining a navigation address provided in an embodiment of the present application.
[0080] FIG3 is a flow chart of a method for determining a navigation address provided in another embodiment of the present application.
[0081] FIG4 is a schematic structural diagram of a data processing device provided in an embodiment of the present application.
[0082] FIG5 is a schematic structural diagram of a data processing device provided in another embodiment of the present application. DETAILED DESCRIPTION
[0083] The embodiments of the present application are described in detail below with reference to the accompanying drawings.
[0084] The following is an illustrative explanation of some technical terms used in the embodiments of this application.
[0085] 1. Point of Interest (POI): POI, also known as a tag of interest, is a term used in geographic information systems (GIS) to refer to geographical objects that can be abstracted as points, particularly those geographical entities closely related to people's lives, such as schools, banks, restaurants, gas stations, hospitals, and supermarkets. POIs are primarily used to describe the addresses of objects or events, significantly enhancing the ability to describe and query their locations, and improving the accuracy and speed of geolocation.
[0086] The following is an introduction to the system architecture used in the embodiments of the present application. It should be noted that the system architecture and business scenarios described in this application are intended to more clearly illustrate the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. It is known to those skilled in the art that with the evolution of the system architecture and the emergence of new business scenarios, the technical solutions provided in this application are also applicable to similar technical problems.
[0087] Please refer to Figure 1, which is a schematic diagram of the architecture of a navigation system provided in an embodiment of the present application. The navigation system includes a terminal device 101 and a data processing device 102. It should be noted that the navigation system may include one or more terminal devices and one or more data processing devices, and the embodiment shown in Figure 1 is merely an example. The terminal device and data processing device are described below in an exemplary manner.
[0088] The terminal device 101 is a device that has data processing capabilities and can interact with users. In some embodiments, the terminal device can interact with the user, such as presenting a user interface, or receiving user input operations, etc. Exemplarily, the terminal device includes but is not limited to handheld terminals, wearable devices, entertainment devices, transportation equipment, etc. Among them, handheld devices are such as mobile phones, tablet computers, laptops, or police communications, etc. Wearable devices are such as smart bracelets, smart watches, or smart glasses, etc. Transportation equipment is such as vehicles, ships, aircraft, rail transportation (such as subways, high-speed railways, etc.), or logistics robots, etc. It should be understood that although Figure 1 describes the terminal device as an example, the present application is also applicable to other devices that have the need to determine the navigation address.
[0089] In some embodiments, the terminal device being able to interact with the user may include at least one of the following: the terminal device outputting information to the user, or the terminal device receiving information input by the user. For example, the terminal device 101 may receive a navigation command input by the user, and the terminal device 101 may also present a prompt to the user to instruct the user to enter a POI address or to indicate that a particular POI address is invalid.
[0090] The data processing device 102 is a device with data processing capabilities. The data processing device can be a physical device, such as a server, cloud, or host, or a virtual device, such as a virtual machine, software, program code, or container. In some embodiments, the data processing device 102 can be integrated into the terminal device 101.
[0091] In some implementations, the data processing device 102 may identify a first POI address in a navigation command and determine a first address type that may correspond to the first POI address based on the address dataset. When determining a navigation address corresponding to the POI address on a map, the first address type may be combined to obtain a more likely navigation address. This may include filtering out addresses of less likely address types, making the determined navigation address more accurate, thereby improving navigation accuracy and enhancing the user experience.
[0092] In some implementations, the navigation command may be provided by the terminal device 101 to the data processing device 102. Specifically, the terminal device 101 and the data processing device 102 have communication capabilities, and a communication connection may exist between the terminal device 101 and the data processing device 102. For example, the terminal device 101 may communicate with the data processing device 102, sending information to the data processing device 102 and / or receiving information from the data processing device 102. For example, the terminal device 101 may send a received user-entered navigation command or an entity or POI address obtained based on the navigation command to the data processing device 102, and the data processing device 102 may return a navigation address or a navigation route to the terminal device 101.
[0093] In some implementations, the terminal device 101 and the data processing device 102 may be directly connected via a wired communication method, such as a tangible medium such as a metal wire or an optical fiber, or may be indirectly connected via a wireless communication method, such as ultra-wideband (UWB) technology, long term evolution (Long Term Evolution) communication technology, fifth generation mobile communication technology (5th generation mobile networks or 5th generation wireless systems, 5th-Generation, 5G or 5G technology for short), global system for mobile communications (GSM), general packet radio service (GPRS), or universal mobile telecommunications system (UMTS).
[0094] The following is a detailed description of the method for determining a navigation address provided in an embodiment of the present application.
[0095] Please refer to Figure 2, which is a flowchart illustrating a method for determining a navigation address provided in an embodiment of the present application. As shown in Figure 2, the method for determining a navigation address includes operations S201-S204. This method can be applied to the navigation system shown in Figure 1; optionally, the method can also be applied to the data processing device 102 in the navigation system shown in Figure 1.
[0096] S201: The data processing device identifies a first POI address in a navigation command.
[0097] A data processing device is a device with data processing and communication capabilities, as described above. A navigation command refers to a command or information input by a user indicating a user's navigation needs. In some embodiments, the user may input a navigation command through voice input, keyboard input, or other means. In some embodiments, the data processing device may directly receive the user's input navigation command, or a terminal device may receive the user's input navigation command and submit it to the data processing device.
[0098] The POI address may be an address indicating a geographical entity of interest. For example, the POI address may be an aunt's home, XX train station, scenic spot A, scenic spot B, or user A's home.
[0099] In some embodiments, a data processing device identifies a navigation command input by a user to obtain an entity in the navigation command, and the data processing device determines the first POI address based on the entity in the navigation command. For example, the navigation command input by the user may be "I want to go to XX train station," where the entities in the navigation command include "I" and "XX train station." Since "XX train station" is an address, "XX train station" is determined as the first POI address.
[0100] In some embodiments, identifying the navigation command input by the user to obtain the entity in the navigation command may be achieved by using a navigation entity recognition model, wherein the navigation entity model has an address entity recognition capability.
[0101] In some embodiments, the navigation entity recognition model is trained based on some or all of the POI addresses in the address dataset. In some embodiments, the navigation entity recognition model can be trained by injecting some or all of the POI addresses into a natural language processing (NLP) entity recognition model. Exemplarily, the address dataset contains multiple POI addresses and address types corresponding to the multiple POI addresses, and the POI addresses in the address dataset are determined based on multiple navigation intentions, multiple real addresses, etc. In some embodiments, the address dataset may also include the actual location corresponding to the POI address, such as longitude and latitude, map tile identification (ID), etc.
[0102] The multiple navigation intents can be the destinations of the navigation process accumulated on the navigation line. For example, in the navigation process of "Go to XX Community", the navigation intent can be "XX Community". Another example is that in the navigation process of "Go to West Lake", the navigation intent can be "West Lake". The multiple real addresses can be common addresses across the country from the address data publicly available online, such as scenic area addresses (scenic spots), hotel addresses, aliases (which can be nicknames of locals), shopping mall addresses, and transportation facility addresses (train stations, subway stations, bus stops, etc.).
[0103] In some embodiments, the data processing device can construct a Faiss (Facebook AI Similarity Search) search library based on some or all POI addresses in the address dataset. This allows for the establishment of an address resource database, a dedicated address dataset, and searches based on the Faiss search library to improve navigation accuracy. Faiss is a clustering and similarity search library that provides efficient similarity search and clustering for dense vectors, supporting searches of billions of vectors.
[0104] S202: The data processing device retrieves N similar POI addresses and address types of the N similar POI addresses from the address data set according to the first POI address.
[0105] Wherein, N is an integer and N≥1. For example, N is equal to 1. For another example, N≥2, in this case, there are multiple similar POI addresses.
[0106] Similar POI addresses are addresses obtained through fuzzy retrieval in the address dataset and may be similar to the first POI address in terms of text, phonetics, semantics, etc. For example, taking textual similarity as an example, a POI address similar to "XX Station" could be "XX East Station," "XX Railway Station," or "XXX High-Speed Railway Station," etc. Taking phonetic similarity as an example, a POI address similar to "Xunli Gate" could be "Xuli Gate," "Xunli Gate," or "Xunli Gate," etc. Taking semantic similarity as an example, a POI address similar to "Window of the World" could be "World Window," "World Window," etc.
[0107] Address type refers to the category of the address. For example, address types may include catering services (such as hotels, restaurants, and tea shops), shopping services (such as shopping malls, pedestrian streets, and comprehensive markets), lifestyle services (travel agencies, business halls, and ticket offices), commercial residences (residential communities and industrial parks), companies (companies, factories, and bases), transportation facilities (bus stations, train stations, and subway stations), scenic spots, and healthcare services (hospitals, clinics, and emergency centers).
[0108] Exemplarily, Table 1 shows each similar POI address and the address type of each similar POI address.
[0109] Table 1 Similar POI addresses and address types of similar POI addresses
[0110] Exemplarily, the data processing device searches for the first POI address in the Faiss search database, and returns five similar POI addresses and the address types of the five similar POI addresses.
[0111] S203: The data processing device determines a first address type according to the address types of N similar POI addresses.
[0112] Specifically, the first address type belongs to one of N address types of similar POI addresses, where N is an integer and N≥1.
[0113] When N=1, the first address type is the address type of the similar POI address.
[0114] In some embodiments, the data processing device calculates a similarity between the similar POI address and the first POI address. If the first similarity between the similar POI address and the first POI address is greater than or equal to a first threshold, the first address type is the address type of the similar POI address. Conversely, if the first similarity between the similar POI address and the first POI address is less than the first threshold, the data processing device outputs a prompt message, the prompt message being used to instruct the user to enter a POI address or to indicate that the first POI address is invalid.
[0115] When N≥2, the data processing device may determine the first address type in the following ways.
[0116] Mode 1: The first address type is an address type corresponding to a POI address whose similarity to the first POI address is greater than or equal to a first threshold among the N similar POI addresses.
[0117] In some embodiments, the data processing device calculates similarity between N similar POI addresses and the first POI address to obtain a first similarity for each of the N similar POI addresses. Furthermore, the data processing device determines the first address type based on the address types of similar POI addresses among the N similar POI addresses whose first similarity is greater than or equal to a first threshold. This allows for more accurate and rapid determination of address types, improving navigation accuracy and, consequently, enhancing the user experience.
[0118] Illustratively, the first similarity may reflect the textual and semantic similarity between the similar POI address and the first POI address.
[0119] For example, the first threshold value is used to indicate the condition that must be met for the similarity between the similar POI address and the first POI address. The first threshold value is predefined or pre-specified. For example, if the first threshold value is 60%, then the first similarity greater than or equal to the first threshold value may be 60%, 75%, 80%, etc.
[0120] For ease of understanding, Table 2 shows the first similarity corresponding to each similar POI address.
[0121] Table 2 Similarity values corresponding to each similar POI address
[0122] For example, if the first threshold is 60%, then the similar POI addresses with a similarity greater than or equal to the first threshold are: Point C, Point D, and Point E. The address types obtained by the data processing device are "shopping services" and "scenic spots." In this case, the first address type is "shopping services" or "scenic spots."
[0123] Mode 2: The first address type is the address type corresponding to the POI address with the highest similarity to the first POI address among the N similar POI addresses.
[0124] For example, the POI address with the highest similarity to the first POI address in Table 2 is point E, and the first address type is point E, and the address type corresponding to point E is "scenic spot".
[0125] Method 3: The first address type is the address type that appears most frequently among the N similar POI addresses. An example is given in conjunction with Table 1.
[0126] For example, as can be seen from Table 1, the address type "scenic spots" appears twice, the address type "shopping services" appears twice, and the address type "commercial residence" appears once. The address types "scenic spots" and "shopping services" appear the most frequently, so the first address type is "scenic spots" or "shopping services".
[0127] It should be understood that the above-mentioned methods are illustrative examples, and the determination of the first address type may be achieved through other methods, which are not limited in this application. In some embodiments, the above-mentioned methods may be combined.
[0128] S204: The data processing device determines a navigation address in the map according to the first POI address and the first address type.
[0129] A map is a graphic or image that selectively depicts certain phenomena on Earth (or other planets) in two-dimensional or multi-dimensional form and means on a plane or sphere according to certain rules. It has a strict mathematical foundation, a symbol system, and textual annotations. It can use cartographic generalization principles to scientifically reflect the distribution characteristics and interrelationships of natural and socioeconomic phenomena. For example, a map can be Amap.
[0130] Determining the navigation address in the map according to the first POI address and the first address type can be achieved in the following ways.
[0131] Method 1: The data processing device determines M map POI addresses in the map based on the first POI address and the first address type, where each of the M map POI addresses has a corresponding real address in the map. The data processing device determines a navigation address based on a second similarity between the M map POI addresses and the first POI address. M is an integer, and M ≥ 1.
[0132] For example, the value of M can be an integer such as 1, 2, 3, or 4.
[0133] Furthermore, determining the navigation address according to the second similarity between the M map POI addresses and the first POI address can be implemented in the following manner.
[0134] The data processing device determines a target address based on the second similarity between the M map POI addresses and the first POI address, where the target address is the map POI address with the highest second similarity to the first POI address among the M map POI addresses; if the second similarity between the target address and the first POI address is greater than a second threshold, the data processing device uses the target address as the navigation address.
[0135] Exemplarily, the data processing device records the maximum similarity as sim and determines the difference between sim and a second threshold. If sim is greater than the second threshold, the data processing device uses the target address as the navigation address. If sim is less than the second threshold, the data processing device performs a second round of address query. After obtaining the address, the address is executed and the address tag is stored.
[0136] Exemplarily, the data processing device inputs a first POI address and a first address type into a map interface. The data processing device then obtains M map POI addresses obtained by searching the map based on the first POI address and the first address type. These M map POI addresses have corresponding real addresses on the map. The data processing device calculates similarity between the M map POI addresses and the first POI address, and selects the map POI address with the highest similarity to the first POI address among the M map POI addresses as the target address. When the second similarity between the target address and the first POI address is greater than a second threshold, the data processing device selects the target address as the navigation address. In this way, the navigation address can be determined more accurately and quickly, improving navigation accuracy and thereby enhancing the user experience.
[0137] For example, the map is AutoNavi Maps. When the data processing device searches for M map POI addresses based on the first POI address and the first address type, the search method may include keyword search, nearby search, input prompt, etc. Specifically, the data processing device may input the first POI address and the first address type into the AutoNavi Maps interface, and receive the M map POI addresses returned by AutoNavi Maps through the AutoNavi Maps interface.
[0138] Specifically, the similarity calculation may reflect the textual and semantic similarities between the M map POI addresses and the first POI address.
[0139] For example, the second threshold value is used to indicate the condition that must be met for the similarity between the map POI address and the first POI address. The second threshold value is predefined or pre-specified. For example, if the second threshold value is 75%, the second similarity greater than the second threshold value may be 80%, 90%, etc.
[0140] For example, taking the first POI address as "Window of the World" and the first address type as "Scenic Spots", M map POI addresses are retrieved from the map, and the (second) similarity between each of the M map POI addresses and "Window of the World" is calculated. Table 3 shows the second similarity between each map POI address and each map POI address, where Dr1-Dr5 are exemplary 5 map POI addresses (i.e., M=5).
[0141] Table 3 The second similarity between each map POI address and each map POI address
[0142] In this example, it can be obtained from the table that the target address is Dr5; taking the second threshold as 75% as an example, the navigation address is Dr5.
[0143] Method 2: The navigation address may be the map POI address with the second highest similarity to the first POI address among the M map POI addresses.
[0144] For example, in Table 3, Dr5 is the map POI address with the second highest similarity to the first POI address, and the navigation address is Dr5.
[0145] Method 3: Search the map for an address that is the first POI address and the address type is the first address type, and use the first address returned by the map as the navigation address, or use the address closest to the user's current location among the addresses returned by the map as the navigation address.
[0146] It should be understood that the above implementations are illustrative examples, and there may be more implementations during the specific implementation process. In addition, multiple implementations may be combined.
[0147] In some embodiments, if the second similarity between the target address and the first POI address is less than a second threshold, the data processing device outputs a prompt message, which instructs the user to enter a POI address or indicates that the first POI address is invalid. Optionally, after receiving the POI address entered by the user, the data processing device may execute navigation and save the address tag. This effectively avoids invalid addresses and improves the user experience.
[0148] For example, if the second threshold is 85%, and according to Table 3, the similarities are all less than the second threshold, the data processing device outputs a prompt message (i.e., a second round of interaction), the prompt message being used to instruct the user to enter a POI address or to indicate that the first POI address is invalid. For example, the prompt message may be "Please re-enter the POI address" or "You have entered an invalid address."
[0149] In some embodiments, the data processing device can generate a navigation route to the navigation address based on the navigation address. This allows for more accurate and intelligent execution of navigation commands. For example, if the aforementioned navigation address is Dr5, the data processing device can generate a navigation route to Dr5.
[0150] In the embodiment shown in FIG2 , the data processing device can identify the first POI address in the navigation command and, based on the address dataset, determine a first address type that the first POI address may correspond to. When determining the navigation address corresponding to the POI address on the map, the data processing device can combine the first address type to obtain a more likely navigation address. This means filtering out addresses of less likely address types, making the determined navigation address more accurate, thereby improving navigation accuracy and enhancing the user experience.
[0151] The embodiment shown in FIG2 includes multiple possible solutions. For ease of understanding, one possible solution is described below. It should be understood that some of the terminology, logic, etc. in the solution shown in FIG3 can be explained with reference to the embodiment shown in FIG2.
[0152] Please refer to Figure 3, which is a flowchart of a method for determining a navigation address provided in another embodiment of the present application. The method shown in Figure 3 can be applied to the navigation system shown in Figure 1, or to the data processing device 102 in the navigation system shown in Figure 1. The method includes one or more operations in S301-S310, as follows.
[0153] S301: Obtain navigation address data.
[0154] Navigation address data can be accumulated online map data or publicly available navigation address data. For example, a data processing device or server can extract online data from an intelligent voice navigation system and filter address data containing POIs in navigation intent as navigation address data. For example, a data processing device can collect publicly available online address data to obtain common addresses nationwide.
[0155] S302: Training an entity recognition model based on the navigation address data.
[0156] Specifically, the navigation address data (e.g., the address data with POI address in 301) is added to the navigation entity recognition model to enhance the entity recognition model's ability to recognize addresses. In this way, when the user initiates a navigation command with a POI address, the corresponding POI address is accurately recognized.
[0157] S303: Constructing a faiss library according to the navigation address data.
[0158] This operation can construct a Faiss library (or Faiss search library) for navigation based on navigation address data. For example, a data entry in the Faiss library might be in the form of "window of the world, scenic spot, longitude and latitude," where "window of the world" is the POI address, "scenic spot" is the address type of the POI address, and longitude and latitude are the actual location corresponding to the POI address. Optionally, the Faiss library can also include the address label, alias, etc. of the POI address.
[0159] S304: Identify the first POI address in the navigation command through the entity recognition model.
[0160] For example, a navigation command might be something like "I want to go to Window of the World." The entity recognition model can identify the entities "I" and "Window of the World." The entity recognition model can also analyze whether the entity is an address. For example, if the entity "I" is not an address, but the entity "Window of the World" is an address, then "Window of the World" is the first POI address.
[0161] It should be noted that the first POI address determined by the entity recognition module may or may not actually be a valid address. For example, in the navigation command "Navigate to Aunt's House", "Aunt's House" is recognized by the entity recognition model as the first POI address, but it may or may not be a valid address.
[0162] S305: Recall similar POI addresses of the first POI address and the address types of the similar POI addresses from the faiss database.
[0163] Specifically, this operation is to search the POI address identified by the entity recognition model in the faiss database to obtain similar addresses. For example, 5 similar POI addresses can be recalled from the faiss database.
[0164] S306: Calculate the similarity between the similar POI address and the first POI address through a similarity module to determine the first address type.
[0165] Specifically, this operation is to calculate the similarity between the address list recalled by faiss and the first POI address identified by the navigation entity recognition model, and obtain the address types of the recalled addresses whose similarity values are higher than a threshold.
[0166] S307: Obtain an Amap address list based on the first POI address and the first address type.
[0167] Specifically, this operation involves inputting the POI address entity identified by the navigation entity recognition model and the address type retrieved through the Faiss search into the AutoNavi map interface to obtain an AutoNavi address list, or search POI list. Search methods include AutoNavi's keyword search, nearby search, and input prompts.
[0168] Of course, Amap can also be replaced by other maps, such as Baidu Maps.
[0169] S308: Calculate the similarity sim between the first POI address and the addresses in the Amap address list using a similarity model.
[0170] Specifically, this operation is to perform similarity calculation between the obtained Amap address list and the entity (ie, the first POI) identified by the entity model, calculate the corresponding similarity values, and take the maximum similarity value as sim.
[0171] When sim is greater than the threshold, execute S310 ; when sim is less than the threshold, execute S309 .
[0172] S309: Second round of interactive inquiry of address.
[0173] Specifically, when the maximum similarity sim calculated between the address list returned by Amap and the entities recognized by the entity model is less than a threshold, the intelligent voice interaction performs a second round of address inquiry to obtain the address.
[0174] Furthermore, after the address is obtained in the second round of interaction, the address label can be saved.
[0175] For example, if the maximum similarity value in the AutoNavi address list corresponding to the POI address "Auntie's House" is less than the threshold, "Auntie's House" is considered an invalid address. In this case, the address is retrieved through a second round of interactive inquiry, and "Auntie's House" is used as the address tag for the address obtained from the inquiry. This facilitates retrieval during the next navigation, improving navigation intelligence and enhancing the user experience.
[0176] S310: Execute navigation.
[0177] In the embodiment shown in FIG3 , the first POI address in the navigation command is identified, similar POI addresses and address types of similar POIs are recalled from the faiss library, the first address type is determined, and an AutoNavi address list is obtained in AutoNavi Maps based on the first POI address and the first address type. The address in the AutoNavi address list that has the highest similarity to the first POI address and is greater than a threshold is used as the navigation address. In this way, addresses of address types with low probability are filtered out, making the determined navigation address more accurate, thereby improving the accuracy of navigation and enhancing the user experience.
[0178] The above describes in detail the method of the embodiment of the present application. The following provides an apparatus of the embodiment of the present application.
[0179] Please refer to Figure 4, which is a schematic diagram of the structure of a data processing device provided in an embodiment of the present application. Referring to Figure 4, the data processing device 40 may include a processing unit 401 and an identification unit 402. The data processing device 40 is used to implement the aforementioned method for determining a navigation address, such as the method for determining a navigation address in the embodiments shown in Figures 2 and 3.
[0180] It should be noted that the division of the above-mentioned multiple units is merely a logical division based on function and does not limit the specific structure of the data processing device 40. In a specific implementation, some functional modules may be subdivided into more small functional modules, and some functional modules may be combined into a single functional module.
[0181] In some embodiments, the identification unit 402 is configured to identify a first interest tag POI address in the navigation command;
[0182] The processing unit 401 is configured to:
[0183] According to the first POI address, N similar POI addresses and address types of the N similar POI addresses are retrieved from an address dataset, wherein the N similar POI addresses are similar to the first POI address, where N is an integer and N≥1;
[0184] Determining a first address type according to the address types of the N similar POI addresses;
[0185] A navigation address is determined in a map according to the first POI address and the first address type.
[0186] In some implementations, the recognition unit 402 is further configured to recognize entities in the navigation command according to the navigation command input by the user using a navigation entity recognition model.
[0187] In some implementations, the processing unit 401 is further configured to determine the first POI address according to an entity in the navigation command.
[0188] In some implementations, the navigation entity recognition model is trained based on part or all of the POI addresses in the address dataset.
[0189] In some embodiments, the address dataset includes a plurality of POI addresses and address types corresponding to the plurality of POI addresses;
[0190] The POI addresses in the address data set are determined according to multiple navigation intentions and / or multiple real addresses.
[0191] In some implementations, the processing unit 401 is further configured to:
[0192] Calculating similarities between the N similar POI addresses and the first POI address respectively to obtain a first similarity for each of the N similar POI addresses;
[0193] The first address type is obtained according to the address types of similar POI addresses having a first similarity greater than a first threshold among the N similar POI addresses.
[0194] In some implementations, the processing unit 401 is further configured to:
[0195] Determine, in a map, M map POI addresses according to the first POI address and the first address type, wherein the M map POI addresses have corresponding real addresses in the map, where M is an integer and M≥1;
[0196] determining a target address according to the second similarity between the M map POI addresses and the first POI address, the target address being the map POI address with the highest second similarity to the first POI address among the M map POI addresses;
[0197] If the second similarity between the target address and the first POI address is greater than a second threshold, the target address is used as the navigation address.
[0198] In some implementations, the navigation address is used to generate a navigation route to the navigation address.
[0199] In some embodiments, the processing unit 401 is further configured to output a prompt message if the similarity between the target address and the first POI address is less than a second threshold, wherein the prompt message is configured to instruct the user to input a POI address or to indicate that the first POI address is an invalid address.
[0200] It should be noted that the above modules (processing unit 401 and identification unit 402) are used to perform the relevant operations of the above method. For example, the identification unit 402 is used to perform the relevant content at S201, and the processing unit 401 is used to perform the relevant content of S202-S204.
[0201] Figure 5 is a schematic diagram of the structure of a data processing device provided in another embodiment of the present application. A data processing device is a device with processing capabilities. The device here can be a physical device, such as a server (such as a rack server), a host, etc., or a virtual device, such as a virtual machine, a container, etc.
[0202] 5 , the data processing device 50 includes: a processor 501, a memory 502, and one or more program codes, and may include a communication interface 503. It should be understood that the present application does not limit the number of processors and memories in the data processing device 50.
[0203] The processor 501 is a module for performing calculations and may include a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), a digital signal processor (DSP), a microcontroller unit (MCU), or one or more integrated circuits for controlling the execution of the above program.
[0204] The memory 502 is used to provide storage space, which can optionally store application data, user data, an operating system, and computer programs. The memory 502 may include, but is not limited to, read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer.
[0205] The memory 502 may exist independently and be connected to the processor 501 via a bus. The memory 502 may also be integrated with the processor 501.
[0206] The communication interface 503 is used to provide information input or output for the at least one processor. And / or, the communication interface 503 can be used to receive data sent externally and / or send data to the outside. The communication interface 503 can be a wired link interface such as an Ethernet cable, or a wireless link interface (Wi-Fi, Bluetooth, general wireless transmission and other wireless communication technologies, etc.). Optionally, the communication interface 503 can also include a transmitter (such as a radio frequency transmitter, antenna, etc.) coupled to the interface, or a receiver, etc.
[0207] In an embodiment of the present application, the one or more programs are stored in the memory 502 in the form of program code and are configured to be executed by the processor 501. The program includes instructions for implementing the aforementioned method for determining the navigation address. For example, the method for determining the navigation address shown in Figures 2 and 3. That is, the memory 502 stores executable instructions, and the processor 501 executes the executable instructions to implement the aforementioned method for determining the navigation address, such as the method for determining the navigation address in the embodiments of Figures 2 and 3. That is, the memory 502 stores instructions for executing the method for determining the navigation address.
[0208] Alternatively, the memory 502 stores executable instructions, and the processor 501 executes the executable instructions to implement the functions of the aforementioned processing unit (or device), thereby realizing the method for determining the navigation address.
[0209] The present application also provides a computer program product including instructions. The computer program product may be software or a program product including instructions that can be run on a computing device or stored in any available medium. The computer program instructions are used to implement the aforementioned method for determining a navigation address, such as the method for determining a navigation address in the embodiment of FIG. 2 .
[0210] The embodiment of the present application further provides a computer-readable storage medium including instructions for implementing the aforementioned method for determining a navigation address, such as the method for determining a navigation address in the embodiment of FIG. 2 .
[0211] The computer-readable storage medium may be any available medium capable of storing data in a data processing device, or a data storage device such as a data center containing one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive).
[0212] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described in this application as "exemplary" or "for example" should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.
[0213] The “at least one” mentioned in the embodiments of this application refers to one or more, and “plurality” refers to two or more. “At least one of the following items” or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, (a and b), (a and c), (b and c), or (a and b and c), where a, b, c can be single or multiple. “And / or” describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character “ / ” generally indicates that the previous and next associated objects are in an “or” relationship.
[0214] Furthermore, unless otherwise specified, ordinal numbers such as "first" and "second" in the embodiments of the present application are used to distinguish multiple objects and are not used to limit the order, timing, priority, or importance of multiple objects. For example, the first content providing device and the second content providing device are only for ease of description and do not indicate differences in the device structure, deployment order, importance, etc. of the first content providing device and the second content providing device.
[0215] Those skilled in the art will understand that all or part of the operations of the above embodiments may be performed by hardware or by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.
[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the protection scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for determining a navigation address, characterized in that: The method comprises: Identifying (S201) a first interest tag POI address in a navigation command; According to the first POI address, retrieve (S202) N similar POI addresses and address types of the N similar POI addresses in an address data set, wherein each of the N similar POI addresses is similar to the first POI address, and N is an integer greater than or equal to 1; Determine (S203) a first address type according to the address types of the N similar POI addresses; and A navigation address is determined in a map ( S204 ) according to the first POI address and the first address type.
2. The method according to claim 1, characterized in that The identifying the first POI address in the navigation command includes: identifying the navigation command input by the user through a navigation entity recognition model to obtain an entity in the navigation command; and The first POI address is determined according to the entity in the navigation command.
3. The method according to claim 1 or 2, characterized in that: The address data set includes a plurality of POI addresses and address types corresponding to the plurality of POI addresses; The POI addresses in the address data set are determined according to at least one of the following: multiple navigation intentions or multiple real addresses.
4. The method according to any one of claims 1 to 3, characterized in that: The determining a first address type according to the address types of the N similar POI addresses includes: Calculating a first similarity between each of the N similar POI addresses and the first POI address; and The first address type is determined according to the first similarity and the address type of each similar POI address of the N similar POI addresses.
5. The method according to claim 4, characterized in that The determining the first address type according to the first similarity and the address type of each similar POI address of the N similar POI addresses includes: The address type of the similar POI addresses having a first similarity greater than or equal to a first threshold among the N similar POI addresses is determined as the first address type.
6. The method according to claim 4, characterized in that The determining the first address type according to the first similarity and the address type of each similar POI address of the N similar POI addresses includes: The address type of a similar POI address with the greatest first similarity among the N similar POI addresses is determined as the first address type.
7. The method according to any one of claims 1 to 3, characterized in that: The determining a first address type according to the address types of the N similar POI addresses includes: The address type of the similar POI address with the largest number of occurrences of the address type among the N similar POI addresses is determined as the first address type.
8. The method according to any one of claims 1 to 7, characterized in that: The determining a navigation address in a map according to the first POI address and the first address type includes: Determine M map POI addresses in a map according to the first POI address and the first address type, wherein the M map POI addresses have corresponding real addresses in the map, and M is an integer greater than or equal to 1; determining a target address according to a second similarity between each of the M map POI addresses and the first POI address; and If the second similarity between the target address and the first POI address is greater than a second threshold, the target address is used as the navigation address.
9. The method according to claim 8, characterized in that The determining the target address according to the second similarity between each of the M map POI addresses and the first POI address includes: The map POI address having the greatest second similarity among the M map POI addresses is determined as the target address.
10. The method according to claim 8, characterized in that The navigation address is used to generate a navigation route to the navigation address.
11. The method according to claim 8, characterized in that The method further comprises: If the second similarity between the target address and the first POI address is less than the second threshold, a prompt message is output, wherein the prompt message is used to instruct the user to input a POI address or to indicate that the first POI address is an invalid address.
12. A data processing device (40), characterized in that: The data processing device comprises an identification unit (402) and a processing unit (401); The identification unit is used to identify the first interest tag POI address in the navigation command; and The processing unit is used for: According to the first POI address, retrieve N similar POI addresses and address types of the N similar POI addresses in the address data set, wherein each of the N similar POI addresses is similar to the first POI address, and N is an integer greater than or equal to 1; Determining a first address type according to the address types of the N similar POI addresses; and A navigation address is determined in a map according to the first POI address and the first address type.
13. A data processing device (50), characterized in that: The data processing device comprises a processor (501) and a memory (502), wherein a program is stored in the memory, and the program comprises instructions for executing the method according to any one of claims 1 to 11.
14. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, wherein the computer program includes instructions for executing the method according to any one of claims 1 to 11.
Citation Information
Patent Citations
Method for determining navigation address and related device
CN118394990A
Address positioning method and device, readable storage medium and electronic equipment
CN110674423A
Address data processing method, device, electronic equipment and computer readable medium
CN110990651A
Address type identification method and device and electronic equipment
CN111435360A
Address resolution method and device, computer equipment and storage medium
CN114169331A
