Place name and address translation tool and method integrating artificial intelligence proper noun and common noun splitting

Abstract

The present application relates to the field of place name and address translation technology, and specifically discloses a place name and address translation tool and method that integrates artificial intelligence proper noun and common noun splitting. The tool adopts deep learning-based artificial intelligence technology to perform semantic analysis and contextual semantic association enhancement on the place name and address to be translated, and facilitates the proper noun and common noun splitting and type annotation of the place name and address to be translated based on the Prompt-driven LLM model. Based on a preset geographic terminology knowledge base, an iterative minimum translation unit determination is performed on each place name and address segment, and a recursive decomposition is performed on non-minimum unit segments until all segments meet the minimum translation granularity requirement. Each translation unit is subjected to transliteration or standard translation conversion, and a standard place name and address translation is output through normalization processing. This can effectively solve the problems of inaccurate semantic splitting and insufficient standardization of translated names in traditional translation, and improve the quality of place name and address translation in scenarios such as cross-border map services and international logistics.

Description

Technical Field

[0001] This application relates to the technical field of place name and address translation, and more specifically, to a place name and address translation tool and method integrating artificial intelligence for splitting proper names and common names. Background Art

[0002] With the in-depth development of globalization and the increasing frequency of international exchanges, the cross-language transmission and understanding of place name and address information have become crucial. Whether it is international trade, logistics transportation, cross-border tourism, emergency response, or cultural exchanges, accurate and efficient place name and address translation is the basis for ensuring smooth communication and effective cooperation. However, as a special language phenomenon, place name and address translation not only involves language conversion, but is also closely related to various factors such as geography, culture, and administrative management, with a high degree of complexity and particularity. When traditional machine translation systems or manual translation methods handle place names and addresses, they often face many challenges. For example, they cannot clearly identify proper names (such as "Chaoyang") and common names (such as "District", "Street") in place names, resulting in translation errors or non-compliance with the expression habits of the target language; they handle nested multi-level administrative divisions, aliases, common names, etc. improperly, causing information loss or ambiguity; and they cannot ensure the unity and authority of standard translations of specific geographical entities, seriously affecting the quality and practicality of place name and address translation.

[0003] Existing place name and address translation technologies mostly adopt translation methods based on rules or statistical models. Although they can handle some simple and standardized place names and addresses, their processing capabilities for complex structures, colloquial expressions, or emerging place names are limited. The maintenance cost of the rule base is high, and it is difficult to adapt to the dynamic changes of place name information. In recent years, although neural machine translation (NMT) has made remarkable progress in the field of general text translation, when directly applied to place name and address translation, due to the lack of in-depth understanding of the unique structure and semantics of place names and addresses, problems such as incorrect translation of proper names, improper handling of common names, or inability to recognize the hierarchical relationship of place names often occur, resulting in rigid, unnatural, or even completely incorrect translation results. For example, the model may not be able to accurately distinguish whether "Beijing Road" refers to the proper name "Beijing" or the "Beijing Municipality", and it is also difficult to properly handle nested place names such as "Middle Road of Taiyanggong", thus affecting the accuracy and usability of the translation.

[0004] Therefore, an optimized place name and address translation tool and method integrating artificial intelligence for splitting proper names and common names are needed to solve the above technical problems. Summary of the Invention

[0005] To solve the above technical problems, this application is proposed.

[0006] According to one aspect of this application, a place name and address translation method integrating artificial intelligence for splitting proper names and common names is provided, which includes:

[0007] S1, obtaining the place name and address to be translated and the specified target translation language input by the user;

[0008] S2, using the LLM model to perform proper noun-common noun splitting and type annotation on the place name and address to be translated to obtain a sequence of annotated place name and address segments to be translated;

[0009] S3, based on a preset geographic terminology knowledge base, querying whether each of the to-be-translated place-name and address segments in the sequence of annotated place-name and address segments is a minimum translation unit; if not, iteratively performing steps S2 and S3 on the to-be-translated place-name and address segments to obtain a sequence of minimum translation units of the annotated place-name and address segments;

[0010] S4, translating each of the minimum translation units of the place name and address to be translated in the sequence of the annotated minimum translation units of the place name and address to be translated to obtain a translation result of the place name and address in the target language;

[0011] S5, normalizing the target language place name and address translation result to obtain a place name and address translation output.

[0012] According to another aspect of the present application, a place name and address translation tool integrating artificial intelligence proper noun and common noun splitting is provided, which includes:

[0013] The user input module is used to obtain the place name and address to be translated and the specified target translation language input by the user;

[0014] A place name preprocessing module is used to use the LLM model to split the place name address to be translated into a specific name and a common name and to mark the type to obtain a sequence of marked place name address segments to be translated;

[0015] A minimum translation unit identification module is used to query, based on a preset geographic terminology knowledge base, whether each place name and address segment to be translated in the sequence of annotated place name and address segments to be translated is a minimum translation unit; if not, cyclically calling the place name preprocessing module and the minimum translation unit identification module to process the place name and address segments to be translated, so as to obtain a sequence of minimum translation units of the annotated place name and address to be translated;

[0016] A translation processing module, configured to perform translation processing on each of the minimum translation units of the place name and address to be translated in the sequence of the annotated minimum translation units of the place name and address to be translated to obtain a translation result of the place name and address in the target language;

[0017] The translation result normalization module is used to perform normalization processing on the target language place name and address translation result to obtain the place name and address translation output.

[0018] Beneficial effect: Compared with the existing technology, the place name and address translation tool and method provided by the present application that integrates artificial intelligence proper noun and common noun splitting uses deep learning-based artificial intelligence technology to perform semantic analysis and contextual semantic association enhancement on the place names and addresses to be translated, and is conducive to the realization of proper noun and common noun splitting and type annotation of the place names and addresses to be translated based on the Prompt-driven LLM model, generating a sequence of annotated place name and address fragments. Subsequently, based on the preset geographic terminology knowledge base, an iterative minimum translation unit determination is performed on each place name and address fragment, and through layer-by-layer query and comparison, recursive decomposition is performed on non-minimum unit fragments until all fragments meet the minimum translation granularity requirements, and then each translation unit is subjected to transliteration or standard translation conversion based on the type annotation results, and the place name and address translation that meets the standard is output through normalization processing. This method can effectively solve the problems of inaccurate semantic splitting and insufficient standardization of translated names in traditional translation, thereby improving the quality of place name and address translation in scenarios such as cross-border map services and international logistics. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0020] Figure 1 The present invention is a flowchart of a place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0021] Figure 2 This is a flowchart of sub-step S2 of the place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0022] Figure 3 This is a data flow diagram of sub-step S2 of the place name and address translation method for integrating artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0023] Figure 4 This is a flowchart of sub-step S23 of the place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0024] Figure 5 This is a flowchart of sub-step S231 of the place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0025] Figure 6 This is a flowchart of sub-step S2312 of the place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application.

[0026] Figure 7 This is a block diagram of a place name and address translation tool that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0028] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0029] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0030] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0031] It should be noted in advance that the acquisition and processing of all information or data in this application are carried out in compliance with relevant data protection laws and policies and with the authorization of the relevant authority managers.

[0032] Figure 1 Flowchart of the place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to the embodiment of the present application. Figure 1As shown, the place name and address translation method integrating artificial intelligence proper noun and general noun splitting includes the following steps: S1, obtaining the place name and address to be translated and the specified target translation language input by the user; S2, using the LLM model to split the place name and address to be translated into proper noun and general noun and type mark the place name and address to be translated to obtain a sequence of marked place name and address segments to be translated; S3, based on a preset geographic terminology knowledge base, querying whether each place name and address segment to be translated in the sequence of marked place name and address segments to be translated is a minimum translation unit; if not, iteratively executing steps S2 and S3 on the place name and address segments to be translated to obtain a sequence of marked place name and address minimum translation units to be translated; S4, translating each place name and address minimum translation unit to be translated in the sequence of marked place name and address minimum translation units to obtain a target language place name and address translation result; S5, normalizing the target language place name and address translation result to obtain a place name and address translation output.

[0033] In the above-mentioned method for translating place names and addresses that integrates artificial intelligence to split proper nouns and common nouns, step S1 obtains the place name and address to be translated and the specified target translation language input by the user. Specifically, in order to ensure the input integrity and adaptability of the target language during the translation process, this application constructs a user interaction interface to receive the place name and address text to be translated and the target language parameters, and performs preliminary verification of the input content (such as format legitimacy and language encoding identification) to ensure the validity and standardization of the input content. In the specific implementation process, by providing a user interaction interface, such as a web form or an application programming interface (API), the user can enter the Chinese place name and address to be translated (such as "No. 1, Zhongguancun East Road, Haidian District, Beijing") through a text box and select the target language (such as "English") from a drop-down list or parameter.

[0034] Specifically, during implementation, users can submit place-name and address information requiring translation to the system through a variety of channels. For example, on a website, users might directly type a Chinese address string into a text input box, such as "No. 1, Zhongguancun East Road, Haidian District, Beijing." On a mobile device, users might enter the address through voice recognition or handwriting input. In enterprise applications or automated system integration scenarios, users are more likely to use programmatic interfaces (such as RESTful APIs) to submit structured address data in batches. Regardless of the method used, the system must have a unified data reception mechanism and be able to dynamically adjust parsing strategies based on different input channels to ensure that data from various sources is correctly read and entered into subsequent processing.

[0035] At the same time, users also need to clearly specify the target translation language, that is, the language version into which they want the original place name address to be translated. This parameter is usually selected by the user through a drop-down menu, radio button, or multilingual identifier field. The system must maintain a complete list of target languages and support ISO standard language codes (such as en for English, es for Spanish, ja for Japanese, etc.) so that subsequent modules can quickly match the corresponding language rule base and terminology resources when performing translation tasks. In addition, for certain special needs, such as bilingual output or regional variant selection (such as American English, British English), the system can also provide additional options for users to configure, thereby improving the adaptability and accuracy of the translation results.

[0036] After receiving the user's input, the system will immediately start the preliminary verification process of the input content, which is an important prerequisite for ensuring the quality of translation. First, the system will check the format of the input address text to determine whether it meets the expected character set specifications (such as only containing Chinese characters, numbers, common punctuation marks, etc.), and exclude interference from illegal characters or maliciously injected content. Secondly, the system will also perform language encoding recognition operations, using natural language processing technology to automatically detect the language type of the input text to ensure that it is a valid Chinese address expression, rather than other unrelated languages or garbled content. If an abnormality is found in the input content, the system should return a corresponding error prompt and guide the user to resubmit the correct data to avoid invalid input interfering with the overall process.

[0037] To enhance the robustness of the system and enhance user experience, this step should also have a certain degree of fault tolerance. For example, if the address format entered by the user is not standardized, the system can attempt to perform a certain degree of automatic repair, such as removing extra spaces, correcting common spelling errors, or completing missing administrative division names. Although this intelligent preprocessing mechanism is not a core function of this step, it can significantly improve the quality of input data without affecting the main process, laying a good foundation for the subsequent separation of proper nouns and common nouns and standardized translation.

[0038] Furthermore, given that place name and address translation tools may be targeted at a global user base, the cultural differences and technical environments of different regions must be fully considered during the input processing phase. For example, in some non-Unicode environments, the system may need to support conversions between multiple character encoding formats, such as UTF-8, GBK, and Big5, to ensure that users from different operating systems or browsers can successfully submit data. At the same time, the system should also have a corresponding mapping mechanism for some special characters or dialect words to avoid translation failures due to invisible or unrecognizable characters.

[0039] In the above-mentioned place name and address translation method integrating artificial intelligence proper noun and common noun splitting, the step S2 uses the LLM model to perform proper noun and common noun splitting and type annotation on the place name and address to be translated to obtain a sequence of annotated place name and address fragments to be translated. It should be understood that since traditional translation methods are difficult to accurately parse the complex combination of proper nouns (such as "Zhongguancun") and common nouns (such as "East Road" and "No. 1 Courtyard") in place name addresses, as well as to identify the geographical semantic types of each fragment (such as roads, administrative districts, points of interest), this directly affects the accuracy and naturalness of the translation. For example, if it is impossible to distinguish between "Nanjing" (the proper noun part of the city name as a road name) in "Nanjing Road" and "Nanjing" (the city name itself) in "Nanjing City", it may cause the road name to be mistranslated as the city name, or vice versa, causing serious translation confusion. Therefore, in order to achieve deep semantic understanding and structured analysis of place name addresses, this application further utilizes the powerful natural language processing capabilities of the large language model (LLM) to achieve intelligent splitting and precise annotation of place name addresses to be translated. Among them, Figure 2 This is a flowchart of sub-step S2 of the place name and address translation method that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application. Figure 3 The data flow diagram of sub-step S2 of the place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to the embodiment of the present application is shown in FIG. Figure 2 and Figure 3 As shown, the step S2 includes the steps of: S21, performing data preprocessing on the place name and address to be translated to obtain a standardized place name and address to be translated; S22, extracting the contextual semantic features of the standardized place name and address to be translated to obtain a contextual semantic coding vector of the place name and address to be translated; S23, performing contextual semantic association enhancement on the contextual semantic coding vector of the place name and address to be translated to obtain a contextual semantic association enhanced coding vector of the place name and address to be translated; S24, inputting the contextual semantic association enhanced coding vector of the place name and address to be translated into a proper noun and common noun splitting and annotation module based on the Prompt-based LLM model to obtain a sequence of the annotated place name and address fragments to be translated.

[0040] Specifically, in step S21, data preprocessing is performed on the to-be-translated place name address to obtain a standardized to-be-translated place name address. Specifically, this application takes into account that the original input of place name addresses often contains non-standard symbols (such as "#", " / "), chaotic formats (such as mixed Chinese and English, missing spaces), or dialect variants (such as "Pudong" written as "Pudong South Road" and "Pudong · South Road" coexisting). Direct input into the model is likely to cause word segmentation errors or semantic ambiguities. Therefore, in order to eliminate input noise and unify the expression form, this application further performs data cleaning and standardization processing on the to-be-translated place name address to obtain a standardized to-be-translated place name address. In the specific implementation process, first, irrelevant symbols are removed (such as retaining core descriptors like "Road", "Number", and filtering advertising text), then full-width / half-width character conversion is performed through Unicode normalization, and the omitted levels are supplemented based on the authoritative place name database (such as the administrative division code of the Ministry of Civil Affairs) (such as "Haidian District, Beijing" is supplemented to "Haidian District, Beijing Municipality"), and finally, a standardized place name address with a unified structure and conforming to geographical naming norms is output.

[0041] Specifically, in a specific example of this application, step S22 includes: performing context semantic encoding on the standardized to-be-translated place name address based on the mBERT model to obtain the context semantic encoding vector of the to-be-translated place name address. It should be understood that considering that the meanings of each lexical unit in a place name address often depend on its context, for example, "East Road" represents a part of the road name in "East Nanjing Road", while in "East Road Primary School" it may represent a part of the school name. Therefore, in order to capture the deep semantics of each component in the standardized to-be-translated place name address and the contextual associations between them, this application uses the transfer learning ability of the pre-trained language model and adopts the mBERT (Multilingual Bidirectional Encoder Representations from Transformers) model to perform deep semantic encoding on the standardized to-be-translated place name address. It should be understood that the mBERT model is based on the bidirectional Transformer architecture and can capture the dependencies between lexical units bidirectionally, thus more accurately understanding the meaning of each word in a specific context. In this application, by inputting the standardized to-be-translated place name address into the mBERT model and using its context representation ability pre-trained on multiple language corpora, word segmentation, word embedding, and context semantic encoding are performed on the standardized to-be-translated place name address, enabling deep semantic understanding of the standardized to-be-translated place name address and generating its context semantic representation, obtaining the context semantic encoding vector of the to-be-translated place name address, providing a solid context semantic basis for subsequent place name address splitting and annotation.

[0042] Specifically, the step S23 performs context semantic association enhancement on the context semantic encoding vector of the place name and address to be translated to obtain the context semantic association enhanced encoding vector of the place name and address to be translated. It should be understood that although the context semantic encoding vector of the place name and address to be translated contains the context semantic information of the place name and address, when processing texts such as place names and addresses with specific structures and high local dependencies, the mBERT model may not be able to capture the fine semantic associations between the internal components of the place name and address, such as hierarchical relationships, adjacent relationships, and the relationship between the part and the whole. Therefore, in order to further strengthen the important associations between the internal components of the place name and address to be translated, the present application proposes a feature enhancement and reconstruction method, which distills and reconstructs the context semantic encoding vector of the place name and address to be translated by utilizing the semantic association topological structure between the local components of each place name and address, so that it can explicitly model the combination rules and boundary constraints between the internal components of the place name while retaining the global context information of the place name and address to be translated, so that the subsequent place name and address splitting and annotation tasks are more focused on the overall semantic structure rather than the surface vocabulary. Among them, Figure 4 Flowchart of sub-step S23 of the place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to the embodiment of the present application. Figure 4 As shown, the step S23 includes the steps of: S231, performing local scale-based semantic feature distillation and refinement on the context semantic coding vector of the place name and address to be translated to obtain a sequence of local semantic feature distillation coding vectors of the place name and address to be translated; S232, performing context association reconstruction on the sequence of local semantic feature distillation coding vectors of the place name and address to be translated to obtain the context semantic association enhanced coding vector of the place name and address to be translated.

[0043] Figure 5 Flowchart of sub-step S231 of the place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to the embodiment of the present application. Figure 5 As shown, the step S231 includes the steps of: S2311, performing semantic decomposition based on one-dimensional convolutional coding on the context semantic coding vector of the place name and address to be translated to obtain a sequence of local semantic feature coding vectors of the place name and address to be translated; S2312, performing feature distillation coding on each local semantic feature coding vector of the place name and address to be translated in the sequence of local semantic feature coding vectors of the place name and address to be translated based on the context semantic association topological structure of the sequence of local semantic feature coding vectors of the place name and address to be translated to obtain a sequence of local semantic feature distillation coding vectors of the place name and address to be translated.

[0044] In a specific example of the present application, step S2311 is expressed as follows:

[0045] ;

[0046] ;

[0047] in, represents the context semantic encoding vector of the place name address to be translated, Indicates based on One-dimensional convolution operation of the convolution kernel, is the scale of the one-dimensional convolution kernel, represents a set of local semantic feature encoding vectors of the place name address to be translated, 、 、 and Respectively represent the first, second, and third in the set of local semantic feature encoding vectors of the place name address to be translated and The local semantic feature encoding vector of the place name to be translated, The number of local semantic feature encoding vectors for the place name address to be translated.

[0048] That is, the local receptive field characteristics of one-dimensional convolutional coding are utilized to scan the contextual semantic encoding vectors of the place-name address to be translated. By applying a one-dimensional convolutional kernel to capture different types of local semantic patterns, the contextual semantic encoding vectors of the place-name address to be translated are projected onto a structured space composed of local feature bases, thereby achieving a transformation from holistic representation to distributed local representation. Based on this, the resulting sequence of local semantic feature encoding vectors of the place-name address to be translated contains a variety of local semantic structure information extracted from the contextual semantic encoding vectors of the place-name address to be translated, which can effectively represent the local semantic features of each component of the place-name address, providing a structured local semantic feature foundation for subsequent processing.

[0049] Figure 6 Flowchart of sub-step S2312 of the place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to the embodiment of the present application. Figure 6As shown, the step S2312 includes the steps of: S23121, calculating the semantic correlation factor between any two local semantic feature coding vectors of the place name and address to be translated in the sequence of local semantic feature coding vectors of the place name and address to be translated to obtain a local semantic feature correlation topology matrix of the place name and address to be translated composed of multiple semantic correlation factors; S23122, inputting the local semantic feature correlation topology matrix of the place name and address to be translated into a gated mask function to obtain a fine-grained correlation mask topology matrix of the local semantic features of the place name and address to be translated; S23123, based on the fine-grained correlation mask topology matrix of the local semantic feature of the place name and address to be translated, performing feature structure feedback modulation on each local semantic feature coding vector of the place name and address to be translated in the sequence of local semantic feature coding vectors of the place name and address to be translated to obtain a sequence of local semantic feature distillation coding vectors of the place name and address to be translated.

[0050] In a specific example of the present application, step S23121 is expressed as follows:

[0051] ;

[0052] in, The first one in the set of local semantic feature encoding vectors of the place name address to be translated The local semantic feature encoding vector of the place name to be translated, represents transpose, Indicates the calculation of the 2-norm of the vector, represents the bandwidth parameter, represents the exponential function with base e, express and The semantic correlation factor between the local semantic features of the place name and address to be translated is the first The element value at position.

[0053] That is, by calculating the semantic correlation factor between any two local semantic feature encoding vectors of the place names and addresses to be translated, quantifying the mutual relationship and proximity between the two in the semantic space, and then arranging multiple semantic correlation factors into a topological matrix, the local semantic feature correlation topological matrix of the place names and addresses to be translated is obtained, which can effectively represent the semantic dependency and geometric distribution relationship between the local semantic features, reveal implicit semantic relationships such as administrative hierarchy, geographical location, and semantic subordination, and provide structured semantic prior knowledge for subsequent operations.

[0054] In a specific example of the present application, step S23122 is expressed as follows:

[0055] ;

[0056] in, represents the gated mask weight matrix, Represents the topological matrix of the local semantic features of the place name to be translated, represents the gated mask bias matrix, represents the sigmoid activation function, Represents the fine-grained association mask topology matrix of the local semantic features of the place name address to be translated.

[0057] That is, for the possible noise associations or non-critical connections in the local semantic feature association topology matrix of the place name and address to be translated, the association strength of the local semantic feature association topology matrix of the place name and address to be translated is modulated by a gated mask function to achieve focusing on important semantic associations and suppressing noise associations, thereby improving the expression purity of semantic relationships and the targetedness of subsequent processing, so that the obtained fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated can more accurately represent the key associations between the local semantics of the place name and address, and provide a more discriminative semantic structure prior for subsequent feature distillation and reconstruction.

[0058] In particular, considering that the geometric coupling configuration distribution of the local semantic feature association topology matrix of the place name and address to be translated on the potential low-dimensional manifold structure will be nonlinear and unsaturated, the global association topology configuration of the local semantic feature association topology matrix of the place name and address to be translated will be compressed due to the nonlinear coupling association, and the association polarization gain effect of the gated mask will become more significant, affecting the essential geometric microstructure association expression effect of the fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated. Therefore, in a preferred example of the present application, the step S23123 includes: performing local structural balanced optimization on the fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated to obtain an optimized fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated.

[0059] Specifically, first, for the local semantic feature fine-grained association mask topology matrix of the place name address to be translated Each eigenvalue of , a vector gradient field is introduced to correct the local non-uniform geometric coupling, thereby realizing the microstructure regularization of the geometric coupling field:

[0060] ;

[0061] in ， express Middle Rank Elements of the column, Represents the fine-grained association mask topology matrix of the local semantic features of the place name to be translated middle The corresponding vector gradient field quantity, express Middle Rank Elements of the column, It means partial derivative.

[0062] Then, the vector gradient field quantity As an external field excitation item, the fine-grained association mask topology matrix of the local semantic features of the place name address to be translated is used. Dynamic adjustment of the mean field of each eigenvalue in :

[0063] ;

[0064] in, is the fine-grained association mask topology matrix of the local semantic features of the place name address to be translated The eigenmean of all eigenvalues of , represents the gain modulation coefficient, Represents the first in the fine-grained correlation mask topology matrix of the optimized received signal strength local temporal feature structure Rank Elements of a column.

[0065] In this way, under the action of the external field excitation term as a multi-order differential gradient, the nonlinear response saturation of the geometric coupling configuration distribution under the macroscopic mean field is reversely promoted, thereby compensating for the sub-configuration correlation decoupling caused by the correlation polarization gain effect through the mean field harmonic response under the mean field, thereby improving the essential geometric micro-correlation topological expression effect of the fine-grained correlation mask topological matrix of the local semantic features of the place name address to be translated.

[0066] Then, the local semantic feature encoding vector of the place name and address to be translated and the optimized fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated are input into the feature dense feedback distillation unit to obtain the distilled encoding vector of the local semantic features of the place name and address to be translated, which is expressed as follows:

[0067] ;

[0068] in, Represents the optimization of the fine-grained association mask topology matrix of the local semantic features of the place name address to be translated, represents the distillation weight matrix, represents dot product, represents the matrix multiplication operation, is a nonlinear activation function, express The characteristic scale value of The first one in the set of local semantic feature distillation encoding vectors of the place name to be translated The local semantic feature distillation encoding vector of the place name to be translated.

[0069] That is, through the feature-intensive feedback distillation mechanism, the local semantic feature encoding vector of each place-name address to be translated is structurally fused with the optimized fine-grained association mask topology matrix of the local semantic features of the place-name address to be translated, achieving information extraction and ambiguity elimination of the local semantic features of the place-name address to be translated, promoting the expression consistency and coordination between the local semantic features of each place-name address to be translated, and thus obtaining a set of distilled encoding vectors of the local semantic features of the place-name address to be translated that is richer in information and more semantically relevant. In this way, based on the association topology between the various components of the place-name address to be translated and guided by global context information, a refined understanding and expression of the local semantic features of each place-name address to be translated can be achieved.

[0070] More specifically, in a specific example of the present application, step S232 includes: inputting the sequence of the local semantic feature distillation encoding vectors of the place name and address to be translated into a feature reconstruction module based on the self-attention mechanism to obtain the contextual semantic association enhanced encoding vector of the place name and address to be translated, which is expressed as follows:

[0071] ;

[0072] ;

[0073] in, Represents the set of distilled encoding vectors of the local semantic features of the place name to be translated. 、 and The first, second and third vectors in the set of local semantic feature distillation encoding vectors of the place name to be translated are represented by The local semantic feature distillation encoding vector of the place name to be translated, represents the feature reconstruction operation, 、 and represent the query matrix, key matrix and value matrix respectively, 、 and denote the query embedding matrix, key embedding matrix and value embedding matrix respectively, represents the softmax activation function, The context semantic association enhanced encoding vector represents the place name and address to be translated.

[0074] That is, by leveraging the global information integration ability of the self-attention mechanism, long-distance semantic dependencies in the sequence of the distilled encoding vectors of the local semantic features of the to-be-translated place names and addresses are dynamically captured. Based on the context weights, the importance of each local semantic feature to the overall representation is automatically judged, and they are dynamically reconstructed into a unified and high-order enhanced encoding vector of the context semantic associations of the to-be-translated place names and addresses, so as to comprehensively represent complex semantic associations such as hierarchical structures and geographical relationships in the place names and addresses, realize the deep abstraction and high-order representation of the context semantics of the place names and addresses, provide a more globally consistent and semantically discriminative feature representation for the subsequent translation process, and improve the translation accuracy and integrity of complex place names and addresses.

[0075] Specifically, in step S24, the enhanced encoding vector of the context semantic associations of the to-be-translated place names and addresses is input into the proper name and common name splitting and annotation module based on the Prompt-based LLM model to obtain the sequence of the to-be-translated place name and address segments with annotations. That is, in order to convert the enhanced context semantic information of the to-be-translated place names and addresses into a clear sequence of proper name and common name segments and their corresponding type tags, this application is based on the Prompt-based Learning paradigm of the large language model. By designing specific prompts, the LLM model after targeted fine-tuning is guided to perform the tasks of proper name and common name splitting and geographical type annotation. In the specific implementation process, first, a structured prompt template is designed, and the designed prompt is embedded into the LLM model so that it can understand and execute specific splitting and annotation tasks. Then, the enhanced encoding vector of the context semantic associations of the to-be-translated place names and addresses is used as the input. After receiving the input, the LLM model will automatically perform in-depth semantic parsing on the contained place name and address information according to the guidance of the prompt to identify the proper names (such as city names, road names, etc.) and common names (such as "road", "street", "district", etc.) therein, and generate the corresponding annotation output (such as [{"text": "Beijing", "PN / GN": "proper name", "type": "city name"}, {"text": "City", "PN / GN": "common name", "type": "suffix of municipal administrative division"}, {"text": "Haidian", "PN / GN": "proper name", "type": "district name"}, {"text": "District", "PN / GN": "common name", "type": "suffix of district-level administrative division"},...]), thus providing accurate structured information for the subsequent place name and address translation. This splitting and annotation method based on the Prompt-based LLM model not only improves the accuracy and naturalness of place name and address translation, but also simplifies the translation process and improves the translation efficiency.

[0076] In the above method for translating place names and addresses by integrating the splitting of proper names and common names of artificial intelligence, in step S3, based on a preset geographical term knowledge base, query whether each place name and address segment to be translated in the sequence of the annotated place names and addresses to be translated is a minimum translation unit. If not, iteratively execute steps S2 and S3 on the place name and address segment to be translated to obtain a sequence of minimum translation units of the annotated place names and addresses to be translated. It should be understood that since there may be multiple levels of nesting in place name and address segments, directly translating un-decomposed composite segments is likely to lead to redundant translations or semantic loss. Therefore, to ensure the optimization of the translation granularity, this application is based on a knowledge base-driven recursive decomposition algorithm. Through a preset geographical term knowledge base (including a standard common name mapping table, an administrative division hierarchy table, etc.), iterative querying and decomposition are performed on each place name and address segment output by the LLM model to ensure that each segment is a non-decomposable minimum translation unit. Specifically, first, each place name and address segment to be translated is compared with the minimum translation units in the preset geographical term knowledge base (such as "city" and "district" are independent common names, and "Taiyanggong" is an indivisible proper name). If the segment can be further decomposed (for example, "Chaoyang District" can be decomposed into the proper name "Chaoyang" and the common name "district"), a recursive call is triggered until all segments meet the minimum granularity requirements. In this way, it is possible to dynamically adapt to the differences in place name structures of different languages (such as the reverse expression of English addresses), ensuring the independence and accuracy of each translation unit in the target language.

[0077] In the above-mentioned place name and address translation method that integrates artificial intelligence proper noun and common noun splitting, the step S4 is to translate each of the minimum translation units of the place name and address to be translated in the sequence of the annotated minimum translation units of the place name and address to be translated to obtain the place name and address translation result in the target language. It should be understood that the proper nouns in the place name and address usually need to be transliterated (such as "Beijing"), while the common nouns need to adopt the standard translation name (such as "Road") according to the target language specification. Therefore, in order to achieve accurate adaptation of the translation strategy, this application is based on the type-driven translation strategy selection principle. According to the type annotation of each minimum translation unit of the place name and address to be translated (such as proper noun, common noun) and the target translation language specified by the user, the method of transliteration or table lookup to obtain the standard translation name for each unit is selected for translation. In the specific implementation process, the system traverses the sequence of the annotated minimum translation units of the place name and address to be translated. For each minimal translation unit of a place name or address to be translated, if its type is marked as "proper noun" and there is no corresponding standard translation name in the knowledge base (for example, the name of a new residential area or shop), the transliteration module (such as converting it into English syllables based on Pinyin transliteration rules) is called for processing; if its type is "common noun" (such as "road," "street," "district," or "province") or it is a "proper noun" with an official standard translation name (such as "China" or "Yellow River"), its standard translation in the target language is searched and extracted from the preset geographic terminology knowledge base. In this way, each place name or address component is ensured to receive the most appropriate translation processing, thereby improving the accuracy and authenticity of the translation and generating a sequence of target language place name and address translation fragments as the preliminary translation result.

[0078] In the place name and address translation method of the above-mentioned integrated artificial intelligence proper noun and common noun splitting, the step S5 is to normalize the target language place name and address translation result to obtain the place name and address translation output. It should be understood that the present application takes into account that directly splicing the various translation segments in the target language place name and address translation result may not conform to the writing habits, word order or format requirements of the target language. For example, the word order of Chinese and English addresses is opposite, and there are often commas between the components in the English address, and there are specific rules for capitalization. Therefore, in order to combine scattered translation segments into a complete, standardized, and easy-to-read target language place name and address, the present application further performs post-processing based on the target language address expression specification on the target language place name and address translation result, and obtains the final place name and address translation output by performing normalization operations such as word order adjustment, punctuation addition, and case conversion on it. In the specific implementation process, first, a structured place name and address standard expression template is designed for different language types, and the components, word order, punctuation, and capitalization rules of the target language place name and address are defined. For example, the standard representation template for an English place name address is "[address number, house number, street name, common name (street / road), district / county name, city / state name, postal code, country name]," where each component is separated by a comma, the first letter of the street name and common name is capitalized, and the rest is lowercase. Next, the corresponding standard representation template for place names and addresses is invoked based on the target language type. Each translated segment in the target language place name address translation result is mapped to the standard representation template, and each segment is correctly cased (e.g., the first letter of the street name and city name is capitalized). For example, if the source language address is Chinese, the original segment order is "[country name] [province / city] [district / county] [street] [house / building name]." When mapping this to the standard representation template for English place names and addresses, the type annotation of each translated segment is first identified (from the LLM model mentioned above). Then, segments of the corresponding type are selected from the sequence of translated segments according to the order defined in the standard representation template and added to the new sequence. For example, it first searches for "house number / building name" segments, then "street" segments, and so on. This way, the final place name and address translation output is not only lexically accurate, but also fully conforms to the target language's expression habits and standards in terms of overall structure and format, providing high-quality, directly applicable translation text.

[0079] In summary, a place name and address translation method based on the integrated artificial intelligence proper noun and common noun splitting of the embodiment of the present application is explained, which adopts artificial intelligence based on deep learning to perform semantic parsing and contextual semantic association reinforcement on the place name and address to be translated, and is conducive to the realization of proper noun and common noun splitting and type annotation of the place name and address to be translated based on the Prompt-driven LLM model, and generates a sequence of annotated place name and address fragments. Subsequently, based on a preset geographic terminology knowledge base, an iterative minimum translation unit determination is performed on each place name and address fragment, and through layer-by-layer query and comparison, recursive decomposition is performed on non-minimum unit fragments until all fragments meet the minimum translation granularity requirements, and then transliteration or standard translation conversion is performed on each translation unit based on the type annotation result, and a place name and address translation that meets the standard is output through normalization processing. This method can effectively solve the problems of inaccurate semantic splitting and insufficient standardization of translated names in traditional translation, thereby improving the quality of place name and address translation in scenarios such as cross-border map services and international logistics.

[0080] Furthermore, a place name and address translation tool that integrates artificial intelligence to split proper nouns and common nouns is also provided.

[0081] Figure 7 This is a block diagram of a place name and address translation tool that integrates artificial intelligence proper noun and common noun splitting according to an embodiment of the present application. Figure 7 As shown, the place name and address translation tool 100 with integrated artificial intelligence proper noun and common noun splitting according to the embodiment of the present application includes: a user input module 110, which is used to obtain the place name and address to be translated and the specified target translation language input by the user; a place name preprocessing module 120, which is used to use the LLM model to perform proper noun and common noun splitting and type annotation on the place name and address to be translated to obtain a sequence of annotated place name and address segments to be translated; a minimum translation unit recognition module 130, which is used to query each place name and address segment to be translated in the sequence of annotated place name and address segments based on a preset geographic terminology knowledge base. Whether the segment is a minimum translation unit, if not, the place name preprocessing module and the minimum translation unit identification module are cyclically called to process the place name and address segment to be translated to obtain a sequence of annotated minimum translation units of place name and address to be translated; the translation processing module 140 is used to translate each place name and address minimum translation unit to be translated in the sequence of annotated minimum translation units of place name and address to be translated to obtain a target language place name and address translation result; the translation result normalization module 150 is used to normalize the target language place name and address translation result to obtain a place name and address translation output.

[0082] The basic principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in the present invention are merely illustrative and non-limiting, and should not be construed as necessarily possessed by each embodiment of the present invention. Furthermore, the specific details of the above embodiments are provided for illustrative purposes and to facilitate understanding, and are not intended to be limiting. These details do not necessarily limit the present invention to being implemented using these specific details.

[0083] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0084] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be encompassed therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0085] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0086] Finally, it should be noted that the above description has been provided for the purpose of illustration and description. In addition, the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to be limiting. Although the technical solutions may be modified or replaced with equivalents with reference to the preferred embodiments, they do not depart from the spirit and scope of the technical solutions of the present invention.

Claims

1. A place name and address translation method integrating artificial intelligence proper noun and common noun splitting, characterized in that: include: S1, obtaining the place name and address to be translated and the specified target translation language input by the user; S2, using the LLM model to perform proper noun and common noun splitting and type annotation on the place name and address to be translated to obtain a sequence of annotated place name and address segments to be translated, which includes: Performing data preprocessing on the place name and address to be translated to obtain a standardized place name and address to be translated; Extracting contextual semantic features of the standardized place-name address to be translated to obtain a contextual semantic encoding vector of the place-name address to be translated; Performing semantic decomposition based on one-dimensional convolutional coding on the context semantic coding vector of the place name and address to be translated to obtain a sequence of local semantic feature coding vectors of the place name and address to be translated; Based on the contextual semantic association topology of the sequence of local semantic feature coding vectors of the place-name and address to be translated, performing feature distillation coding on each local semantic feature coding vector of the place-name and address to be translated in the sequence of local semantic feature coding vectors of the place-name and address to be translated to obtain a sequence of local semantic feature distillation coding vectors of the place-name and address to be translated; Performing context-association reconstruction on the sequence of the local semantic feature distillation coding vectors of the place name and address to be translated to obtain the context-semantic association enhanced coding vector of the place name and address to be translated; Inputting the context semantic association enhanced encoding vector of the place name and address to be translated into a proper noun and common noun splitting and annotation module based on the Prompt-based LLM model to obtain a sequence of the annotated place name and address segments to be translated; S3, based on a preset geographic terminology knowledge base, querying whether each of the to-be-translated place-name and address segments in the sequence of annotated place-name and address segments is a minimum translation unit; if not, iteratively performing steps S2 and S3 on the to-be-translated place-name and address segments to obtain a sequence of minimum translation units of the annotated place-name and address segments; S4, translating each of the minimum translation units of the place name and address to be translated in the sequence of the annotated minimum translation units of the place name and address to be translated to obtain a translation result of the place name and address in the target language; S5, normalizing the target language place name and address translation result to obtain a place name and address translation output.

2. The place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to claim 1 is characterized in that: Extracting the contextual semantic features of the standardized place name to be translated to obtain a contextual semantic encoding vector of the place name to be translated, including: The standardized place name and address to be translated is subjected to contextual semantic encoding based on the mBERT model to obtain a contextual semantic encoding vector of the place name and address to be translated.

3. The place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to claim 2 is characterized in that: Based on the contextual semantic association topology of the sequence of local semantic feature coding vectors of the place-name and address to be translated, feature distillation coding is performed on each local semantic feature coding vector of the place-name and address to be translated in the sequence of local semantic feature coding vectors of the place-name and address to be translated to obtain the sequence of local semantic feature distillation coding vectors of the place-name and address to be translated, including: Calculating the semantic correlation factor between any two local semantic feature coding vectors of the place-name and address to be translated in the sequence of the local semantic feature coding vectors of the place-name and address to be translated to obtain a local semantic feature correlation topology matrix of the place-name and address to be translated composed of a plurality of semantic correlation factors; Inputting the local semantic feature association topology matrix of the place name and address to be translated into a gated mask function to obtain a fine-grained association mask topology matrix of the local semantic feature of the place name and address to be translated; Based on the fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated, feature structure feedback modulation is performed on each local semantic feature encoding vector of the place name and address to be translated in the sequence of local semantic feature encoding vectors of the place name and address to be translated to obtain a sequence of distilled encoding vectors of the local semantic features of the place name and address to be translated.

4. The place name and address translation method integrating artificial intelligence proper noun and common noun separation according to claim 3 is characterized in that: Based on the fine-grained association mask topology matrix of the local semantic features of the place-name and address to be translated, feature structure feedback modulation is performed on each local semantic feature encoding vector of the place-name and address to be translated in the sequence of local semantic feature encoding vectors of the place-name and address to be translated to obtain a sequence of distilled encoding vectors of the local semantic features of the place-name and address to be translated, including: Performing local structural balance optimization on the fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated to obtain an optimized fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated; The local semantic feature encoding vector of the place name and address to be translated and the optimized fine-grained association mask topology matrix of the local semantic features of the place name and address to be translated are input into a feature dense feedback distillation unit to obtain the local semantic feature distillation encoding vector of the place name and address to be translated.

5. The place name and address translation method integrating artificial intelligence proper noun and common noun separation according to claim 4 is characterized in that: Performing context-related reconstruction on the sequence of the local semantic feature distilled encoding vectors of the place name and address to be translated to obtain the context-related semantic enhancement encoding vector of the place name and address to be translated, including: The sequence of the local semantic feature distillation encoding vectors of the place name and address to be translated is input into a feature reconstruction module based on the self-attention mechanism to obtain the contextual semantic association enhanced encoding vector of the place name and address to be translated.

6. The place name and address translation method integrating artificial intelligence proper noun and common noun splitting according to claim 1 is characterized in that: The step S4 comprises: Based on the type annotation of each place name and address minimum translation unit to be translated in the sequence of annotated place name and address minimum translation units and the target translation language, each place name and address minimum translation unit to be translated is transliterated or translated into a standard name to obtain the place name and address translation result in the target language.

7. A place name and address translation tool that integrates artificial intelligence to separate proper nouns and common nouns, characterized by: include: The user input module is used to obtain the place name and address to be translated and the specified target translation language input by the user; The place name and address preprocessing module is used to use the LLM model to split the place name and address to be translated into proper nouns and common nouns and to mark the type to obtain a sequence of marked place name and address segments to be translated, which includes: Performing data preprocessing on the place name and address to be translated to obtain a standardized place name and address to be translated; Extracting contextual semantic features of the standardized place-name address to be translated to obtain a contextual semantic encoding vector of the place-name address to be translated; Performing semantic decomposition based on one-dimensional convolutional coding on the context semantic coding vector of the place name and address to be translated to obtain a sequence of local semantic feature coding vectors of the place name and address to be translated; Based on the contextual semantic association topology of the sequence of local semantic feature coding vectors of the place-name and address to be translated, performing feature distillation coding on each local semantic feature coding vector of the place-name and address to be translated in the sequence of local semantic feature coding vectors of the place-name and address to be translated to obtain a sequence of local semantic feature distillation coding vectors of the place-name and address to be translated; Performing context-association reconstruction on the sequence of the local semantic feature distillation coding vectors of the place name and address to be translated to obtain the context-semantic association enhanced coding vector of the place name and address to be translated; Inputting the context semantic association enhanced encoding vector of the place name and address to be translated into a proper noun and common noun splitting and annotation module based on the Prompt-based LLM model to obtain a sequence of the annotated place name and address segments to be translated; A minimum translation unit identification module is used to query, based on a preset geographic terminology knowledge base, whether each place name and address segment to be translated in the sequence of annotated place name and address segments to be translated is a minimum translation unit; if not, cyclically calling the place name preprocessing module and the minimum translation unit identification module to process the place name and address segments to be translated, so as to obtain a sequence of minimum translation units of the annotated place name and address to be translated; A translation processing module, configured to perform translation processing on each of the minimum translation units of the place name and address to be translated in the sequence of the annotated minimum translation units of the place name and address to be translated to obtain a translation result of the place name and address in the target language; The translation result normalization module is used to perform normalization processing on the target language place name and address translation result to obtain the place name and address translation output.

Images