A multi-language rapid processing method for ERP system

Abstract

The present invention discloses a method for rapid multilingual processing for an ERP system, which relates to the field of multilingual processing technology. The method comprises collecting source language text and inputting the source language text into a constructed feature recognition model, wherein the feature recognition model establishes a bidirectional language mapping matrix based on a knowledge graph; segmenting the source language text using a semantic segmenter, extracting semantic feature groups based on the bidirectional language mapping matrix, and generating a semantic probability distribution; constructing a fusion transformation network, performing feature matching between the semantic probability distribution and a target language template library, and outputting a multilingual candidate sequence with confidence; screening the multilingual candidate sequences based on the confidence, and performing syntactic reorganization using a grammar optimizer, and updating the reorganized multilingual candidate sequences to the ERP system. By constructing a bidirectional language mapping matrix based on a knowledge graph and a fusion transformation network, the present invention achieves a key technological breakthrough in multilingual processing for ERP systems.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-language processing, in particular to a multi-language rapid processing method for an ERP system. Background Art

[0002] Multilingual processing technology has become a key enabler for bridging diverse language environments and facilitating international business communication. In recent years, with the rapid development of artificial intelligence and natural language processing, machine translation and intelligent language processing technologies have shown increasing promise in enterprise applications. However, traditional multilingual processing methods often suffer from technical bottlenecks such as insufficient semantic accuracy, low processing efficiency, and insufficient depth of contextual understanding.

[0003] Existing multilingual processing technologies primarily rely on statistical and rule-based translation models, which often encounter significant limitations when processing complex semantics and specialized terminology. Traditional methods typically employ static dictionary mappings or simple neural network translation models, which struggle to accurately capture subtle semantic differences and contextual associations between languages. Particularly in application scenarios such as ERP systems, which are highly sensitive to specialized terminology and business context, existing technologies often fail to achieve high-precision, low-latency, and rapid conversion of multilingual text. These technical shortcomings not only hinder effective communication between businesses but can also lead to semantic distortion and reduced translation quality of critical business information, thereby increasing the communication costs and potential risks of cross-language collaboration.

[0004] Therefore, a fast multilingual processing method for ERP system is proposed to solve the key problems of insufficient semantic accuracy and processing efficiency in traditional multilingual processing technology. Summary of the Invention

[0005] In view of the problems of insufficient semantic accuracy, low processing efficiency and insufficient depth of context understanding in existing multilingual processing technologies, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to achieve deep semantic mapping based on knowledge graphs, improve the accuracy and processing speed of multilingual text conversion in ERP systems, and effectively reduce the cost of semantic understanding in cross-language business communication.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a multilingual rapid processing method for an ERP system, comprising: collecting source language text and inputting the source language text into a constructed feature recognition model, wherein the feature recognition model establishes a bidirectional language mapping matrix based on a knowledge graph; segmenting the source language text using a semantic segmenter, extracting semantic feature groups based on the bidirectional language mapping matrix, and generating a semantic probability distribution; constructing a fusion transformation network, performing feature matching on the semantic probability distribution and a target language template library, and outputting a multilingual candidate sequence with confidence; screening the multilingual candidate sequence according to the confidence, performing syntactic reorganization using a grammar optimizer, and updating the reorganized multilingual candidate sequence to the ERP system.

[0009] As a preferred solution of the multilingual rapid processing method for the ERP system of the present invention, wherein: the multilingual candidate sequences are screened according to the confidence, and the syntax is reorganized by the grammar optimizer, and the reorganized multilingual candidate sequences are updated to the ERP system, including: by setting a dynamic threshold function based on historical translation quality data, the multilingual candidate sequences are arranged in descending order according to the confidence, and the minimum confidence requirement is determined; the adjacent candidate sequences are detected for repetition using a sliding window algorithm, and the multilingual candidate sequences with a confidence greater than the threshold and a repetition less than the set value are retained; a hybrid architecture based on rules and statistics is used to establish a grammar optimizer, wherein the grammar optimizer includes a grammar analyzer, a grammar analyzer, and a grammar analyzer. The method comprises the following steps: a grammatical analysis module, a structural reorganization module, and a grammatical verification module; constructing a syntactic tree for the screened multilingual candidate sequences through the syntactic analysis module; adjusting the sentence structure based on the grammatical norms and syntactic tree of the target language through the structural reorganization module, and extracting the syntactic dependencies of the multilingual candidate sequences; adjusting the positions of sentence components according to the word order rules of the target language, optimizing the hierarchical structure of long sentences, supplementing grammatical components, and verifying the adjusted and reorganized multilingual candidate sequences through the grammatical verification module; setting a life cycle and access rights for the verified multilingual candidate sequences, and simultaneously updating the data to several cache nodes through an asynchronous write mechanism to establish a mapping relationship between the source language text and the target language text.

[0010] As a preferred solution of the multilingual rapid processing method for the ERP system described in the present invention, the method for obtaining the multilingual candidate sequence is as follows: a fusion transformation network is established based on the encoder architecture and the decoder architecture, and a multi-layer bidirectional feature extraction module is used for the encoder architecture to capture the deep semantic features of the source language text, wherein the fusion transformation network includes an input layer and an output layer; the target language knowledge is integrated through the decoder architecture, and a target language template library is constructed based on a large-scale parallel corpus; based on the feature embedding matrix of the input layer, the probability distribution of the normalized language feature vector is mapped to a high-dimensional feature space; based on the high-dimensional feature space, a multi-headed feature extraction module is used to extract the target language template library. The cross-attention layer performs feature matching and calculates the similarity between the source language features and the target language template through the attention head. Based on the similarity, the input features are projected into different subspaces through the transformation matrix, the attention score is calculated, and the features are reorganized. Based on the feature reorganization, a residual connection structure is used to establish short-circuit connections between the layers of the fusion transformation network, so that shallow-layer features are transmitted to deep layers. The residual connection structure includes a feature addition module and a layer normalization module. The current layer output features and input features are added by the feature addition module, and the features are normalized by the layer normalization module. Based on the normalized features, a beam search algorithm is used to generate a multilingual candidate sequence with confidence. The confidence score is calculated based on the multilingual candidate sequence, and the confidence score is used to sort and screen the multilingual candidate sequences.

[0011] As a preferred embodiment of the multilingual fast processing method for an ERP system according to the present invention, the multi-head cross attention layer includes a transformation matrix; the transformation matrix includes a query transformation matrix, a key transformation matrix, and a value transformation matrix; the specific formula of the confidence score is as follows:

[0012] ;

[0013] in, is the source language feature vector, is the target language feature vector, is the total dimension of the feature

[0014] , is the eigenvalue of the i-th dimension of the source language, is the eigenvalue of the target language i-th dimension, is the feature weight of the i-th dimension

[0015] , is the angle between the source language feature vector and the target language feature vector, is the similarity enhancement function, the specific formula is as follows:

[0016] ;

[0017] in, is the feature difference vector, is the feature space dimension, is the edit distance, is the maximum allowed edit distance.

[0018] As a preferred solution of the multi-language rapid processing method for ERP system described in the present invention, wherein: the method for generating the language feature vector is to establish a semantic segmenter based on the hierarchical structure of the bidirectional long short-term memory network, identify the natural paragraph boundaries of the source language text through the semantic segmenter, and divide the long paragraph into text units based on the semantic coherence index, retaining the hierarchical relationship and context information between the natural paragraphs; extract grammatical structure markers through the bidirectional language mapping matrix, and match the extracted grammatical structure markers with the segmented text units, wherein the bidirectional language mapping matrix includes grammatical structure markers, word meaning relationship weights and context association rules; according to the word meaning relationship weights of the matching results, the word meaning relationship weights are used to match the text units. The semantic association strength between words is recalculated, and the contextual features of the text unit are analyzed based on the contextual association rules; the grammatical features, semantic association strength and contextual features are combined to form a semantic feature group; the semantic feature group is processed in parallel through a self-attention network to capture semantic relationships, and the dot product similarity between the query vector, the key vector and the value vector is calculated to obtain the attention weight between the features; the feature vectors are weighted and combined according to the attention weight to generate a language feature vector; the language feature vector is quantized and converted into a probability value using a softmax function, and the probability value is mapped to the interval [0,1] through a normalization method.

[0019] As a preferred solution of the multi-language rapid processing method for ERP system of the present invention, wherein: the grammatical structure tag includes part-of-speech tagging, syntactic dependency and grammatical component analysis results; the word meaning relationship weight represents the semantic relevance between words; the context association rule represents the usage pattern of words in different contexts; the language feature vector The specific formula is as follows:

[0020] ;

[0021] in, is the attention head index, is the weight of the p-th attention head, is the query transformation matrix of the p-th attention head, is the key transformation matrix of the p-th attention head, is the value transformation matrix of the p-th attention head, is the input feature, is the number of attention heads, is the scaling factor, is the activation function, is the attention calculation function.

[0022] As a preferred solution of the multilingual rapid processing method for the ERP system described in the present invention, the method for constructing the bidirectional language mapping matrix is ​​as follows: the source language text of the ERP system is collected, and the source language text is formatted by a text parser to remove special characters and redundant tags; the formatted source language text is divided into basic language units according to the syntactic structure, semantic annotation is performed, and entity relationships and attribute information are extracted; language knowledge nodes are constructed based on the annotated basic language units, and a network structure is formed between the nodes through semantic relationship edges to establish a knowledge graph, wherein the knowledge graph continuously optimizes the correlation strength between nodes through an iterative update mechanism; a feature recognition model is constructed based on a deep neural network model based on the Transformer architecture according to the knowledge graph, wherein the feature recognition model includes a multi-layer self-attention mechanism and an encoding layer; the multi-layer self-attention mechanism is used to capture the long-distance dependency relationship of the source language text; the encoding layer maps the source language text into a high-dimensional feature vector; based on the feature recognition model, language knowledge node pairs in the knowledge graph are extracted, the semantic similarity between the nodes is calculated, and mapping weights are set to construct a bidirectional language mapping matrix .

[0023] As a preferred solution of the multi-language rapid processing method for ERP system of the present invention, wherein: the bidirectional language mapping matrix The specific formula is as follows:

[0024] ;

[0025] in, is the source language word index, t is the target language word index, is the source language word vector, is the target language word vector, is the grammatical matching degree, is the position distance, is the distance attenuation coefficient, is the context relevance, is the normalized component, n is the vector dimension, and k is the number of knowledge graph relations.

[0026] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the multilingual rapid processing method for an ERP system as described in the first aspect of the present invention are implemented.

[0027] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the multilingual rapid processing method for an ERP system as described in the first aspect of the present invention are implemented.

[0028] The beneficial effects of the present invention are as follows: the present invention can accurately capture subtle semantic differences across languages ​​through a multi-head cross-attention layer and a residual connection structure; combined with a dynamic threshold function and a grammar optimizer, it can intelligently screen and reorganize multilingual candidate sequences to ensure the grammatical fluency and semantic consistency of the translation results; through the iterative update mechanism of the knowledge graph, it realizes continuous learning and adaptive optimization of the multilingual processing model, providing enterprises with a highly intelligent multilingual text rapid conversion solution, reducing the technical barriers and communication costs of cross-language business communication, and achieving a key technological breakthrough in multilingual processing of ERP systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0030] Figure 1 This is a flow chart of the multi-language rapid processing method for an ERP system according to Example 1. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0033] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0034] Example 1

[0035] Reference Figure 1, which is the first embodiment of the present invention, provides a multi-language rapid processing method for an ERP system, comprising:

[0036] S1: Collect source language text and input the source language text into a constructed feature recognition model, wherein the feature recognition model establishes a bidirectional language mapping matrix based on a knowledge graph.

[0037] Specifically, the method for constructing a bidirectional language mapping matrix is ​​to collect source language text from the ERP system, and use a text parser to format the source language text to remove special characters and redundant tags; the formatted source language text is divided into basic language units according to the syntactic structure, and semantic annotation is performed to extract entity relationships and attribute information.

[0038] Furthermore, language knowledge nodes are constructed based on the annotated basic language units, and the nodes are connected to form a network structure through semantic relationship edges to establish a knowledge graph, in which the knowledge graph continuously optimizes the association strength between nodes through an iterative update mechanism.

[0039] Furthermore, a deep neural network model based on the Transformer architecture constructs a feature recognition model based on the knowledge graph, where the feature recognition model includes a multi-layer self-attention mechanism and an encoding layer; the multi-layer self-attention mechanism is used to capture the long-distance dependencies of the source language text; the encoding layer maps the source language text into a high-dimensional feature vector.

[0040] Specifically, based on the feature recognition model, we extract the language knowledge node pairs in the knowledge graph, calculate the semantic similarity between the nodes, set the mapping weights, and construct a bidirectional language mapping matrix. , the specific formula is as follows:

[0041] ;

[0042] in, is the source language word index, t is the target language word index, is the source language word vector, is the target language word vector, is the grammatical matching degree, is the position distance, is the distance attenuation coefficient, is the context relevance, is the normalized component, n is the vector dimension, and k is the number of knowledge graph relations.

[0043] S2: Segmenting the source language text through a semantic segmenter, extracting semantic feature groups based on the bidirectional language mapping matrix, and generating a semantic probability distribution.

[0044] Specifically, the method for generating language feature vectors is to establish a semantic segmenter based on the hierarchical structure of a bidirectional long short-term memory network, identify the natural paragraph boundaries of the source language text through the semantic segmenter, and divide long paragraphs into text units based on semantic coherence indicators, while retaining the hierarchical relationship and contextual information between natural paragraphs.

[0045] Furthermore, grammatical structure markers are extracted through a bidirectional language mapping matrix, and the extracted grammatical structure markers are matched with the segmented text units; the semantic association strength between words is calculated according to the word-meaning relationship weights of the matching results, and the contextual features of the text units are analyzed based on the contextual association rules; the grammatical features, semantic association strength and contextual features are combined to form a semantic feature group.

[0046] It should be noted that the bidirectional language mapping matrix includes grammatical structure markers, word meaning relationship weights and context association rules; grammatical structure markers include part-of-speech tagging, syntactic dependency and grammatical component analysis results; word meaning relationship weights represent the degree of semantic relevance between words; context association rules represent the usage patterns of words in different contexts.

[0047] Furthermore, the semantic feature groups are processed in parallel through the self-attention network to capture the semantic relationship and calculate the dot product similarity between the query vector, key vector and value vector to obtain the attention weight between the features; the feature vectors are weighted and combined according to the attention weight to generate the language feature vector , the specific formula is as follows:

[0048] ;

[0049] in, is the attention head index, is the weight of the p-th attention head, is the query transformation matrix of the p-th attention head, is the key transformation matrix of the p-th attention head, is the value transformation matrix of the p-th attention head, is the input feature, is the number of attention heads, is the scaling factor, is the activation function, is the attention calculation function.

[0050] Specifically, the language feature vector is quantized and converted into a probability value using the softmax function, and the probability value is mapped to the [0, 1] interval through a normalization method.

[0051] Furthermore, the specific formula for the probability distribution of language feature vectors is as follows:

[0052] ;

[0053] in, is the word of the current index m, h is the index of the candidate word, is the index of the characteristic function, is the weight coefficient of the f-th characteristic function, is the fth characteristic function, is the total number of candidate words, is the word of candidate word h.

[0054] S3: Construct a fusion transformation network, perform feature matching between the semantic probability distribution and the target language template library, and output a multilingual candidate sequence with confidence.

[0055] Specifically, the method for obtaining multilingual candidate sequences is to establish a fusion transformation network based on the encoder architecture and the decoder architecture, and adopt a multi-layer bidirectional feature extraction module for the encoder architecture to capture the deep semantic features of the source language text, where the fusion transformation network includes an input layer and an output layer; integrate the target language knowledge through the decoder architecture, and build a target language template library based on large-scale parallel corpus; based on the feature embedding matrix of the input layer, map the normalized language feature vector probability distribution to a high-dimensional feature space.

[0056] It should be noted that the target language template library includes a language pair feature mapping table, a sentence structure template, and semantic conversion rules; the language pair feature mapping table stores the correspondence between the source language and the target language; the sentence structure template contains typical expressions of the target language; and the semantic conversion rules define the translation strategy in a specific context.

[0057] Furthermore, a multi-head cross-attention layer is used for feature matching based on the high-dimensional feature space, and the similarity between the source language features and the target language template is calculated through the attention head; based on the similarity, the input features are projected into different subspaces through the transformation matrix, the attention score is calculated and the features are reorganized; based on the feature reorganization, a short-circuit connection is established between the layers of the fusion transformation network through the residual connection structure, so that the shallow features are transferred to the deep layers; the output features of the current layer and the input features are added through the feature addition module, and the features are standardized through the layer normalization module.

[0058] It should be noted that the multi-head cross attention layer includes a transformation matrix; the transformation matrix includes a query transformation matrix, a key transformation matrix, and a value transformation matrix; the residual connection structure includes a feature summation module and a layer normalization module;

[0059] Furthermore, based on the standardized features, a beam search algorithm is used to generate multilingual candidate sequences with confidence. Confidence scores are calculated based on the multilingual candidate sequences, and the confidence scores are used to sort and screen the multilingual candidate sequences.

[0060] Specifically, the specific formula for the confidence score is as follows:

[0061] ;

[0062] in, is the source language feature vector, is the target language feature vector, is the total dimension of the feature

[0063] , is the eigenvalue of the i-th dimension of the source language, is the eigenvalue of the target language i-th dimension, is the feature weight of the i-th dimension

[0064] , is the angle between the source language feature vector and the target language feature vector, is the similarity enhancement function, the specific formula is as follows:

[0065] ;

[0066] in, is the feature difference vector, is the feature space dimension, is the edit distance, is the maximum allowed edit distance.

[0067] S4: Screening the multilingual candidate sequences according to the confidence level, performing syntactic reorganization through a grammar optimizer, and updating the reorganized multilingual candidate sequences to the ERP system.

[0068] Specifically, by setting a dynamic threshold function based on historical translation quality data, multilingual candidate sequences are sorted in descending order according to confidence, and the minimum confidence requirement is determined; a sliding window algorithm is used to detect the repetition of adjacent candidate sequences, and multilingual candidate sequences with confidence greater than the threshold and repetition less than the set value are retained.

[0069] It should be noted that when the system receives a new multilingual candidate sequence, it immediately calculates the confidence score of each sequence and sorts all sequences in descending order of confidence; if the confidence of a multilingual candidate sequence is less than a dynamic threshold set based on historical translation quality, it will be deleted from the candidate queue; if the confidence of a multilingual candidate sequence is greater than or equal to the dynamic threshold set based on historical translation quality, a sliding window is started to calculate the repetition of adjacent sequences; if the repetition between adjacent sequences is greater than the set threshold, the multilingual candidate sequence with the highest confidence is retained; if the difference in confidence between repeated sequences is less than a preset difference, the multilingual candidate sequence with the highest position is retained; if the repeated content spans several multilingual candidate sequences, the entire set of sequences is evaluated to retain the optimal sequence combination; if the repetition between adjacent sequences is less than or equal to the set threshold, the screened multilingual candidate sequence is added to the sequence pool; if the number of sequences in the sequence pool is greater than the capacity limit, the sequence with the lowest confidence is removed.

[0070] Furthermore, a grammar optimizer is established by adopting a hybrid architecture combining rule-based and statistical methods, wherein the grammar optimizer includes a syntactic analysis module, a structural reorganization module and a grammar verification module; a syntactic tree is constructed for the screened multilingual candidate sequences through the syntactic analysis module; based on the grammatical norms and syntactic tree of the target language, the sentence structure is adjusted through the structural reorganization module to extract the syntactic dependencies of the multilingual candidate sequences; the positions of sentence components are adjusted according to the word order rules of the target language, the hierarchical structure of long sentences is optimized, grammatical components are supplemented, and the adjusted and reorganized multilingual candidate sequences are verified through the grammar verification module.

[0071] It should be noted that when the sentence length of a multilingual candidate sequence is greater than a preset threshold, the sentence is decomposed into several sentences for processing; when the sentence length of a multilingual candidate sequence is less than or equal to the preset threshold, the structure is reorganized based on the grammatical norms of the target language; if a component that does not conform to the word order of the target language is detected, the position is adjusted according to the word order rules; if a component that conforms to the word order of the target language is detected, the grammar verification phase is entered; if a grammatical error is found, corresponding corrections are made according to the error type; if multiple possible expressions appear, the multilingual candidate sequence with the highest confidence is selected for marking, and this content is manually reviewed.

[0072] Furthermore, the verified multilingual candidate sequences are given lifecycles and access permissions, and the data is simultaneously updated to several cache nodes through an asynchronous write mechanism to establish a mapping relationship between the source language text and the target language text.

[0073] Furthermore, this embodiment also provides a computer device suitable for the multi-language rapid processing method for an ERP system, comprising a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-language rapid processing method for an ERP system as proposed in the above embodiment.

[0074] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0075] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the following steps: collecting source language text and inputting the source language text into a constructed feature recognition model, wherein the feature recognition model establishes a bidirectional language mapping matrix based on a knowledge graph; segmenting the source language text using a semantic segmenter, extracting semantic feature groups based on the bidirectional language mapping matrix, and generating a semantic probability distribution; constructing a fusion transformation network, performing feature matching between the semantic probability distribution and a target language template library, and outputting a multilingual candidate sequence with a confidence score; screening the multilingual candidate sequence based on the confidence score, performing syntactic reorganization using a grammar optimizer, and updating the reorganized multilingual candidate sequence to an ERP system.

[0076] In summary, the present invention uses a multi-head cross-attention layer and a residual connection structure to accurately capture subtle semantic differences across languages. Combined with a dynamic threshold function and a grammar optimizer, it can intelligently screen and reorganize multilingual candidate sequences to ensure the grammatical fluency and semantic consistency of the translation results. Through the iterative update mechanism of the knowledge graph, it enables continuous learning and adaptive optimization of the multilingual processing model, providing enterprises with a highly intelligent solution for rapid multilingual text conversion, reducing the technical barriers and communication costs of cross-language business communication, and achieving a key technological breakthrough in multilingual processing in ERP systems.

[0077] Example 2

[0078] Referring to Table 1, which is a second embodiment of the present invention, this embodiment provides a multi-language rapid processing method for an ERP system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0079] Specifically, the test corpus consisted of 6,000 business texts from a multinational enterprise's ERP system, covering six languages: English, Chinese, Japanese, German, French, and Spanish. To ensure data representativeness, the test corpus included text from multiple domains, including product descriptions, technical specifications, business processes, and customer feedback. The experimental environment utilized a server equipped with an Intel Xeon 8-core CPU, 64GB of memory, and an NVIDIA Tesla V100 GPU.

[0080] Furthermore, data was collected from the ERP system at a rate of 200 records per minute to ensure system load balancing. The Stanford CoreNLP toolkit was used to preprocess the source language text, including word segmentation, part-of-speech tagging, and dependency syntax analysis. During the knowledge graph construction phase, the Neo4j graph database was used to store language knowledge nodes, and Python scripts were used to automate the construction and updating of the knowledge graph. The association strength between nodes was calculated based on co-occurrence frequency and semantic similarity, with cosine similarity used as the baseline metric.

[0081] Furthermore, the feature recognition model uses a 12-layer Transformer encoder structure, with each layer containing 8 attention heads and a hidden layer dimension of 512. The model is trained using the Adam optimizer with a learning rate of 0.0001 and a batch size of 64. Weight decay and dropout mechanisms are introduced during training to prevent overfitting. The semantic segmenter is built based on a BiLSTM network with a hidden layer dimension of 256 and optimized using the cross-entropy loss function.

[0082] Furthermore, the fusion transform network adopts an encoder-decoder architecture, with the encoder and decoder each consisting of six layers, each using 16 attention heads. The target language template library is constructed based on 10 million parallel corpus pairs, using the TF-IDF algorithm for feature matching. The beam search algorithm uses a beam width of 4 to generate a diverse range of translation candidates. The grammar optimizer uses a rule-based syntactic restructuring method combined with statistical models for grammatical verification to ensure the fluency and accuracy of the output text.

[0083] Specifically, as shown in Table 1, the proposed method achieves a BLEU score of 42.6 in terms of translation quality, a 36.5% improvement over the 31.2 score of the traditional method. This improvement is primarily due to the knowledge graph and bidirectional language mapping matrix introduced by the proposed method, which enable the system to more accurately understand and convert the semantic content of the source language text.

[0084] Table 1. Comparison between the present invention and traditional methods

[0085] Performance indicators Traditional methods Method of the present invention Performance improvement ratio% BLEU score 31.2 42.6 36.5 Translation speed (words / second 156 242 55.1 Memory usage GB 45 35 -22.2 Accuracy% 82.5 91.3 10.7 Recall rate% 80.3 89.7 11.7 F1 score % 81.4 90.5 11.2 Average response time in ms 85 55 -35.3

[0086] Furthermore, in terms of system performance, the present invention achieves a processing speed of 242 words per second, a 55.1% improvement over the 156 words per second of traditional methods. Furthermore, through an optimized architectural design, the present invention's memory usage is only 35GB, a 22.2% reduction compared to traditional methods, demonstrating greater resource efficiency. The present invention's average response time is only 55ms, a 35.3% reduction compared to the 85ms of traditional methods. This rapid response capability is of great practical significance for applications requiring real-time translation processing.

[0087] Furthermore, in terms of translation accuracy, the accuracy rate increased from 82.5% of the traditional method to 91.3%, an increase of 10.7 percentage points; the recall rate increased from 80.3% to 89.7%, an increase of 11.7 percentage points; and the F1 score increased from 81.4% to 90.5%, an increase of 11.2 percentage points. These data fully demonstrate that the present invention can not only accurately identify and convert the semantic features of the source language text, but also maintain a high level of translation coverage.

[0088] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-language rapid processing method for an ERP system, characterized by: include, Collecting source language text and inputting the source language text into a constructed feature recognition model, wherein the feature recognition model establishes a bidirectional language mapping matrix based on a knowledge graph; Segmenting the source language text using a semantic segmenter, extracting semantic feature groups based on the bidirectional language mapping matrix, and generating a semantic probability distribution; Constructing a fusion transformation network to perform feature matching between the semantic probability distribution and the target language template library, and outputting a multilingual candidate sequence with confidence; The multilingual candidate sequences are screened according to the confidence level, and syntax is reorganized by a grammar optimizer, and the reorganized multilingual candidate sequences are updated to the ERP system.

2. The multi-language rapid processing method for an ERP system according to claim 1, characterized in that: The multilingual candidate sequences are screened according to the confidence level, and syntax is reorganized by a grammar optimizer, and the reorganized multilingual candidate sequences are updated to the ERP system, including: By setting a dynamic threshold function based on historical translation quality data, multilingual candidate sequences are sorted in descending order of confidence to determine the minimum confidence requirement; Use the sliding window algorithm to detect the repetition of adjacent candidate sequences, and retain the multilingual candidate sequences with a confidence level greater than a threshold and a repetition level less than a set value; A grammar optimizer is established by adopting a hybrid architecture combining rule-based and statistical methods, wherein the grammar optimizer includes a syntax analysis module, a structure reorganization module, and a grammar checking module; Constructing a syntax tree for the screened multilingual candidate sequences through the syntax analysis module; Based on the grammatical norms and syntax tree of the target language, the sentence structure is adjusted by the structure reorganization module to extract the syntactic dependency of the multilingual candidate sequences; Adjusting the positions of sentence components according to the word order rules of the target language, optimizing the hierarchical structure of long sentences, supplementing grammatical components, and verifying the adjusted and reorganized multilingual candidate sequences through the grammar verification module; The verified multilingual candidate sequences are assigned lifecycles and access permissions, and the data is simultaneously updated to several cache nodes through an asynchronous write mechanism to establish a mapping relationship between the source language text and the target language text.

3. The multi-language rapid processing method for an ERP system according to claim 2, wherein: The method for obtaining the multilingual candidate sequence is: Establishing a fusion transformation network based on an encoder architecture and a decoder architecture, and applying a multi-layer bidirectional feature extraction module to the encoder architecture to capture deep semantic features of the source language text, wherein the fusion transformation network includes an input layer and an output layer; Integrate target language knowledge through the decoder architecture and build a target language template library based on large-scale parallel corpus; Based on the feature embedding matrix of the input layer, mapping the normalized language feature vector probability distribution to a high-dimensional feature space; Perform feature matching using a multi-head cross-attention layer based on the high-dimensional feature space, and calculate the similarity between the source language features and the target language template through the attention heads; Based on the similarity, the input features are projected into different subspaces through the transformation matrix, the attention scores are calculated and the features are reorganized; Based on the feature reorganization, short-circuit connections are established between the layers of the fusion transformation network through a residual connection structure, so that shallow layer features are transferred to deep layers, wherein the residual connection structure includes a feature summation module and a layer normalization module; Adding the current layer output features and input features through the feature summation module, and normalizing the features through the layer normalization module; Based on the standardized features, a beam search algorithm is used to generate multilingual candidate sequences with confidence. A confidence score is calculated based on the multilingual candidate sequences, wherein the confidence score is used for sorting and screening the multilingual candidate sequences.

4. The multi-language rapid processing method for an ERP system according to claim 3, wherein: The multi-head cross attention layer includes a transformation matrix; the transformation matrix includes a query transformation matrix, a key transformation matrix, and a value transformation matrix; the specific formula of the confidence score is as follows: ； in, is the source language feature vector, is the target language feature vector, is the total dimension of the feature is the eigenvalue of the i-th dimension of the source language, is the eigenvalue of the target language i-th dimension, is the feature weight of the i-th dimension is the angle between the source language feature vector and the target language feature vector, is the similarity enhancement function, the specific formula is as follows: ； in, is the feature difference vector, is the feature space dimension, is the edit distance, is the maximum allowed edit distance.

5. The multi-language rapid processing method for an ERP system according to claim 3, wherein: The method for generating the language feature vector is: A semantic segmenter is established based on the hierarchical structure of a bidirectional long short-term memory network. The semantic segmenter identifies natural paragraph boundaries in the source language text and segments long paragraphs into text units based on semantic coherence indicators, preserving the hierarchical relationship and contextual information between natural paragraphs. Extracting grammatical structure markers through a bidirectional language mapping matrix, and matching the extracted grammatical structure markers with the segmented text units, wherein the bidirectional language mapping matrix includes grammatical structure markers, word meaning relationship weights, and context association rules; Calculating the semantic association strength between words based on the word meaning relationship weights of the matching results, and analyzing the contextual features of the text unit based on the context association rules; Combining grammatical features, semantic association strength, and contextual features to form semantic feature groups; The semantic feature groups are processed in parallel through a self-attention network to capture semantic relationships, and the dot product similarity between the query vector, the key vector, and the value vector is calculated to obtain the attention weights between the features; Performing weighted combination on the feature vectors according to the attention weights to generate a language feature vector; The language feature vector is quantized and converted into a probability value by using a softmax function, and the probability value is mapped to the interval [0, 1] by a normalization method.

6. The multi-language rapid processing method for an ERP system according to claim 5, characterized in that: The grammatical structure tag includes part-of-speech tagging, syntactic dependency and grammatical component analysis results; the word meaning relationship weight represents the semantic relevance between words; the context association rule represents the usage rules of words in different contexts; the language feature vector The specific formula is as follows: ； in, is the attention head index, is the weight of the p-th attention head, is the query transformation matrix of the p-th attention head, is the key transformation matrix of the p-th attention head, is the value transformation matrix of the p-th attention head, is the input feature, is the number of attention heads, is the scaling factor, is the activation function, is the attention calculation function.

7. The multi-language rapid processing method for an ERP system according to claim 5, wherein: The method for constructing the bidirectional language mapping matrix is: Collecting source language text from the ERP system and formatting the source language text using a text parser to remove special characters and redundant tags; The formatted source language text is divided into basic language units according to the syntactic structure, semantic annotation is performed, and entity relationships and attribute information are extracted; Construct language knowledge nodes based on the annotated basic language units, and connect the nodes to form a network structure through semantic relationship edges to establish a knowledge graph, wherein the knowledge graph continuously optimizes the association strength between nodes through an iterative update mechanism; A deep neural network model based on the Transformer architecture constructs a feature recognition model based on the knowledge graph, wherein the feature recognition model includes a multi-layer self-attention mechanism and an encoding layer; the multi-layer self-attention mechanism is used to capture the long-range dependencies of the source language text; the encoding layer maps the source language text into a high-dimensional feature vector; Based on the feature recognition model, the language knowledge node pairs in the knowledge graph are extracted, the semantic similarity between the nodes is calculated, and the mapping weights are set to construct a bidirectional language mapping matrix. .

8. The multi-language rapid processing method for an ERP system according to claim 7, characterized in that: The bidirectional language mapping matrix The specific formula is as follows: ； in, is the source language word index, t is the target language word index, is the source language word vector, is the target language word vector, is the grammatical matching degree, is the position distance, is the distance attenuation coefficient, is the context relevance, is the normalized component, n is the vector dimension, and k is the number of knowledge graph relations.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multilingual rapid processing method for an ERP system according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multilingual rapid processing method for an ERP system according to any one of claims 1 to 8 are implemented.

Images