A multi-dimensional cross-cultural English communication ability evaluation method, system, device and medium

Abstract

The application relates to the technical field of data processing, in particular to a multi-dimensional cross-cultural English communication ability evaluation method, system, device and medium. The method comprises the following steps: obtaining conversation scene data, and presenting a conversation scene to an evaluation object based on the conversation scene data, wherein the conversation scene data comprises controllable cultural conflict elements. Multimodal behavior data generated by the evaluation object in the conversation process is collected. First feature data associated with the controllable cultural conflict elements and second feature data irrelevant to the controllable cultural conflict elements are extracted based on the multimodal behavior data. Fusion analysis is performed based on the first feature data and the second feature data, and a comprehensive evaluation index representing the cross-cultural adaptation ability of the evaluation object is obtained. The method can comprehensively and objectively quantify the performance of the evaluation object in cultural conflict response, language use, non-verbal behavior and other multi-dimensional aspects, significantly reduces the evaluation bias, and improves the reliability and validity of the evaluation.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a multi-dimensional cross-cultural English communicative competence assessment method, system, device, and medium. Background Technology

[0002] In existing technologies, the assessment of cross-cultural English communicative competence often relies on static language tests or highly subjective expert ratings, lacking effective simulation of dynamic and real-world cultural conflict situations. At the same time, existing methods usually only focus on single-modal data such as language expression, failing to comprehensively and synchronously collect and integrate multi-dimensional information such as non-verbal behavior and emotional responses, resulting in one-sided and insufficiently objective assessment results. Summary of the Invention

[0003] To overcome the shortcomings of the existing technology, this application provides a multi-dimensional cross-cultural English communicative competence assessment method, system, device and medium, aiming to solve the problems of existing English communicative competence assessment methods relying on single data and incomplete dimensions, resulting in low assessment accuracy and insufficient objectivity.

[0004] The first aspect of this application provides a multi-dimensional cross-cultural English communicative competence assessment method, the method comprising:

[0005] Acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data, wherein the conversation scene data includes controllable cultural conflict elements; Collect and evaluate multimodal behavioral data generated by the subject during the session; Based on the multimodal behavioral data, extract first feature data associated with the controllable cultural conflict elements and second feature data unrelated to the controllable cultural conflict elements; Based on the fusion analysis of the first feature data and the second feature data, a comprehensive evaluation index representing the cross-cultural adaptability of the evaluation object is obtained.

[0006] Optionally, obtaining the session scenario data includes: Acquire quantitative indicators for scenario type and target cultural dimension; Determine the initial session scenario corresponding to the scenario type; Based on the quantitative indicators of the target cultural dimension, the controllable cultural conflict elements for the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements are determined. Based on the controllable cultural conflict elements, the scene injection time, and the initial conversation scene, conversation scene data is generated.

[0007] Optionally, determining the controllable cultural conflict elements for the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements based on the quantitative indicators of the target cultural dimension includes: Based on the quantitative indicators of the target cultural dimension, one or more behavioral elements are retrieved from the preset cultural dimension-conflict behavior mapping library; A candidate element set is constituted based on the one or more behavioral elements; Based on the scenario type, target elements that match the initial conversation scenario are selected from the candidate element set as controllable cultural conflict elements. Identify key session nodes in the initial session scenario, and determine the scene injection time of the controllable cultural conflict element based on the key session nodes.

[0008] Optionally, the first feature data includes attention shift features related to cultural conflict, and the methods for extracting the attention shift features include: The time of scene injection of the controllable cultural conflict elements is taken as the reference time point; Based on the eye-tracking data in the multimodal behavioral data, determine the time point at which the gaze focus of the assessment subject shifts to a new focus associated with the controllable cultural conflict element; The time difference between the reference time point and the transfer time point is calculated to obtain the attention transfer feature.

[0009] Optionally, the first feature data includes semantic content appropriateness features, and the extraction methods for the semantic content appropriateness features include: Convert the audio data in the multimodal behavioral data into text data; The text data was analyzed using a natural language processing model trained on a cross-cultural corpus to obtain a semantic appropriateness score. The semantic appropriateness feature is obtained based on the semantic appropriateness score.

[0010] Optionally, the extraction method for the second feature data includes: From the audio data in the multimodal behavioral data, extract speech fluency features, terminology accuracy features, and sentence complexity features before the injection of the controllable cultural conflict elements; From the visual behavior data in the multimodal behavior data, extract the behavioral coordination features before the injection of the controllable cultural conflict elements; The second feature data is obtained based on the speech fluency feature, the terminology accuracy feature, the sentence complexity feature, and the behavior coordination feature.

[0011] Optionally, the comprehensive assessment index characterizing the cross-cultural adaptability of the assessment object by performing fusion analysis based on the first feature data and the second feature data includes: The first feature data and the second feature data are standardized respectively; Determine the weight coefficients corresponding to the first and second feature data after standardization. The comprehensive evaluation index is obtained by calculating the first and second feature data after standardization and their corresponding weight coefficients.

[0012] A second aspect of this application provides a multi-dimensional cross-cultural English communicative competence assessment device, the device comprising: The presentation module is used to acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data, wherein the conversation scene data includes controllable cultural conflict elements; The data acquisition module is used to collect multimodal behavioral data generated by the evaluation object during the session; The extraction module is used to extract first feature data associated with the controllable cultural conflict elements and second feature data unrelated to the controllable cultural conflict elements based on the multimodal behavioral data. The assessment module is used to perform fusion analysis based on the first feature data and the second feature data to obtain a comprehensive assessment index that characterizes the cross-cultural adaptability of the assessment object.

[0013] A third aspect of this application provides an electronic device comprising a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement the multidimensional cross-cultural English communicative competence assessment method as described above.

[0014] A fourth aspect of this application provides a computer-readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the multidimensional cross-cultural English communicative competence assessment method as described above.

[0015] This application has at least the following beneficial effects: The multi-dimensional cross-cultural English communicative competence assessment method provided in this application constructs an assessment environment that highly simulates real cross-cultural communicative situations by introducing conversational scenario data containing controllable cultural conflict elements. This effectively stimulates and captures the actual behavioral responses of the assessed individuals under cultural conflict. By collecting multimodal behavioral data generated during the conversation and further distinguishing and extracting primary feature data associated with controllable cultural conflict elements and secondary feature data unrelated to them, targeted identification and isolation of factors influencing cross-cultural adaptability are achieved. Based on this, by fusing and analyzing the two types of feature data, it is possible not only to comprehensively and objectively quantify the assessed individuals' performance in multiple dimensions such as cultural conflict coping, language use, and non-verbal behavior, but also to significantly reduce assessment bias caused by individual general communication habits or scenario-irrelevant factors. Ultimately, a comprehensive assessment index characterizing their cross-cultural adaptability is obtained, and this method significantly improves the reliability and validity of the assessment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating a multi-dimensional cross-cultural English communicative competence assessment method provided in Embodiment 1 of this application; Figure 2 A schematic diagram of the multi-dimensional cross-cultural English communication competence assessment device provided in Embodiment 2 of this application; Figure 3 This is a schematic diagram of the electronic device provided in Embodiment 3 of this application. Detailed Implementation

[0018] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. Unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the specification of this application is for the purpose of describing an embodiment in one alternative implementation and is not intended to be limiting of the application.

[0020] Example 1 like Figure 1 As shown, this application provides a multi-dimensional cross-cultural English communicative competence assessment method, including: S1. Acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data, which includes controllable cultural conflict elements.

[0021] Specifically, S1 "Acquire Session Context Data" includes: S11, obtain quantitative indicators of scenario type and target cultural dimension.

[0022] The scenario type is selected from a pre-set scenario library. The quantitative indicators for the target cultural dimension are set based on mainstream cross-cultural theoretical models such as Hofstede's cultural dimension theory, and are assigned specific quantitative target values. The indicators and types can be entered by the evaluation system administrator through the configuration interface or loaded from pre-configured evaluation schemes.

[0023] Examples of scenarios include cross-cultural negotiations, dispute resolution in international team projects, and clarifying needs with clients from high-context cultural backgrounds. The target cultural dimension quantification indicators include two dimensions: power distance and uncertainty aversion, and assign specific quantitative target values ​​to them, such as simulating communicators in cultural contexts with high power distance (PDI=80) and low uncertainty aversion (UAI=30).

[0024] Step S11 limits the scope of the assessment context by selecting a specific scenario type; by setting quantifiable cultural dimension indicators, the subsequent simulation of cultural conflicts has a theoretical basis and a controllable, standardized intensity value.

[0025] S12, determine the initial session scenario corresponding to the scenario type.

[0026] The initial conversation scenario is a neutral English dialogue script that conforms to the basic logic of the selected scenario type.

[0027] For example, in a cross-cultural negotiation scenario, the initial conversation may include text descriptions and role-playing dialogue scripts for the basic steps such as opening greetings, statements of positions by both parties, bargaining, concessions and compromises, and reaching a preliminary consensus. These scripts are stored in a scenario template library and do not contain deliberate or strong points of cultural conflict to ensure the universality of their basic structure.

[0028] This step S12 provides a foundation for injecting controllable cultural conflict elements by determining the initial conversation scenario corresponding to the scenario type.

[0029] S13, based on the quantitative indicators of the target cultural dimension, determine the controllable cultural conflict elements for the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements.

[0030] The system uses quantitative indicators of the target cultural dimension to retrieve a list of potential conflict behaviors from a pre-set library, filters out controllable cultural conflict elements, and analyzes the text logic of the initial conversation scenario to determine the timing of injecting controllable cultural conflict elements into the scenario.

[0031] For example, if the target dimension is "high power distance", the system may select "unilaterally setting a deadline using an imperative tone" as a conflict element from the mapping library and inject it into the key node "both parties negotiate the time arrangement" in the initial scenario.

[0032] This step S13 automatically and structurally transforms abstract cultural theory indicators into specific, embeddable standard stimuli in the conversation flow, ensuring that the introduction of cultural conflict in the assessment is theoretically sound, standardized, and repeatable, thus laying the foundation for subsequent objective measurement.

[0033] S14 generates conversation scene data based on controllable cultural conflict elements, scene injection time, and initial conversation scene.

[0034] The system inserts selected controllable cultural conflict elements (such as a specific line of dialogue or a specific nonverbal behavioral instruction) and their injection timestamps into the corresponding positions in the initial conversation scene in the form of structured data tags. The final generated conversation scene data is a complete data package containing dialogue text sequences, character behavioral instructions, conflict element tags, and timestamps, which can be used to drive the subsequent evaluation interface or virtual character presentation.

[0035] This application, through steps S11-S14, pre-obtains the scenario type and quantified target cultural dimension indicators, ensuring that the constructed initial conversation scenario has clear targeting and cultural orientation. Based on this, controllable cultural conflict elements and their specific injection points in the conversation flow are precisely determined according to the quantitative indicators of the target cultural dimension, achieving a scientific and reproducible integration and scheduling between cultural conflict elements and scenario content. This technical approach makes the generated conversation scenario data highly structured and standardized, ensuring consistency in the situations faced by different evaluation subjects, thereby greatly improving the fairness and horizontal comparability of the evaluation results. Simultaneously, through refined and quantitative control of the conflict element type and introduction timing, it can systematically simulate cultural conflicts of different intensities and dimensions, providing a high-quality input foundation with clear layers and variables for subsequent multimodal behavioral data collection and analysis, significantly enhancing the scientific rigor, systematicity, and depth of the entire evaluation process and its conclusions.

[0036] Furthermore, S13, based on the quantitative indicators of the target cultural dimension, determines the controllable cultural conflict elements used in the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements, including: S131, based on the quantitative indicators of the target cultural dimension, retrieve one or more behavioral elements from the preset cultural dimension-conflict behavior mapping library.

[0037] The cultural dimension-conflict behavior mapping library is a structured database that stores the correspondence between specific behavioral descriptions such as "high power distance (PDI>70)" and "frequent use of imperative statements, directly interrupting the other party." Based on the input PDI=80 index, the system automatically retrieves a list of all behavioral elements associated with that high value.

[0038] S132, a candidate element set is formed based on one or more behavioral elements.

[0039] S133, based on scenario type, selects target elements that match the initial conversation scenario from the candidate element set as controllable cultural conflict elements.

[0040] For example, from the retrieved "high power distance" behavioral elements, combined with the "negotiation" scenario, the system may filter out the specific behavior of "simulating the other party's role to interrupt our statement with an absolute tone of 'This is the final offer, no discussion allowed' during the negotiation process" and mark it as a controllable cultural conflict element.

[0041] S134, identify key conversation nodes in the initial conversation scenario, and determine the scene injection time of controllable cultural conflict elements based on the key conversation nodes.

[0042] The system analyzes the initial conversation script using natural language processing technology, identifies the position of the statement "during the negotiation process, the two parties first have a significant disagreement on the price terms", and sets this time point as the scenario injection time T0 for the aforementioned conflict element.

[0043] Steps S131-S134 of this application transform specific, operable, and embeddable behavioral stimuli into the conversation flow through structured mapping and scenario-based filtering. The pipeline approach of theoretical quantification → behavioral mapping → scenario adaptation → timing anchoring ensures that each generated evaluation conflict is standardized, reproducible, and highly relevant to the evaluation objective, laying the foundation for subsequent objective measurement.

[0044] S2 collects multimodal behavioral data generated by the evaluation object during the session.

[0045] Specifically, when the assessment subject interacts with a simulated interlocutor (which could be a recorded video, a virtual character, or a real evaluator) through a computer interface, the system synchronously collects multimodal behavioral data. This includes at least: 1) Audio data: all speech of the assessment subject is collected via a microphone, with a sampling rate of at least 16kHz; 2) Visual behavioral data: facial expressions, upper body gestures, and postures of the assessment subject are recorded via a camera, with a video frame rate of at least 25fps; 3) Eye-tracking data: the assessment subject's gaze points, gaze duration, and saccade paths on a specific area of ​​interest (AOI) on the screen are collected via an integrated or external eye tracker (such as the Tobii eye tracker), with a sampling rate determined based on device performance (e.g., at least 60Hz). All data streams are timestamped by a time synchronization server to ensure time alignment for subsequent analysis.

[0046] Step S2 aims to comprehensively and objectively record the real-time responses of the assessment subjects to standardized stimuli, providing a raw data foundation for quantitative analysis.

[0047] S3, based on multimodal behavioral data, extracts first feature data associated with controllable cultural conflict elements and second feature data unrelated to controllable cultural conflict elements.

[0048] The goal of step S3 is to extract two types of features from the raw data: one type directly reflects the response to cultural conflict stimuli (first feature data), and the other type reflects the baseline capabilities of the assessment subject in the absence of conflict interference (second feature data).

[0049] In one embodiment, the first feature data includes attention shift features related to cultural conflict, S31, the methods for extracting attention shift features include: S311, with the time of scene injection of controllable cultural conflict elements as the benchmark time point T0.

[0050] S312, Based on eye-tracking data in multimodal behavioral data, determine the time point T1 when the subject's gaze shifts to a new focus associated with a controllable cultural conflict element.

[0051] S313, calculate the time difference Δt=T1-T0 between the reference time point and the transfer time point to obtain the attention transfer characteristics.

[0052] The "new focus associated with controllable cultural conflict elements" is predefined in the system. For example, if the conflict element is "the other party presenting a chart with obvious cultural symbolism," the area on the screen displaying the chart is defined as the area of ​​interest (AOI). The system uses eye-tracking data analysis algorithms to determine when the object of evaluation first enters and stabilizes (e.g., maintains a gaze for more than 200ms) within the predefined AOI; this moment is T1.

[0053] The smaller the time difference Δt, the faster the assessed subject captures attentional signals of cultural conflict. Attentional shift characteristics quantify an individual's early cognitive alertness to potential cultural conflict cues.

[0054] This application, through the design of steps S311-S313, uses the preset injection time of controllable cultural conflict elements as a benchmark, and simultaneously utilizes eye-tracking data to capture the specific time point at which the assessment subject's gaze focus actually shifts to the conflict-related elements, thereby calculating the time difference between the two to obtain attention shift characteristics. This method effectively overcomes the limitations of traditional assessments that rely on subjective observation or post-event self-report to infer attention changes, achieving millisecond-level objective measurement of the speed and efficiency of an individual's attention resource allocation under cross-cultural conflict stimuli. The attention shift characteristics extracted in this way, as key quantitative indicators in the first feature data, not only significantly enhance the objectivity and precision of cultural conflict response behavior analysis, but also provide highly reliable time-series behavioral data for subsequent fusion analysis, enabling the comprehensive assessment index to more sensitively and reliably reflect the assessment subject's subconscious cultural sensitivity and immediate cognitive adjustment ability, thereby comprehensively improving the scientific depth and accuracy of cross-cultural adaptability assessment.

[0055] In another embodiment, the first feature data includes semantic content appropriateness features, S32, the extraction method of semantic content appropriateness features includes: S321 converts audio data in multimodal behavioral data into text data.

[0056] For example, the English audio stream of the evaluation subject can be converted into text in real time or post-conversion using an automatic speech recognition engine, such as iFlytek, Google Speech-to-Text API, or an open-source engine such as DeepSpeech. After conversion, basic text cleaning is required, such as removing filler words (um, ah) and correcting recognition errors.

[0057] S322 uses a natural language processing model trained on a cross-cultural corpus to analyze text data and obtain a semantic appropriateness score.

[0058] The NLP model can be a text classification or sequence labeling model fine-tuned based on the Transformer architecture (such as BERT, RoBERTa). The training corpus comes from real cross-cultural communication records and case studies, and experts have labeled the "appropriateness" levels of discourse in different cultural contexts (e.g., 1-5 points). After learning, the model can score the input text (the discourse of the evaluation object) in a specific conflict context (e.g., "facing direct instructions from a high power distance culture") based on aspects such as formality of word choice, politeness strategies, and expression of viewpoints, and output a semantic appropriateness score S_semantic (e.g., ranging from 0-1).

[0059] S323, semantic content appropriateness features are obtained based on semantic appropriateness scores.

[0060] This feature can be the score S_semantic directly, or its normalized value. The semantic content appropriateness feature measures the ability of the assessed individual to respond appropriately and effectively using language under cultural conflict pressure.

[0061] This application, through the design of steps S321-S323, converts audio data from multimodal behavioral data into text data and utilizes a natural language processing model trained on professional cross-cultural corpora to perform semantic appropriateness analysis. This enables an objective and intelligent assessment of the appropriateness and effectiveness of the language expression content of the evaluated object in a cross-cultural context. This method overcomes the subjectivity, inconsistency, and efficiency bottlenecks of traditional manual evaluation, and can accurately quantify whether the language content conforms to specific cultural norms and etiquette. The semantic appropriateness features extracted in this way, as the core dimension in the first feature data, not only deepen the assessment from superficial fluency and correctness to a more practically communicative cultural pragmatic level, but also contribute key language content quality indicators to the comprehensive evaluation index, significantly improving the scientific rigor, precision, and reliability of the entire evaluation system in measuring cross-cultural communicative competence.

[0062] Furthermore, in S33, the methods for extracting the second feature data include: S331 extracts speech fluency features, terminology accuracy features, and sentence complexity features from audio data in multimodal behavioral data before the injection of controllable cultural conflict elements.

[0063] Specifically, audio data from a stable time window (e.g., the first 30 seconds) preceding the conflict injection point T0 is analyzed. 1) Speech fluency features: quantified by calculating the number of words spoken per minute, average speech rate, frequency and average duration of silent pauses. 2) Terminology accuracy features: the proportion of correctly used terms among all term mentions is calculated by comparing the speech recognition text with a scene-related professional terminology dictionary. 3) Sentence complexity features: syntactic analysis of the text is performed to calculate indicators such as average sentence length and frequency of subordinate clauses.

[0064] S332 extracts behavioral coordination features from visual behavioral data in multimodal behavioral data before the injection of controllable cultural conflict elements.

[0065] Within the same time window, visual data is analyzed. Behavioral coordination features can be quantified by calculating the synchronicity of gestures and speech key points (such as the temporal correlation between gesture peaks and stress occurrences) and the stability of body posture (such as the amplitude and frequency of trunk movements).

[0066] S333, the second feature data is obtained based on speech fluency features, terminology accuracy features, sentence complexity features, and behavioral coordination features.

[0067] The second feature data is a vector composed of the above-mentioned basic features, representing the baseline level of English communicative competence of the assessed subject in the absence of specific cultural conflicts.

[0068] Through the design of steps S331-S333, this application effectively ensures the irrelevance between the extracted features and the preset cultural conflict stimuli by precisely defining the temporal boundaries of data collection. This provides a clean and reliable baseline reference for individual abilities for subsequent fusion analysis, and achieves objective quantification of the basic language proficiency of the assessment subjects, such as fluency, accuracy, complexity, and non-verbal behavior coordination.

[0069] S4, based on the fusion analysis of the first feature data and the second feature data, obtains a comprehensive evaluation index that characterizes the cross-cultural adaptability of the evaluated object.

[0070] This step S4 aims to integrate characteristic data reflecting different levels (conflict response and baseline capabilities) into a single, interpretable, comprehensive indicator.

[0071] Specifically, S4, based on the fusion analysis of the first feature data and the second feature data, yields a comprehensive assessment index characterizing the cross-cultural adaptability of the assessed object, including: S41, standardize the first feature data and the second feature data respectively.

[0072] Methods such as Z-score standardization or Min-Max normalization are used to eliminate the influence of different units and value ranges of each feature value. For example, the attention shift time difference Δt (in milliseconds, value range 100-2000) and semantic appropriateness score S_semantic (value range 0-1) in the first feature data are transformed to the same scale (such as the 0-1 interval).

[0073] S42, determine the weight coefficients corresponding to the first and second feature data after standardization.

[0074] The weighting coefficients (W1, W2) can be determined based on expert scoring or by learning through statistical methods (such as principal component analysis, PCA) based on historical evaluation data.

[0075] For example, if the assessment focuses more on the ability to cope with cultural conflicts, the primary characteristic data can be given a higher weight (e.g., W1=0.7, W2=0.3). The weighting coefficients can be adjusted for different assessment objectives.

[0076] S43, based on the standardized first and second feature data and their respective weight coefficients, calculates the comprehensive evaluation index.

[0077] The formula for calculating the Comprehensive Evaluation Index (CCI) is: CCI = W1 Σ(Standardized first feature_i) Sub-weights_i)+W2 Σ(Standardized second feature_j) Sub-weights (j). Here, Σ represents the weighted sum of sub-features within the same category, such as attention shift and semantic appropriateness. The final CCI is a scalar value, typically ranging from 0 to 100 or 1 to 10. A higher value indicates a stronger overall assessment of cross-cultural adaptability.

[0078] This application, through the design of steps S41-S43, effectively eliminates the comparability barriers caused by differences in dimensions and magnitudes between different characteristics, laying a mathematical foundation for the scientific integration of the two types of data. Furthermore, by configuring differentiated weighting coefficients for the two standardized types of data, a quantitative model that can flexibly adjust the assessment focus is constructed, scientifically regulating the contribution ratio of cultural conflict response ability and basic communication ability in the final evaluation. Finally, the comprehensive assessment index calculated based on standardized data and weighting coefficients not only ensures that the assessment results can more balancedly and accurately reflect the full picture of the cross-cultural adaptability of the assessed individuals, but also significantly enhances the customizability and practical value of this assessment method in different application scenarios.

[0079] In summary, the multi-dimensional cross-cultural English communicative competence assessment method provided in this application constructs an assessment environment that highly simulates real cross-cultural communicative situations by introducing conversational scenario data containing controllable cultural conflict elements. This effectively stimulates and captures the actual behavioral responses of the assessed individuals under cultural conflict. By collecting multimodal behavioral data generated during the conversation and further distinguishing and extracting primary feature data associated with controllable cultural conflict elements and secondary feature data unrelated to them, targeted isolation and identification of factors influencing cross-cultural adaptability are achieved. Based on this, by fusing and analyzing the two types of feature data, it is possible not only to comprehensively and objectively quantify the performance of the assessed individuals in multiple dimensions such as cultural conflict coping, language use, and non-verbal behavior, but also to significantly reduce assessment bias caused by individual general communication habits or scenario-irrelevant factors. Ultimately, a comprehensive assessment index characterizing their cross-cultural adaptability is obtained, and this method significantly improves the reliability and validity of the assessment.

[0080] Example 2 like Figure 2 As shown, a multi-dimensional cross-cultural English communicative competence assessment device includes: Presentation Module 1 is used to acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data. The conversation scene data includes controllable cultural conflict elements. Acquisition module 2 is used to collect multimodal behavioral data generated by the evaluation object during the session; Extraction module 3 is used to extract first feature data associated with controllable cultural conflict elements and second feature data unrelated to controllable cultural conflict elements based on multimodal behavioral data; Assessment module 4 is used to perform fusion analysis based on the first feature data and the second feature data to obtain a comprehensive assessment index characterizing the cross-cultural adaptability of the assessed object. The device of this application, through the cooperation of presentation module 1, acquisition module 2, extraction module 3, and assessment module 4, comprehensively and objectively quantifies the performance of the assessed object in multiple dimensions such as cultural conflict coping, language use, and non-verbal behavior, significantly reducing assessment bias and improving the reliability and validity of the assessment.

[0081] It should be understood that the various variations and specific embodiments of the multi-dimensional cross-cultural English communicative competence assessment method provided in Embodiment 1 above are also applicable to the multimodal non-performing asset risk assessment device in this embodiment. Through the detailed description of the aforementioned multi-dimensional cross-cultural English communicative competence assessment method, those skilled in the art can clearly understand the implementation process of the multimodal non-performing asset risk assessment device in this embodiment. For the sake of brevity, it will not be described in detail here.

[0082] Example 3 like Figure 3 As shown, an electronic device includes a processor 5 and a memory 6. When the processor 5 executes a computer program stored in the memory, it implements the multi-dimensional cross-cultural English communicative competence assessment method as described in Example 1. This method comprehensively and objectively quantifies the performance of the assessment subjects in multiple dimensions such as cultural conflict response, language use, and non-verbal behavior, significantly reducing assessment bias and improving the reliability and validity of the assessment.

[0083] Those skilled in the art should understand that the structure of the electronic device in this application does not constitute a limitation of the embodiments of this application. It can be a bus structure or a star structure. The electronic device may also include more or fewer other hardware or software than shown in the figure, or different component arrangements.

[0084] In some embodiments, the electronic device is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), digital processors, and embedded devices. The electronic device may also include user equipment, which includes, but is not limited to, any electronic product capable of human-computer interaction with a user via a keyboard, mouse, remote control, touchpad, or voice control device, such as VR glasses or VR headsets.

[0085] Example 4 A computer-readable storage medium storing a computer program is characterized in that, when executed by a processor, the computer program implements a multi-dimensional cross-cultural English communicative competence assessment method as described in Example 1, which comprehensively and objectively quantifies the performance of the assessment subjects in multiple dimensions such as cultural conflict response, language use, and non-verbal behavior, significantly reducing assessment bias and improving the reliability and validity of the assessment.

[0086] In the embodiments provided in this application, it should be understood that the disclosed methods, apparatuses, computer-readable storage media, and electronic devices can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple components or modules may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices, components, or modules may be electrical, mechanical, or other forms.

[0087] The components described as separate parts may or may not be physically separate. The components shown as components may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the components can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each component can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0089] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the intelligent data backup method for electronic devices described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0090] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0091] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0092] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A multi-dimensional method for assessing cross-cultural English communicative competence, characterized in that, The method includes: Acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data, wherein the conversation scene data includes controllable cultural conflict elements; Collect and evaluate multimodal behavioral data generated by the subject during the session; Based on the multimodal behavioral data, extract first feature data associated with the controllable cultural conflict elements and second feature data unrelated to the controllable cultural conflict elements; Based on the fusion analysis of the first feature data and the second feature data, a comprehensive evaluation index representing the cross-cultural adaptability of the evaluation object is obtained.

2. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 1, characterized in that, The acquisition of session scenario data includes: Acquire quantitative indicators for scenario type and target cultural dimension; Determine the initial session scenario corresponding to the scenario type; Based on the quantitative indicators of the target cultural dimension, the controllable cultural conflict elements for the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements are determined. Based on the controllable cultural conflict elements, the scene injection time, and the initial conversation scene, conversation scene data is generated.

3. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 2, characterized in that, The determination of the controllable cultural conflict elements for the initial conversation scenario and the scenario injection time of the controllable cultural conflict elements based on the quantitative indicators of the target cultural dimension includes: Based on the quantitative indicators of the target cultural dimension, one or more behavioral elements are retrieved from the preset cultural dimension-conflict behavior mapping library; A candidate element set is constituted based on the one or more behavioral elements; Based on the scenario type, target elements that match the initial conversation scenario are selected from the candidate element set as controllable cultural conflict elements. Identify key session nodes in the initial session scenario, and determine the scene injection time of the controllable cultural conflict element based on the key session nodes.

4. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 2 or 3, characterized in that, The first feature data includes attention shift features related to cultural conflict, and the methods for extracting the attention shift features include: The time of scene injection of the controllable cultural conflict elements is taken as the reference time point; Based on the eye-tracking data in the multimodal behavioral data, determine the time point at which the gaze focus of the assessment subject shifts to a new focus associated with the controllable cultural conflict element; The time difference between the reference time point and the transfer time point is calculated to obtain the attention transfer feature.

5. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 2 or 3, characterized in that, The first feature data includes semantic content appropriateness features, and the extraction methods for the semantic content appropriateness features include: Convert the audio data in the multimodal behavioral data into text data; The text data was analyzed using a natural language processing model trained on a cross-cultural corpus to obtain a semantic appropriateness score. The semantic appropriateness feature is obtained based on the semantic appropriateness score.

6. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 1, characterized in that, The methods for extracting the second feature data include: From the audio data in the multimodal behavioral data, extract speech fluency features, terminology accuracy features, and sentence complexity features before the injection of the controllable cultural conflict elements; From the visual behavior data in the multimodal behavior data, extract the behavioral coordination features before the injection of the controllable cultural conflict elements; The second feature data is obtained based on the speech fluency feature, the terminology accuracy feature, the sentence complexity feature, and the behavior coordination feature.

7. The multi-dimensional cross-cultural English communicative competence assessment method according to claim 1, characterized in that, The comprehensive assessment index representing the cross-cultural adaptability of the assessment object, obtained by fusing and analyzing the first feature data and the second feature data, includes: The first feature data and the second feature data are standardized respectively; Determine the weight coefficients corresponding to the first and second feature data after standardization. The comprehensive evaluation index is obtained by calculating the first and second feature data after standardization and their corresponding weight coefficients.

8. A multi-dimensional cross-cultural English communicative competence assessment device, characterized in that, The device includes: The presentation module is used to acquire conversation scene data and present the conversation scene to the evaluation object based on the conversation scene data, wherein the conversation scene data includes controllable cultural conflict elements; The data acquisition module is used to collect multimodal behavioral data generated by the evaluation object during the session; The extraction module is used to extract first feature data associated with the controllable cultural conflict elements and second feature data unrelated to the controllable cultural conflict elements based on the multimodal behavioral data. The assessment module is used to perform fusion analysis based on the first feature data and the second feature data to obtain a comprehensive assessment index that characterizes the cross-cultural adaptability of the assessment object.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory, the processor being used to implement the multidimensional cross-cultural English communicative competence assessment method according to any one of claims 1 to 7 when executing a computer program stored in the memory.

10. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the multidimensional cross-cultural English communicative competence assessment method according to any one of claims 1 to 7.

Images