Cultural transmission integrating degree evaluation system

Abstract

The invention relates to a cultural transmission integrating degree evaluation system, which comprises a multi-modal data acquisition module for acquiring text, image, audio and video transmission contents, and a cultural feature knowledge base for storing cultural dimension parameters and taboo rules of a target area. According to the cultural transmission integrating degree evaluation system, through a cultural attention mechanism and cross-media association analysis, the cultural symbol recognition accuracy is improved, synchronous analysis of 87 languages and visual symbols is supported, the problem of misjudgment caused by single-modal analysis is solved, cultural transition is dynamically adapted, and the evaluation accuracy is improved. A weight adjustment mechanism is adopted to enable the system to shorten the updating speed of an evaluation standard when a culture hotspot event occurs, a culture entropy model can quantitatively monitor the diffusion / attenuation trend of culture elements, and through the three-dimensional quantitative evaluation system, the cognitive layer culture element recognition coverage is high, and the behavior layer: the propagation conversion rate prediction error is low; the culture conflict detection response time is short, the efficiency is greatly improved compared with manual auditing, and the automatic correction suggestion adoption rate is high.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cultural communication, in particular to a cultural communication fit degree evaluation system. BACKGROUND

[0002] Cultural communication, also known as cultural diffusion, refers to the process of human culture radiating outward from the cultural source or spreading from one social group to another. It can be divided into direct communication and indirect communication. The former usually involves the direct transmission of cultural content in the form of spirit or material by people with culture through trade teams, military forces, etc., while the latter represents a more complex cultural diffusion force, mainly referring to the stimulation of cultural creation activities by a social group using principles from foreign cultural characteristics.

[0003] The process of cultural diffusion depends on various factors such as the practical value, difficulty level, prestige of civilization, adaptability to the times, and resistance of culture, so the evaluation of its fit degree is particularly important. For example, Chinese patent 201810023222.3 discloses a cultural communication and promotion system that realizes double communication between cultural communication enterprises and users and efficient communication of culture.

[0004] However, traditional cultural communication systems have the following three major defects despite having fit degree evaluation:

[0005] 1. The fit degree is usually evaluated using questionnaires, which have poor timeliness and limited sample size.

[0006] 2. The evaluation system based on single text analysis cannot handle multimedia content.

[0007] 3. Current cultural dimension theory application systems lack dynamic adaptability.

[0008] Therefore, there is an urgent need for a cultural content communication effect quantitative evaluation system that combines deep learning and semantic networks to solve the above problems. SUMMARY

[0009] To overcome the deficiencies of the prior art, the present application provides a cultural communication fit degree evaluation system that has the advantages of multi-modal fusion to improve evaluation accuracy, dynamic adaptation to cultural changes, and improved evaluation efficiency, and solves the problems of poor timeliness and limited sample size of questionnaire survey method, the inability of evaluation systems based on single text analysis to handle multimedia content, and the lack of dynamic adaptability of cultural dimension theory application systems.

[0010] To achieve the above purpose, the present application provides the following technical solution: a cultural communication fit degree evaluation system, comprising:

[0011] A multi-modal data acquisition module: acquires text, image, audio, and video communication content.

[0012] Cultural feature knowledge base: store cultural dimension parameters and taboo rules of target region;

[0013] Dynamic evaluation engine: calculate the matching degree of propagation content and cultural standards based on deep learning;

[0014] Visual feedback interface: generate three-dimensional evaluation report and optimization suggestions.

[0015] Further, the multi-modal data acquisition module comprises:

[0016] Cultural symbol detection unit: identify visual elements such as religious totems and national costumes;

[0017] Semantic sensitive word mining unit: build a dynamically updated taboo corpus;

[0018] Cross-media association analysis unit: establish semantic mapping relationship between text, picture and video.

[0019] Further, the cultural symbol detection unit adopts an improved YOLOv5 network structure, wherein a cultural attention layer (CulturalAttentionLayer) is added to enhance the extraction of specific cultural features, and a cultural weight factor ω = 1 + |S_c-S_t| is added to the loss function, S_c is the cultural saliency score.

[0020] Further, the cultural feature knowledge base comprises:

[0021] Static cultural dimension data: based on Hofstede's six cultural dimension theory;

[0022] Dynamic public opinion data: real-time capture of cultural sensitivity changes on social media;

[0023] User-defined rule engine: allows setting specific evaluation thresholds.

[0024] Further, the dynamic evaluation engine comprises the following steps:

[0025] S1, decompose the input content into N cultural feature vectors;

[0026] S2, calculate the Euclidean distance of each vector and the target culture

[0027] S3, get the comprehensive fit score F = 1-∑W_id_i through fuzzy logic algorithm.

[0028] Further, the determination of the weight w_i is made by analyzing historical propagation effect data based on an LSTM network time series prediction model, combined with a reinforcement learning mechanism based on user feedback, and the weight is automatically updated every 100 evaluations;

[0029] The three-dimensional evaluation adopts a three-dimensional evaluation model, comprising:

[0030] Cognitive layer: cultural element recognition accuracy;

[0031] Emotional layer: audience emotional tendency matching degree;

[0032] Behavior layer: expected propagation conversion rate prediction.

[0033] Further, it further comprises:

[0034] Culture conflict warning module: triggering multi-level alarm when high-risk content is detected;

[0035] Automatic correction suggestion generator; provide editable solutions for alternative text / images.

[0036] Further, it further comprises a cultural propagation entropy value calculation method applied to the above-mentioned system, comprising:

[0037] S1, establish a cultural element probability distribution model P(x);

[0038] S2, calculate the entropy value H = -∑P(x)log_bP(x), wherein the base b is set according to the cultural circle layer difference:

[0039] S3, take b = 5 for high-context cultural circle

[0040] S3, take b = 2 for low-context cultural circle.

[0041] Further, when the system is implemented in hardware, FPGA is used to accelerate the cultural feature extraction process, edge computing nodes are deployed to realize real-time evaluation, and a cultural data security isolation area is set.

[0042] Compared with the prior art, the technical scheme of the present application has the following beneficial effects:

[0043] 1. The cultural propagation fit degree evaluation system improves the cultural symbol recognition accuracy through cultural attention mechanism and cross-media association analysis, supports synchronous analysis of 87 languages and visual symbols, and solves the misjudgment problem caused by single modal analysis.

[0044] 2. The cultural propagation fit degree evaluation system dynamically adapts to cultural changes, uses a weight adjustment mechanism to shorten the evaluation standard update speed when cultural hot events occur, and the cultural entropy model can quantitatively monitor the diffusion / decay trend of cultural elements.

[0045] 3. The cultural propagation fit degree evaluation system, through a three-dimensional quantitative evaluation system, the cognitive layer cultural element recognition coverage is high, the behavior layer: the propagation conversion rate prediction error is low, the culture conflict detection response time is fast, the efficiency is greatly improved compared with manual review, and the automatic correction suggestion adoption rate is high. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 The block diagram of the cultural communication compatibility evaluation system. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0048] Please refer to Figure 1 The cultural communication compatibility evaluation system in the embodiment includes:

[0049] The multi-modal data acquisition module acquires text, image, audio and video communication content, including:

[0050] The cultural symbol detection unit identifies visual elements such as religious totems and national costumes, and adopts an improved YOLOv5 network structure, in which a cultural attention layer (Cultural Attention Layer) is newly added to enhance the extraction of specific cultural features, and a cultural weight factor ω = 1 + |S_c-S_t| is added to the loss function, S_c being a cultural saliency score.

[0051] The cultural feature knowledge base includes:

[0052] The static cultural dimension data is constructed based on Hofstede's six-dimensional culture theory;

[0053] The dynamic public opinion data is real-time captured from social media for cultural sensitivity changes;

[0054] The user-defined rule engine allows to set specific evaluation thresholds;

[0055] The semantic sensitive word mining unit constructs a dynamically updated taboo corpus;

[0056] The cross-media association analysis unit establishes semantic mapping relationships among texts, images and videos.

[0057] The cultural feature knowledge base stores cultural dimension parameters and taboo rules of the target area;

[0058] The dynamic evaluation engine calculates the matching degree of the communication content and the cultural standard based on deep learning, and the evaluation method includes the following steps:

[0059] S1, decompose the input content into N cultural feature vectors;

[0060] S2, calculate the Euclidean distance of each vector and the target culture

[0061] S3, get the comprehensive fit degree score F = 1-∑W_id_i through the fuzzy logic algorithm.

[0062] The determination of the weight w_i is made by analyzing historical propagation effect data based on an LSTM network time series prediction model, combined with a reinforcement learning mechanism based on user feedback, and the weight is automatically updated every 100 evaluations;

[0063] Visual feedback interface: generate a three-dimensional evaluation report and optimization suggestions, use a three-dimensional evaluation model, including:

[0064] Cognitive layer: cultural element recognition accuracy;

[0065] Emotional layer: audience emotional tendency matching degree;

[0066] Behavior layer: expected propagation conversion rate prediction.

[0067] A culture propagation fit degree evaluation system also includes:

[0068] Culture conflict warning module: trigger multi-level alarms when high-risk content is detected;

[0069] Automatic correction suggestion generator; provide editable alternatives for text / images.

[0070] A culture propagation fit degree evaluation system also includes a culture propagation entropy value calculation method, which is applied to the above system, including:

[0071] S1, establish a cultural element probability distribution model P(x);

[0072] S2, calculate the entropy value H = -∑P(x)log_bP(x), where the base b is set according to the cultural circle difference:

[0073] S3, take b = 5 for high-context culture circle

[0074] S3, take b = 2 for low-context culture circle.

[0075] Specifically, when the system is implemented in hardware, FPGA is used to accelerate the cultural feature extraction process, edge computing nodes are deployed to realize real-time evaluation, and a cultural data security isolation area is set.

[0076] By adopting the above scheme, the cultural symbol recognition accuracy is improved through cultural attention mechanism and cross-media association analysis, supporting synchronous analysis of 87 languages and visual symbols, and solving the misjudgment problem caused by single modal analysis. Dynamically adapting to cultural changes, the weight adjustment mechanism is adopted to shorten the evaluation standard update speed when cultural hot events occur, and the cultural entropy model can quantitatively monitor the diffusion / decay trend of cultural elements. Through the three-dimensional quantitative evaluation system, the cultural element recognition coverage of the cognitive layer is high, the behavior layer has low prediction error of transmission conversion rate, the cultural conflict detection response time is fast, and the efficiency is greatly improved compared with manual review, and the automatic correction suggestion adoption rate is high.

[0077] It should be noted that the cultural transmission fit degree refers to the matching degree of the transmission content and the cultural values, cognitive habits, social norms and other elements of the target audience when information is transmitted in different cultural backgrounds. The higher the fit degree, the better the transmission effect, otherwise it may lead to misunderstanding, resistance or failure. This concept is particularly important in cross-cultural communication, international marketing, brand globalization, and film and television work export.

[0078] Example 1: Cross-border brand advertising cultural adaptation evaluation

[0079] 1. Application scenario

[0080] A certain international beverage brand conducts a compliance review before launching Ramadan-themed advertisements in Southeast Asia.

[0081] 2. System workflow

[0082] 2.1. Data input

[0083] Upload the advertisement video (including a picture of a Muslim family having dinner), select the target area: Malaysia, Indonesia, and load the Islamic culture feature library (including 58 taboo rules).

[0084] 2.2. Multi-modal analysis

[0085] Visual layer: detects a left-hand cup-holding picture of tableware (Islamic culture taboo), identifies that the headscarf coverage of the female character is less than 90% (does not meet local standards).

[0086] 2.3. Audio layer: background music contains pig squeal samples (trigger sensitive word library), Arabic dubbing accent deviation degree reaches 0.7 (exceeds threshold 0.5)

[0087] 2.4. Dynamic evaluation

[0088] Calculate the cultural conflict index: 72 / 100 (high risk)

[0089] Generate a three-dimensional evaluation report:

[0090] [Cognitive level] Identified 4 types of Islamic cultural elements with an accuracy rate of 98%;

[0091] [Emotional Layer] Predicts a 63% probability of negative emotions;

[0092] [Behavioral layer] The expected conversion rate loss is 41%;

[0093] 2.5 Automatic Optimization

[0094] Suggested modifications to the output: Replace the left-handed character with a right-handed character (provide a 3D model replacement solution), change the headdress material to a semi-transparent gauze (to conform to local aesthetics and not violate religious doctrine), and recommend using the Quran-licensed nasheed music as a replacement.

[0095] Implementation results: The revised advertisement improved cultural fit to 91 / 100, sales in the Malaysian market increased by 37% year-on-year, and the potential loss of $2.5 million in brand value due to religious disputes was avoided.

[0096] Example 2: Digital Dissemination of Intangible Cultural Heritage

[0097] 1. Application Scenarios

[0098] Optimization of International Dissemination Strategies for Chinese Kunqu Opera on YouTube

[0099] 2. System Implementation Process

[0100] 2.1 Cultural DNA Modeling

[0101] Extracting the core elements of Kunqu Opera:

[0102] Visual characteristics: length of the water sleeves (standard 2.4 meters), facial makeup color scheme;

[0103] Auditory characteristics: flute pitch fluctuation range (±15 cents);

[0104] Semantic features: 100 classic song lyrics and allusions.

[0105] 2.2 Cross-cultural matching

[0106] Establish an audience culture mapping matrix:

[0107] Cultural circle Best entry point Risk warning Western Europe Opera lovers (0.82 similarity) Avoid direct translation of "Zhongxiao" North America Modern dance audience (0.76 similarity) Simplify drum piece

[0108] 3. Intelligent Adaptation

[0109] Generate differentiated versions:

[0110] Western version: Compressed transition music (from 3 minutes to 1.5 minutes);

[0111] Japanese and Korean versions: Enhanced close-up shots of clothing textures (30% more camera angles);

[0112] Middle East version: Remove all flute solos (replace with Oud timbre).

[0113] Effect verification

[0114] Comparison of propagation data:

[0115] Indicator Traditional version System optimization version Upgrade Completion rate 12％ 38％ 217％ Sharing rate 1.2％ 5.7％ 375％ Cultural misunderstanding comments 23 2 -91％

[0116] Example 3: International social media crisis warning

[0117] 1. Application scenario

[0118] Monitor the potential risks of a mobile phone brand's promotional poster released in the Indian market

[0119] 2. Real-time processing process

[0120] 2.1 Risk factor capture

[0121] Identify the cow pattern (Hindu sacred object) in the poster in the same frame with the leather phone case;

[0122] Detect the Hindi translation of the slogan "Break the Shackles" containing the semantics of liberation.

[0123] 2.2 Cultural entropy calculation

[0124] Establish a communication prediction model:

[0125] Current entropy H = 0.87 (exceeding the safety threshold of 0.65);

[0126] Probability of protest within 6 hours: 78%.

[0127] 3. Graded response

[0128] Implement a three-level emergency plan: immediately remove the original poster (system automatically contact the publishing platform API), generate alternative solutions: replace the cow pattern with a peacock (the national bird of India), and start KOL apology video automatic generation (including namaste gesture recognition).

[0129] 4. Loss control effect: from risk identification to crisis resolution, it takes 47 minutes (traditional team needs 8 hours), avoids the #Boycott brand topic from being hot searched (saves crisis public relations cost $1.8 million), and the subsequent user emotion recovery speed is increased by 3 times.

[0130] In summary, the cultural dissemination fitting degree evaluation system improves the cultural symbol recognition accuracy rate through the cultural attention mechanism and cross-media association analysis, supports the synchronous analysis of 87 languages and visual symbols, and solves the misjudgment problem caused by single modal analysis. The system dynamically adapts to cultural changes, and the weight adjustment mechanism shortens the evaluation standard update speed when cultural hot events occur. The cultural entropy model can quantitatively monitor the diffusion / decay trend of cultural elements. Through the three-dimensional quantitative evaluation system, the cognitive layer cultural element recognition coverage reaches 98.5%, the emotional layer audience emotional tendency prediction F1-score is 0.91, the behavior layer transmission conversion rate prediction error is less than 5%, the cultural conflict detection response time is less than 0.2 seconds, the efficiency is improved by 340 times compared with manual review, the automatic correction suggestion adoption rate is as high as 76%, and the system has the advantages of multi-modal fusion to improve evaluation accuracy, dynamic adaptation to cultural changes, and improvement of evaluation efficiency, and solves the problems of poor timeliness and limited sample size of questionnaire survey method, the evaluation system based on single text analysis cannot process multimedia content, and the cultural dimension theory application system lacks dynamic adaptability.

[0131] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0132] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A cultural communication fit assessment system, characterized in that, include: Multimodal data acquisition module: Acquires text, images, audio, and video content. Cultural Feature Knowledge Base: Stores cultural dimension parameters and taboo rules for the target region; Dynamic evaluation engine: Based on deep learning, it calculates the matching degree between disseminated content and cultural standards; Visual feedback interface: Generates 3D evaluation reports and optimization suggestions.

2. The cultural communication compatibility assessment system according to claim 1, characterized in that, The multimodal data acquisition module includes: Cultural symbol detection unit: identifies visual elements such as religious totems and ethnic costumes; Semantic sensitive word mining unit: Constructing a dynamically updated taboo corpus; Cross-media association analysis unit: Establish semantic mapping relationships between text, images, and audio.

3. The cultural communication compatibility assessment system according to claim 2, characterized in that, The cultural symbol detection unit adopts an improved YOLOv5 network structure, in which a new Cultural Attention Layer is added to enhance the extraction of specific cultural features. The loss function incorporates a cultural weight factor ω = 1 + |S_c - S_t|, where S_c is the cultural saliency score.

4. The cultural communication compatibility assessment system according to claim 1, characterized in that, The cultural feature knowledge base includes: Static cultural dimension data: constructed based on Hofstede's six-dimensional cultural theory; Dynamic public opinion data: Real-time capture of changes in cultural sensitivity on social media; User-defined rules engine: Allows setting specific evaluation thresholds.

5. The cultural communication fit assessment system according to claim 1, characterized in that, The dynamic evaluation engine's evaluation method includes the following steps: S1. Decompose the input content into N cultural feature vectors; S2. Calculate the Euclidean distance between each vector and the target culture. S3. Obtain the comprehensive fit score F = 1 - ∑W_id_i through fuzzy logic algorithm.

6. The cultural communication compatibility assessment system according to claim 5, characterized in that, The weight w_i is determined by analyzing historical propagation effect data through a time series prediction model based on an LSTM network, combined with a reinforcement learning mechanism based on user feedback, and the weight is automatically updated every 100 evaluations. The three-dimensional evaluation adopts a three-dimensional evaluation model, including: Cognitive level: Accuracy of cultural element recognition; Emotional layer: Matching degree of audience's emotional inclination; Behavioral layer: Predicting the expected conversion rate of the message.

7. The cultural communication fit assessment system according to claim 1, characterized in that, Also includes: Cultural conflict early warning module: Triggers multi-level alerts when high-risk content is detected; Automatic correction suggestion generator; Provides alternative editable solutions for text / images.

8. The cultural communication compatibility assessment system according to claim 1, characterized in that, It also includes a method for calculating the entropy of cultural transmission, applied to the system described in claim 1, comprising: S1. Establish a probability distribution model P(x) for cultural elements; S2. Calculate the entropy value H = -∑P(x)log_bP(x), where the base b is set according to the differences in cultural spheres: S3, High-context cultural sphere b=5 S3, low-context cultural sphere b=2.

9. A cultural communication fit assessment system according to claim 1, characterized in that, The system is implemented in hardware using FPGA to accelerate the cultural feature extraction process, deploying edge computing nodes to achieve real-time evaluation, and setting up a cultural data security isolation zone.

Images