Artificial intelligence-based radio station event analysis method and system

Abstract

The application discloses a radio station event analysis method and system based on artificial intelligence, and the method comprises the following steps: data acquisition, feature extraction, radio station event evaluation, radio station event classification processing and result feedback optimization. The application belongs to the technical field of radio station event analysis, and specifically relates to a radio station event analysis method and system based on artificial intelligence. In the scheme, important features are selected to construct a decision tree structure, features and branches are determined by calculating information gain, the uncertainty of overall radio station event data is evaluated by calculating cumulative residual entropy, the complexity of data is understood, the overall features of radio station events are grasped, the relationship between features and evaluation results is evaluated by calculating conditional cumulative residual entropy, and the evaluation accuracy is improved. The classification consistency is ensured by repositioning, the role of radio station events is referred to, a uniqueness index is introduced, the comprehensive evaluation value of an individual is calculated to determine the processing priority of radio station events and resource allocation.

Description

Technical Field

[0001] This invention belongs to the field of radio event analysis technology, specifically a radio event analysis method and system based on artificial intelligence. Background Technology

[0002] Radio event analysis is a method that uses signal monitoring and speech recognition technologies to acquire content broadcast by radio stations. Natural language processing technology is then used to analyze the text of the content, extracting key information, viewpoints, and sentiments. The aim is to understand public opinion on specific events and topics through radio channels, and further combine big data analysis to gain insights into the development trends and related factors of public opinion, so as to better respond and make decisions.

[0003] However, existing radio station event analysis methods suffer from several technical problems. First, the radio event data is complex and the overall characteristics are unclear, leading to inaccurate evaluation of radio events, affecting the operation and development of radio stations, and causing poor listener experience. Second, different radio events are classified inconsistently, making it difficult to reasonably arrange the order of events among numerous events, resulting in unreasonable resource allocation. Third, the analysis of the uniqueness of radio events is incomplete and prone to subjective judgment bias. Summary of the Invention

[0004] To address the aforementioned issues and overcome the shortcomings of existing technologies, this invention provides an artificial intelligence-based radio station event analysis method and system. Specifically, it addresses the technical problem of complex radio event data with unclear overall characteristics, leading to inaccurate event assessments, impacting radio station operation and development, and resulting in a poor listener experience. The method employs a radio event assessment approach, which involves data preprocessing, selecting key features to construct a decision tree structure, calculating information gain to determine features and branches, calculating cumulative residual entropy to assess the uncertainty of the overall radio event data, helping to understand the complexity of the data, grasp the overall characteristics of radio events, and calculating conditional cumulative residual entropy to assess the relationship between features and assessment results, thereby improving assessment accuracy. Furthermore, it addresses the technical problem of inconsistent classification of different radio events, difficulty in rationally prioritizing numerous events, resulting in unreasonable resource allocation, incomplete analysis of the uniqueness of radio events, and susceptibility to subjective judgment bias. This approach employs repositioning to ensure classification consistency, and, referring to the role of radio events and introducing uniqueness indicators, calculates the comprehensive assessment value of individual events to determine the priority of radio event processing and resource allocation.

[0005] The technical solution adopted by this invention is as follows: The radio station event analysis method based on artificial intelligence provided by this invention includes the following steps:

[0006] Step S1: Data acquisition, specifically, acquiring the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset;

[0007] Step S2: Feature extraction, specifically, extracting audio features and time features to analyze the patterns in radio event data;

[0008] Step S3: Radio event assessment, specifically involves data preprocessing, selecting important features to construct a decision tree structure, determining features and branches by calculating information gain, calculating cumulative residual entropy to assess the uncertainty of the overall radio event data, helping to understand the complexity of the data, grasp the overall characteristics of radio events, and calculating conditional cumulative residual entropy to assess the relationship between features and assessment results, thereby improving assessment accuracy.

[0009] Step S4: Radio event classification and processing, specifically, repositioning to ensure classification consistency, and referring to the role of radio events, introducing uniqueness indicators, calculating the comprehensive evaluation value of individuals to determine the priority of radio event processing and resource allocation;

[0010] Step S5: Result feedback optimization, specifically, rationally allocating human and material resources according to the event type based on the radio event classification results.

[0011] Further, in step S1, the data acquisition includes the following steps:

[0012] Step S11: Use a dedicated radio signal receiving device to receive radio signals in real time, record the received radio audio, and obtain raw data.

[0013] Step S12: Use timestamp information to segment and organize the received signal for subsequent analysis of events in different time periods;

[0014] Step S13: Collect the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset.

[0015] Furthermore, in step S2, the feature extraction specifically involves extracting audio features from the audio data, extracting the time interval and frequency of events within the time interval from the timestamp information, performing semantic analysis in conjunction with the audio content, and extracting event theme and keyword features for further understanding and analysis of the patterns in the radio event data.

[0016] The audio features include intonation, timbre, audio intensity variations, and audio duration.

[0017] Further, in step S3, the radio event assessment includes the following steps:

[0018] Step S31: Data preprocessing. The collected radio event information is preprocessed and feature selected. The features that are important for evaluating radio events are used as nodes of the decision tree to generate a decision tree structure. The branches represent different feature conditions and judgment paths.

[0019] The key features for evaluating radio events include the event topic, speaker identity, key words in the speech, emotional keywords, the level of dissemination, and the timing of the event.

[0020] Step S32: Construct a decision tree model, select the feature with the largest information gain as the root node of the decision tree, subdivide the data for the value of the feature to form child nodes, calculate the new information gain at each child node and select a feature to continue building branches until the predetermined tree depth is reached. The formula used is as follows:

[0021] ;

[0022] In the formula, Indicates information gain. This represents the event characteristic information of the i-th radio station. It is the cumulative residual entropy of the radio event evaluation results in the training dataset, where Y represents a continuous random variable related to the radio events. It is the conditional cumulative residual entropy;

[0023] Step S33: Train the decision tree by adjusting its parameters using training data with known radio event results to better classify and predict radio events.

[0024] Step S34: Evaluate the performance of the decision tree through the validation set, optimize and adjust it to ensure that the decision tree can accurately evaluate and judge new radio event data input, including judging the popularity and attention of new events;

[0025] Step S35: Calculate the cumulative residual entropy to assess the uncertainty of the overall radio event data. The formula used is as follows:

[0026] ;

[0027] In the formula, i represents the index of the feature information, n represents the total number of feature information, and y i The value y represents the result of the radio event evaluation. i-1 Indicates with y i The value of the evaluation result of the adjacent previous radio event, This represents the probability that a continuous random variable is greater than or equal to the evaluation result of a radio event.

[0028] Step S36: Calculate the conditional cumulative residual entropy to more accurately assess the relationship between features and radio event evaluation results. The formula used is as follows:

[0029] ;

[0030] In the formula, Indicates in features The conditional cumulative residual entropy is given by the condition, where J represents the number of different class features and j represents the index of the different class features. This indicates the size of the entire radio event evaluation results dataset. Indicates in features This is a subset of the radio event evaluation results corresponding to the j-th category.

[0031] Further, in step S4, the radio event classification processing includes the following steps:

[0032] Step S41: Repositioning, used to ensure consistency in the classification of different radio events and avoid confusion and contradictions, using the following formula:

[0033] ;

[0034] In the formula, This represents a classification location vector for radio event standards, where the location is a relative position determined based on different signal characteristics, frequency ranges, and event types. This represents the actual classification location vector of the radio event, where m represents the total number of target dimensions used for radio event classification during the relocation process. This represents the category location vector of the radio event after relocation, where k represents the index for traversing each target dimension;

[0035] Step S42: Random diffusion, using the following formula:

[0036] ;

[0037] In the formula, x represents an individual radio event, SDE represents the random diffusion mechanism, and SF represents the location modification operation. This refers to the first reference radio station event. This refers to the second reference radio station incident. This represents the (N-1)th reference radio station event. The distance metric between an individual radio event and its first reference radio event, where N represents the total number of radio events;

[0038] The reference radio event serves as a comparison and reference during the diffusion process. By comparing and measuring with the reference radio event, the positional relationship and degree of change of individual radio events in random diffusion can be determined.

[0039] Step S43: Introduce a uniqueness index to quickly measure the relative position and uniqueness of an individual radio event within the entire group of radio events. This helps to quickly locate and distinguish radio events with special characteristics. The formula used is as follows:

[0040] ;

[0041] In the formula, A1 represents the uniqueness index, and X represents the set of radio events;

[0042] Step S44: Comprehensive evaluation. Calculate the comprehensive evaluation value for each radio event to determine which events to prioritize and how resources to allocate. The formula used is as follows:

[0043] ;

[0044] In the formula, This represents the overall assessment value of an individual radio event. This indicates the severity of the radio incident. This indicates the minimum severity of the radio incident. This indicates the maximum severity level of the radio incident;

[0045] The severity level indicates the degree to which a radio incident has caused widespread controversy, negative social repercussions, and misled public perception. Radio incidents with a high degree of severity require more urgent and priority handling to reduce negative impacts and harm.

[0046] Furthermore, in step S5, the result feedback optimization specifically involves rationally allocating human and material resources based on the radio event classification results and the event type, improving the early warning system for types of events that are prone to occur, and enhancing the ability of relevant personnel to handle events.

[0047] The radio station event analysis system based on artificial intelligence provided by this invention includes a data acquisition module, a feature extraction module, a radio station event evaluation module, a radio station event classification and processing module, and a result feedback and optimization module.

[0048] The data acquisition module specifically acquires the audio data and corresponding timestamp information data of the radio station signal to obtain a radio station event dataset;

[0049] The feature extraction module specifically extracts audio features and time features to analyze the patterns in radio event data;

[0050] The radio event assessment module specifically performs data preprocessing, selects important features to construct a decision tree structure, determines features and branches by calculating information gain, calculates cumulative residual entropy to assess the uncertainty of the overall radio event data, helps to understand the complexity of the data, grasp the overall characteristics of the radio events, calculates conditional cumulative residual entropy to assess the relationship between features and assessment results, and improves assessment accuracy.

[0051] The radio event classification and processing module specifically repositions events to ensure classification consistency, and, with reference to the role of radio events and the introduction of uniqueness indicators, calculates the comprehensive evaluation value of each individual event to determine the priority of radio event processing and resource allocation.

[0052] The result feedback optimization module specifically allocates human and material resources reasonably according to the event type based on the radio event classification results.

[0053] The beneficial results achieved by the present invention using the above solution are as follows:

[0054] (1) To address the technical problem that radio event data is complex and its overall characteristics are unclear, leading to inaccurate radio event assessment, affecting the operation and development direction of radio stations, and causing poor listener experience, radio event assessment is adopted. Specifically, data preprocessing is performed, important features are selected to construct a decision tree structure, features and branches are determined by calculating information gain, and cumulative residual entropy is calculated to assess the uncertainty of the overall radio event data, which helps to understand the complexity of the data, grasp the overall characteristics of radio events, and calculate the conditional cumulative residual entropy to assess the relationship between features and assessment results, thereby improving the accuracy of assessment.

[0055] (2) In view of the technical problems that there are inconsistencies in the classification of different radio events, it is difficult to arrange the order of events reasonably, resulting in unreasonable resource allocation, incomplete analysis of the uniqueness of radio events, and easy to produce subjective judgment bias, the repositioning is adopted to ensure the consistency of classification, and the role of radio events is referred to, the uniqueness index is introduced, and the comprehensive evaluation value of individuals is calculated to determine the priority of radio event processing and resource allocation. Attached Figure Description

[0056] Figure 1 A flowchart illustrating the AI-based radio station event analysis method provided by this invention;

[0057] Figure 2 A schematic diagram of the artificial intelligence-based radio station event analysis system provided by the present invention;

[0058] Figure 3 This is a flowchart illustrating step S3;

[0059] Figure 4 This is a flowchart illustrating step S4.

[0060] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0061] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0062] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0063] Example 1, see Figure 1 The present invention provides an artificial intelligence-based radio station event analysis method, which includes the following steps:

[0064] Step S1: Data acquisition, specifically, acquiring the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset;

[0065] Step S2: Feature extraction, specifically, extracting audio features and time features to analyze the patterns in radio event data;

[0066] Step S3: Radio event assessment, specifically involves data preprocessing, selecting important features to construct a decision tree structure, determining features and branches by calculating information gain, calculating cumulative residual entropy to assess the uncertainty of the overall radio event data, helping to understand the complexity of the data, grasp the overall characteristics of radio events, and calculating conditional cumulative residual entropy to assess the relationship between features and assessment results, thereby improving assessment accuracy.

[0067] Step S4: Radio event classification and processing, specifically, repositioning to ensure classification consistency, and referring to the role of radio events, introducing uniqueness indicators, calculating the comprehensive evaluation value of individuals to determine the priority of radio event processing and resource allocation;

[0068] Step S5: Result feedback optimization, specifically, rationally allocating human and material resources according to the event type based on the radio event classification results.

[0069] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the data acquisition includes the following steps:

[0070] Step S11: Use a dedicated radio signal receiving device to receive radio signals in real time, record the received radio audio, and obtain raw data.

[0071] Step S12: Use timestamp information to segment and organize the received signal for subsequent analysis of events in different time periods;

[0072] Step S13: Collect the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset.

[0073] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, the feature extraction specifically involves extracting audio features from audio data, extracting the time interval of events and the frequency of events within the time interval from timestamp information, performing semantic analysis in conjunction with audio content, and extracting event theme and keyword features for further understanding and analysis of the patterns in radio event data.

[0074] The audio features include intonation, timbre, audio intensity variations, and audio duration.

[0075] Example 4, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S3, the radio event assessment includes the following steps:

[0076] Step S31: Data preprocessing. The collected radio event information is preprocessed and feature selected. The features that are important for evaluating radio events are used as nodes of the decision tree to generate a decision tree structure. The branches represent different feature conditions and judgment paths.

[0077] The key features for evaluating radio events include the event topic, speaker identity, key words in the speech, emotional keywords, the level of dissemination, and the timing of the event.

[0078] Step S32: Construct a decision tree model, select the feature with the largest information gain as the root node of the decision tree, subdivide the data for the value of the feature to form child nodes, calculate the new information gain at each child node and select a feature to continue building branches until the predetermined tree depth is reached. The formula used is as follows:

[0079] ;

[0080] In the formula, Indicates information gain. This represents the event characteristic information of the i-th radio station. It is the cumulative residual entropy of the radio event evaluation results in the training dataset, where Y represents a continuous random variable related to the radio events. It is the conditional cumulative residual entropy;

[0081] Step S33: Train the decision tree by adjusting its parameters using training data with known radio event results to better classify and predict radio events.

[0082] Step S34: Evaluate the performance of the decision tree through the validation set, optimize and adjust it to ensure that the decision tree can accurately evaluate and judge new radio event data input, including judging the popularity and attention of new events;

[0083] Step S35: Calculate the cumulative residual entropy to assess the uncertainty of the overall radio event data. The formula used is as follows:

[0084] ;

[0085] In the formula, i represents the index of the feature information, n represents the total number of feature information, and y i The value y represents the result of the radio event evaluation. i-1 Indicates with y i The value of the evaluation result of the adjacent previous radio event, This represents the probability that a continuous random variable is greater than or equal to the evaluation result of a radio event.

[0086] Step S36: Calculate the conditional cumulative residual entropy to more accurately assess the relationship between features and radio event evaluation results. The formula used is as follows:

[0087] ;

[0088] In the formula, Indicates in features The conditional cumulative residual entropy is given by the condition, where J represents the number of different class features and j represents the index of the different class features. This indicates the size of the entire radio event evaluation results dataset. Indicates in features This is a subset of the radio event evaluation results corresponding to the j-th category.

[0089] By performing the above operations, radio event assessment is adopted. Specifically, data preprocessing is carried out, important features are selected to construct a decision tree structure, features and branches are determined by calculating information gain, and cumulative residual entropy is calculated to assess the uncertainty of the overall radio event data. This helps to understand the complexity of the data, grasp the overall characteristics of radio events, and calculate the conditional cumulative residual entropy to assess the relationship between features and assessment results, thereby improving the accuracy of assessment. This solves the technical problem that the complexity of radio event data and the ambiguity of overall characteristics lead to inaccurate radio event assessment, affecting the operation and development direction of radio stations and causing poor listener experience.

[0090] Example 5, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S4, the radio event classification processing includes the following steps:

[0091] Step S41: Repositioning, used to ensure consistency in the classification of different radio events and avoid confusion and contradictions, using the following formula:

[0092] ;

[0093] In the formula, This represents a classification location vector for radio event standards, where the location is a relative position determined based on different signal characteristics, frequency ranges, and event types. This represents the actual classification location vector of the radio event, where m represents the total number of target dimensions used for radio event classification during the relocation process. This represents the category location vector of the radio event after relocation, where k represents the index for traversing each target dimension;

[0094] Step S42: Random diffusion, using the following formula:

[0095] ;

[0096] In the formula, x represents an individual radio event, SDE represents the random diffusion mechanism, and SF represents the location modification operation. This refers to the first reference radio station event. This refers to the second reference radio station incident. This represents the (N-1)th reference radio station event. The distance metric between an individual radio event and its first reference radio event, where N represents the total number of radio events;

[0097] The reference radio event serves as a comparison and reference during the diffusion process. By comparing and measuring with the reference radio event, the positional relationship and degree of change of individual radio events in random diffusion can be determined.

[0098] Step S43: Introduce a uniqueness index to quickly measure the relative position and uniqueness of an individual radio event within the entire group of radio events. This helps to quickly locate and distinguish radio events with special characteristics. The formula used is as follows:

[0099] ;

[0100] In the formula, A1 represents the uniqueness index, and X represents the set of radio events;

[0101] Step S44: Comprehensive evaluation. Calculate the comprehensive evaluation value for each radio event to determine which events to prioritize and how resources to allocate. The formula used is as follows:

[0102] ;

[0103] In the formula, This represents the overall assessment value of an individual radio event. This indicates the severity of the radio incident. This indicates the minimum severity of the radio incident. This indicates the maximum severity level of the radio incident;

[0104] The severity level indicates the degree to which a radio incident has caused widespread controversy, negative social repercussions, and misled public perception. Radio incidents with a high degree of severity require more urgent and priority handling to reduce negative impacts and harm.

[0105] By performing the above operations, repositioning is adopted to ensure classification consistency. By referring to the role of radio events and introducing uniqueness indicators, the comprehensive evaluation value of individuals is calculated to determine the priority of radio event processing and resource allocation. This solves the technical problems of inconsistent classification of different radio events, difficulty in reasonably arranging the order of events when faced with a large number of events, unreasonable resource allocation, incomplete analysis of the uniqueness of radio events, and easy subjective judgment bias.

[0106] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S5, the result feedback optimization specifically involves rationally allocating human and material resources according to the radio event classification results and event types, improving the early warning system for types of events that are prone to occur, and enhancing the ability of relevant personnel to handle events.

[0107] Example 7, see Figure 2 Based on the above embodiments, the artificial intelligence-based radio station event analysis system provided by the present invention includes a data acquisition module, a feature extraction module, a radio station event evaluation module, a radio station event classification and processing module, and a result feedback optimization module.

[0108] The data acquisition module specifically acquires the audio data and corresponding timestamp information data of the radio station signal to obtain a radio station event dataset;

[0109] The feature extraction module specifically extracts audio features and time features to analyze the patterns in radio event data;

[0110] The radio event assessment module specifically performs data preprocessing, selects important features to construct a decision tree structure, determines features and branches by calculating information gain, calculates cumulative residual entropy to assess the uncertainty of the overall radio event data, helps to understand the complexity of the data, grasp the overall characteristics of the radio events, calculates conditional cumulative residual entropy to assess the relationship between features and assessment results, and improves assessment accuracy.

[0111] The radio event classification and processing module specifically repositions events to ensure classification consistency, and, with reference to the role of radio events and the introduction of uniqueness indicators, calculates the comprehensive evaluation value of each individual event to determine the priority of radio event processing and resource allocation.

[0112] The result feedback optimization module specifically allocates human and material resources reasonably according to the event type based on the radio event classification results.

[0113] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0114] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0115] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A radio station event analysis method based on artificial intelligence, characterized in that: The method includes the following steps: Step S1: Data acquisition, specifically, acquiring the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset; Step S2: Feature extraction, specifically, extracting audio features and time features to analyze the patterns in radio event data; Step S3: Radio event assessment, specifically involves data preprocessing, selecting important features to construct a decision tree structure, determining features and branches by calculating information gain, calculating cumulative residual entropy to assess the uncertainty of the overall radio event data, grasping the overall characteristics of radio events, and calculating conditional cumulative residual entropy to assess the relationship between features and assessment results. Step S4: Radio event classification and processing, specifically, repositioning to ensure classification consistency, and referring to the role of radio events, introducing uniqueness indicators, calculating the comprehensive evaluation value of individuals to determine the priority of radio event processing and resource allocation; Step S5: Result feedback optimization, specifically, rationally allocating human and material resources according to the event type based on the radio event classification results.

2. The radio station event analysis method based on artificial intelligence according to claim 1, characterized in that: In step S3, the radio event assessment includes the following steps: Step S31: Data preprocessing. The collected radio event information is preprocessed and feature selected. The features that are important for evaluating radio events are used as nodes of the decision tree to generate a decision tree structure. The branches represent different feature conditions and judgment paths. The key features for evaluating radio events include the event topic, speaker identity, key words in the speech, emotional keywords, the level of dissemination, and the timing of the event. Step S32: Construct a decision tree model, select the feature with the largest information gain as the root node of the decision tree, subdivide the data for the value of the feature to form child nodes, calculate the new information gain at each child node and select a feature to continue building branches until the predetermined tree depth is reached. The formula used is as follows: ； In the formula, Indicates information gain. This represents the event characteristic information of the i-th radio station. It is the cumulative residual entropy of the radio event evaluation results in the training dataset, where Y represents a continuous random variable related to the radio events. It is the conditional cumulative residual entropy; Step S33: Train the decision tree by adjusting its parameters using training data with known radio event results to better classify and predict radio events. Step S34: Evaluate the performance of the decision tree using the validation set, and optimize and adjust it to ensure that the decision tree can accurately evaluate and judge new radio event data inputs, including judging the degree of attention given to new events; Step S35: Calculate the cumulative residual entropy to assess the uncertainty of the overall radio event data. The formula used is as follows: ； In the formula, i represents the index of the feature information, n represents the total number of feature information, and y i The value y represents the result of the radio event assessment. i-1 Indicates with y i The value of the evaluation result of the adjacent previous radio event, This represents the probability that a continuous random variable is greater than or equal to the evaluation result of a radio event. Step S36: Calculate the conditional cumulative residual entropy to more accurately assess the relationship between features and radio event evaluation results. The formula used is as follows: ； In the formula, Indicates in features The conditional cumulative residual entropy is given by the condition, where J represents the number of different class features and j represents the index of the different class features. This indicates the size of the entire radio event evaluation results dataset. Indicates in features This is a subset of the radio event evaluation results corresponding to the j-th category.

3. The radio station event analysis method based on artificial intelligence according to claim 1, characterized in that: In step S4, the radio event classification and processing includes the following steps: Step S41: Repositioning, used to ensure consistency in the classification of different radio events and avoid confusion and contradictions, using the following formula: ； In the formula, This represents a classification location vector for radio event standards, where the location is a relative position determined based on different signal characteristics, frequency ranges, and event types. This represents the actual classification location vector of the radio event, where m represents the total number of target dimensions used for radio event classification during the relocation process. This represents the category location vector of the radio event after relocation, where k represents the index for traversing each target dimension; Step S42: Random diffusion, using the following formula: ； In the formula, x represents an individual radio event, SDE represents the random diffusion mechanism, and SF represents the location modification operation. This refers to the first reference radio station event. This refers to the second reference radio station incident. This represents the (N-1)th reference radio station event. The distance metric between an individual radio event and its first reference radio event, where N represents the total number of radio events; The reference radio event serves as a comparison and reference during the diffusion process. By comparing and measuring with the reference radio event, the positional relationship and degree of change of individual radio events in random diffusion can be determined. Step S43: Introduce a uniqueness index to quickly measure the relative position and uniqueness of an individual radio event within the entire group of radio events. This helps to quickly locate and distinguish radio events with special characteristics. The formula used is as follows: ； In the formula, A1 represents the uniqueness index, and X represents the set of radio events; Step S44: Comprehensive evaluation. Calculate the comprehensive evaluation value for each radio event using the following formula: ； In the formula, This represents the overall assessment value of an individual radio event. This indicates the severity of the radio incident. This indicates the minimum severity of the radio incident. This indicates the maximum severity level of the radio incident; The severity level indicates the extent to which the radio incident has caused widespread controversy, negative social repercussions, and misled public perception.

4. The radio station event analysis method based on artificial intelligence according to claim 1, characterized in that: In step S1, the data acquisition includes the following steps: Step S11: Use a dedicated radio signal receiving device to receive radio signals in real time, record the received radio audio, and obtain raw data. Step S12: Use timestamp information to segment and organize the received signal for subsequent analysis of events in different time periods; Step S13: Collect the audio data and corresponding timestamp information of the radio station signal to obtain the radio station event dataset.

5. The radio station event analysis method based on artificial intelligence according to claim 1, characterized in that: In step S2, the feature extraction specifically involves extracting audio features from audio data, extracting the time interval and frequency of events within the time interval from timestamp information, performing semantic analysis in conjunction with audio content, and extracting event theme and keyword features for further understanding and analysis of the patterns in radio event data. The audio features include intonation, timbre, audio intensity variations, and audio duration.

6. The radio station event analysis method based on artificial intelligence according to claim 1, characterized in that: In step S5, the result feedback optimization specifically involves allocating human and material resources reasonably based on the radio event classification results and the event type, improving the early warning system for common event types, and enhancing the ability of relevant personnel to handle events.

7. An artificial intelligence-based radio station event analysis system, used to implement the artificial intelligence-based radio station event analysis method as described in any one of claims 1-6, characterized in that: It includes a data acquisition module, a feature extraction module, a radio event evaluation module, a radio event classification and processing module, and a result feedback and optimization module.

8. The artificial intelligence-based radio station event analysis system according to claim 7, characterized in that: The data acquisition module specifically acquires the audio data and corresponding timestamp information data of the radio station signal to obtain a radio station event dataset; The feature extraction module specifically extracts audio features and time features to analyze the patterns in radio event data; The radio event assessment module specifically involves data preprocessing, selecting important features to construct a decision tree structure, determining features and branches by calculating information gain, calculating cumulative residual entropy to assess the uncertainty of the overall radio event data, grasping the overall characteristics of the radio events, and calculating conditional cumulative residual entropy to assess the relationship between features and assessment results. The radio event classification and processing module specifically repositions events to ensure classification consistency, and, with reference to the role of radio events and the introduction of uniqueness indicators, calculates the comprehensive evaluation value of each individual event to determine the priority of radio event processing and resource allocation. The result feedback optimization module specifically allocates human and material resources reasonably according to the event type based on the radio event classification results.

Images