Intelligent news broadcasting system and method

Abstract

The invention relates to the technical field of data analysis, in particular to an intelligent news broadcasting system and method.The method can comprise the steps that a generated news broadcasting text is obtained, and core keywords corresponding to the news broadcasting text are obtained; based on the vocabulary sensitivity degree mean value of the core key vocabularies and the manual generation participation degree of the news broadcast text, calculating to obtain a quantitative deletable degree index of the news broadcast text; based on the quantitative deletable degree index and the broadcast timeliness index of the news broadcast text, calculating to obtain a quantitative deferrable degree index of the news broadcast text; the visualization presents a quantized deletable degree indicator and / or a quantized delayable degree indicator. According to the technical scheme provided by the invention, the news events with higher priority and more core broadcast text content can be broadcasted emphatically by utilizing limited news broadcast time in an emergency, the broadcast sequence of the news broadcast text to be broadcasted / the final broadcast text content can be modified flexibly, and the method has a promotable value.

Description

Technical Field

[0001] This disclosure relates to the field of data analysis technology, specifically to a news intelligent broadcasting system and method. Background Technology

[0002] By allowing for greater flexibility in editing news broadcast texts, precise resource allocation can be achieved when urgent news needs to be inserted. Highly automated briefings can be quickly compressed or delayed to make room for breaking news; while important in-depth reports can have their length flexibly adjusted while retaining core facts. This differentiated processing mechanism significantly enhances the broadcasting system's emergency response capabilities, ensuring both the timely release of major news and maintaining the overall smoothness and integrity of the broadcast.

[0003] Traditional methods analyze the extent to which news articles can be edited based solely on their length, neglecting to consider the proportion of core content in different news texts and the sensitivity of the news. This may lead to inaccuracies in the edited text, resulting in misunderstandings among the public. Therefore, how to flexibly, accurately, and appropriately edit news broadcasts and adjust their broadcast order in emergency situations has become a pressing practical problem that needs to be solved. Summary of the Invention

[0004] To address the technical problems mentioned in the background section, the purpose of this disclosure is to provide a news intelligent broadcasting system and method, the specific technical solution of which is as follows: The first aspect of this disclosure provides a method for intelligent news broadcasting, which specifically may include: Obtain the generated news broadcast text and the corresponding keyword set; Based on preset filtering rules, core keywords with a high degree of relevance to the news broadcast text are selected from the keyword set. Based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text, a quantitative indicator of the degree of editability of news broadcast text is calculated. Based on the quantitative deletability index and the broadcast timeliness index of the news broadcast text, the quantitative delayability index of the news broadcast text is calculated. The indicators of quantifiable scalability and quantifiable deferability are presented visually.

[0005] In one possible implementation of the first aspect described above, this intelligent news broadcasting method may further include: Adjust the broadcast order and / or final broadcast content of news broadcast text based on quantitative indicators of the degree of deletion and / or quantitative indicators of the degree of delay.

[0006] In one possible implementation of the first aspect above, the process of selecting core keywords with a high degree of relevance to the news broadcast text from the keyword set based on preset filtering rules includes the following steps: From the keyword set, obtain the word frequency data and inverse document frequency data of each keyword in the news broadcast text, and use the product of the word frequency data and the inverse document frequency data as the important parameters of the word; In the keyword set, a text abstract graph corresponding to the news broadcast text is constructed with each keyword as a node. The weight value of each edge in the text abstract graph is positively correlated with the co-occurrence frequency of the keywords corresponding to the two ends of the edge. The semantic association parameters are determined based on the distribution of the weight values ​​of the connected edges of the keywords in the text abstract graph. The core degree parameter is calculated based on the lexical importance parameter and semantic association parameter, and the keywords with a core degree parameter greater than a preset threshold are designated as core keywords.

[0007] In one possible implementation of the first aspect above, during the process of calculating the core degree parameter based on the lexical importance parameter and the semantic association parameter, the core degree parameter is positively correlated with the lexical importance parameter and the core degree parameter is positively correlated with the semantic association parameter.

[0008] In one possible implementation of the first aspect mentioned above, in the process of calculating and obtaining a quantitative indicator of the editability of news broadcast text based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text, the acquisition of the degree of human involvement includes the following steps: Based on the news broadcast text generation process, obtain the first ratio of manually provided reference materials to all reference materials; Based on the generation process of news broadcast text, the number of non-human decision-making steps is obtained, and the ratio of the number of times core keywords appear in the news broadcast text to the number of non-human decision-making steps is used as the second ratio. The degree of artificial generation participation is calculated based on the first ratio and the second ratio. The value of the degree of artificial generation participation is positively correlated with the first ratio and the second ratio.

[0009] In one possible implementation of the first aspect mentioned above, the process of calculating a quantitative indicator of the editability of news broadcast text based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text includes the following steps: Obtain the first difference between the maximum word sensitivity and the average word sensitivity of all news broadcast texts to be broadcast, and use the ratio of the first difference to the degree of human participation as the third ratio. Obtain the fourth ratio of the total number of characters in the news broadcast text to the maximum number of characters in multiple news broadcast texts; The quantitative reduction index is calculated based on the third and fourth ratios. The value of the quantitative reduction index is positively correlated with the third ratio and the fourth ratio.

[0010] In one possible implementation of the first aspect above, in the process of calculating the quantitative delayability index of the news broadcast text based on the quantitative scalability index and the broadcast timeliness index of the news broadcast text, the acquisition of the broadcast timeliness index includes the following steps: Get the fifth ratio of the interval between the current time and the current time of the news broadcast text to the maximum interval of all news broadcast texts to be broadcast; Obtain the second difference between the distance between the news broadcast text and the current broadcast location and the minimum distance between all news broadcast texts to be broadcast; Obtain the growth rate of user interaction behavior in the news broadcast text, and use the ratio of the user interaction behavior growth rate to the second difference as the sixth ratio. The broadcast timeliness index is calculated based on the fifth ratio and the sixth ratio. The value of the broadcast timeliness index is positively correlated with the fifth ratio and the sixth ratio.

[0011] In one possible implementation of the first aspect above, the process of calculating the quantitative delayability index of the news broadcast text based on the quantitative scalability index and the broadcast timeliness index of the news broadcast text includes the following steps: Obtain the third difference between the quantitative cutoff index and the minimum quantitative cutoff index of all news broadcast texts to be broadcast; The quantitative delayability index is calculated based on the third difference and the broadcast timeliness index. The value of the quantitative delayability index is inversely correlated with the third difference and the value of the broadcast timeliness index.

[0012] A second aspect of this disclosure provides a news intelligent broadcasting system, which may specifically include: The news broadcast text generation module is used to obtain the generated news broadcast text and the keyword set corresponding to the news broadcast text. The core keyword filtering module is used to filter out core keywords that are highly relevant to the news broadcast text from the keyword set based on preset filtering rules. The module for assessing the degree of editability is used to calculate a quantitative indicator of the degree of editability of news broadcast text based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text. The delayability assessment module is used to calculate and obtain the quantitative delayability index of the news broadcast text based on the quantitative scalability index and the broadcast timeliness index of the news broadcast text. The visualization module is used to visualize the quantitative indicators of quantifiable scalability and quantifiable delayability.

[0013] In one possible implementation of the second aspect above, the intelligent news broadcasting system may further include: The news broadcast adjustment module is used to adjust the broadcast order and / or final broadcast content of news broadcast text based on quantitative indicators of the degree of deletion and / or quantitative indicators of the degree of delay.

[0014] Compared with the prior art, this disclosure has the following beneficial effects: The technical solution provided in this disclosure can identify, judge, and extract core keywords from news broadcast texts partially involving artificial intelligence. Based on the degree of human involvement in the generation process of the news broadcast text, and considering the proportion of core keywords, word sensitivity, and degree of human involvement, it quantifies the editability of the news broadcast text. Furthermore, based on the quantified editability of different news broadcast texts and the timeliness of the news events corresponding to the news broadcast texts, it quantifies the delayability of the news broadcast texts. Finally, it visualizes the quantified editability and delayability, enabling news professionals to accurately and intuitively understand the information. By assessing the editability and delayability of each news broadcast text, this method can prioritize and focus on higher-priority news events with more core content within limited broadcast time during emergency situations involving a concentration of news events. It allows for flexible modification of the broadcast order and final content of news broadcast texts, enabling a more rational arrangement of these elements in emergency situations. This effectively avoids negative consequences such as reduced accuracy, public misunderstanding, and delayed reporting of urgent news events caused by improper text editing, and has significant potential for wider application. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1This is a flowchart illustrating a news intelligent broadcasting method according to an embodiment of the present disclosure.

[0017] Figure 2 According to embodiments of this disclosure, a flowchart is provided to filter core keywords that are highly relevant to news broadcast text from a keyword set based on preset filtering rules.

[0018] Figure 3 According to an embodiment of this disclosure, a flowchart is provided to obtain the degree of human participation.

[0019] Figure 4 According to embodiments of this disclosure, a flowchart is provided to calculate a quantitative indicator of the editability of news broadcast text based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text.

[0020] Figure 5 According to an embodiment of this disclosure, a flowchart for obtaining broadcast timeliness indicators is provided.

[0021] Figure 6 According to embodiments of this disclosure, a flowchart is provided to calculate a quantitative delayability index for a news broadcast text based on a quantitative scalability index and a broadcast timeliness index for the news broadcast text.

[0022] Figure 7 According to an embodiment of this disclosure, a schematic diagram of the structure of a news intelligent broadcasting system is provided. Detailed Implementation

[0023] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, provides a detailed explanation of the specific implementation, structure, features, and effects of a news intelligent broadcasting method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0024] 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 invention pertains.

[0025] To address the problems raised in the background art, this disclosure provides a news intelligent broadcasting system and method that, in emergency situations, utilizes limited news broadcasting time to focus on broadcasting news events with higher priority and more core content, and flexibly modifies the broadcasting order / final content of the news broadcast text. Specifically, in some embodiments of this disclosure, Figure 1 A flowchart illustrating a method for intelligent news broadcasting is shown, such as... Figure 1 As shown, the specific steps may include the following: Step 100: Obtain the generated news broadcast text and the corresponding keyword set. In some embodiments, during the generation of the news broadcast text, multi-source information can be automatically discovered and monitored through intelligent crawling and classification models. Relevant news content can be identified in real time according to the requirements of the news. Then, multimodal parsing technology can be used to process various materials uploaded manually and identified by the system (including but not limited to deep learning models extracting webpage text, speech recognition technology converting audio and video, and OCR processing of image text). Furthermore, a pre-trained language model is used for structured information extraction, identifying key entities, clarifying the event context, and dividing the text hierarchy. Finally, text purification, intelligent summarization, and knowledge graphs are used to enhance content quality, optimize readability and information density, and achieve cross-source deduplication and event aggregation based on semantic similarity. A multi-dimensional quality evaluation system is established to ensure the output meets the standards, which is not limited here.

[0026] Step 200: Based on the preset filtering rules, select core keywords that are highly relevant to the news broadcast text from the keyword set.

[0027] Step 300: Based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of news broadcast text, calculate and obtain a quantitative indicator of the degree of editability of news broadcast text.

[0028] Step 400: Calculate the quantitative delayability index of the news broadcast text based on the quantitative deletion index and the broadcast timeliness index of the news broadcast text.

[0029] Step 500: Visualize the quantitative indicators of the degree of quantification that can be reduced and the quantitative indicators of the degree of quantification that can be delayed.

[0030] In some embodiments of this disclosure, such as Figure 1As shown, this intelligent news broadcasting method may further include an optional step 600: adjusting the broadcasting order and / or final broadcasting content of the news broadcast texts based on a quantitative scalability index and / or a quantitative delay index. It is understood that, based on visually obtaining the quantitative scalability index and / or the quantitative delay index, the broadcasting order of each news broadcast text can be adaptively adjusted according to the values ​​of the quantitative scalability index and / or the quantitative delay index. Simultaneously, when broadcasting time is limited, the broadcasting content of each news broadcast text with a quantitative scalability index value is adaptively modified to ensure complete broadcasting of all news events requiring broadcasting within a limited broadcasting time. The specific implementation of steps 100 to 600 above will be further explained below with reference to specific embodiments.

[0031] In some embodiments of this disclosure, specifically in the concrete implementation of the aforementioned step 200, Figure 2 This document illustrates a flowchart of a process for selecting core keywords highly relevant to news broadcast text from a keyword set based on preset filtering rules. Figure 2 As shown, the specific steps may include the following: Step 210: From the keyword set, obtain the word frequency data and inverse document frequency data for each keyword in the news broadcast text, and use the product of the word frequency data and the inverse document frequency data as the vocabulary importance parameter. In some embodiments, word frequency data refers to the frequency of a word's occurrence in the current document, and inverse document frequency data refers to the general importance of a word in the entire document set, which can be represented by the following mathematical expression: ; in, Keywords In all the news broadcast texts to be played Inverse document frequency data; For all news broadcast texts to be played The number of news broadcast texts in China For all news broadcast texts to be played Keywords appear in The number of news broadcast texts. Understandably, keywords... In all the news broadcast texts to be played Inverse document frequency data The higher the value, the more important the keywords are. In all the news broadcast texts to be played The higher the importance of it; This represents the logarithmic function with base 2.

[0032] Step 220: In the keyword set, construct a text abstract graph corresponding to the news broadcast text, using each keyword as a node, and determine semantic association parameters based on the distribution of the weight values ​​of the connected edges between the keywords in the text abstract graph. In some embodiments, the weight values ​​of each edge in the text abstract graph in step 220 are positively correlated with the co-occurrence frequency of the keywords corresponding to the two endpoints of the edge. In some embodiments, the news broadcast text can be... The keywords in the text form a node in the text abstract graph: a sliding window is defined to traverse the text, and an edge is established between co-occurring keyword pairs within the window; the weight of the edge is determined by the co-occurrence frequency of the keyword pair, the higher the frequency, the greater the edge weight, indicating a closer semantic relationship, thereby obtaining the news broadcast text. The text abstraction diagram is not limited here.

[0033] In some embodiments, steps 210 and 220 can be executed synchronously or asynchronously, and no limitation is made here; in some embodiments, when steps 210 and 220 are executed asynchronously, the execution order of steps 210 and 220 is not limited.

[0034] Step 230: Calculate the core importance parameter based on the lexical importance parameter and semantic association parameter, and designate keywords with a core importance parameter greater than a preset threshold as core keywords. In some embodiments, the core importance parameter can be obtained based on the following mathematical expression: ; in, Keywords News broadcast text Coreness parameter; For each keyword The corresponding lexical importance parameter, which is the keyword. The product of the corresponding word frequency data and inverse document frequency data is the total text of the news broadcast to be played. The maximum value of the vocabulary importance parameter corresponding to the keywords in the text; Keywords The sum of the weights of all edges connected to it in the text abstract graph. For all news broadcast texts to be played Each keyword node in the middle The corresponding weight sum The maximum value, Keywords The sum of the weights of all nodes connected by only one edge in the text abstract graph, where the addition of 0.01 is a set minimum value used to avoid keywords. The sum of the weights of all edges connected to it in the text abstract graph is equal to the total weight of all news broadcast text to be played. Each keyword node in the middle The corresponding weight sum The maximum values ​​are the same, leading to the anomaly of a denominator of zero. This is understandable. (Keywords) News broadcast text Coreness parameters The higher the value, the better the keyword. The more representative the news broadcast text is The actual content. In some embodiments, the core degree parameter can be determined using the minimum-maximum normalization method. Normalization is performed, and the normalized core intensity parameter can be located in the interval [0,1]. When the normalized core intensity parameter is greater than a preset threshold (e.g., 0.8, 0.9, 0.95, etc., not limited here), the corresponding keywords are... As core keywords, these keywords serve as an important reference during text deletion, ensuring the semantic accuracy of the news broadcast text during the editing process. Based on the aforementioned mathematical expression, it can be understood that in calculating the core degree parameter based on the lexical importance parameter and the semantic association parameter, the core degree parameter is positively correlated with the lexical importance parameter, and simultaneously, the core degree parameter is positively correlated with the semantic association parameter.

[0035] In some embodiments of this disclosure, specifically in the concrete implementation of the aforementioned step 300, Figure 3 A flowchart illustrating a process for obtaining the level of human involvement in the generated content is shown, such as... Figure 3 As shown, the specific steps may include the following: Step 310: Based on the news broadcast text generation process, obtain the first ratio of manually provided reference materials to all reference materials. In some embodiments, artificial intelligence can be used to assist in the generation of specific broadcast text content during the news broadcast text generation process. Considering that if the content provided by artificial intelligence is too involved in the news broadcast text generation process, it is easy to generate seemingly reasonable but actually completely wrong content in the generated news broadcast text, i.e., "illusions," which may include fabricating non-existent quotations, data, or event details, seriously damaging the authenticity of the news. To avoid this situation, the generated text should be carefully reviewed by humans. Therefore, the degree of human involvement can be used as an indicator to judge the ratio of manually provided materials to human autonomous decision-making in the news broadcast text generation process as one of the bases for whether the content of the news broadcast text can be deleted or modified subsequently: the higher the degree of human involvement, the lower the degree of deletion or modification of the news broadcast text content can be set.

[0036] Step 320: Based on the news broadcast text generation process, obtain the number of non-human decision-making steps, and use the ratio of the number of occurrences of core keywords in the news broadcast text to the number of non-human decision-making steps as a second ratio. In some embodiments, steps 310 and 320 can be executed synchronously or asynchronously, which is not limited here; in some embodiments, when steps 310 and 320 are executed asynchronously, the execution order of steps 310 and 320 is not limited.

[0037] Step 330: Calculate the degree of artificial generation participation based on the first ratio and the second ratio, wherein the value of the degree of artificial generation participation is positively correlated with the first ratio and positively correlated with the second ratio. In some embodiments, the degree of artificial generation participation can be represented by the following mathematical expression: ; in, For news broadcast text The degree of human involvement in the generation; For news broadcast text The percentage of reference materials provided by humans The percentage of manually provided reference materials in all news broadcast texts to be broadcast. The first ratio; Core keywords in news broadcast text The number of times it appears in For news broadcast text The total number of steps in the generation process. For news broadcast text The number of human decision-making steps in the generation process, ( (This is the text for a news broadcast) The number of non-human decision-making steps in the generation process, where the increase of 0.01 is a set minimum value, used to avoid news broadcast text The number of non-human decision-making steps in the generation process is zero, resulting in an anomaly where the denominator is zero. This is the second ratio. It's understandable that this applies to news broadcast texts. The degree of artificial participation The larger the value, the lower the degree of artificial intelligence involvement in the generation process of the news broadcast text, and the lower the quantitatively reducible degree of the news broadcast text that can be subsequently deleted or modified.

[0038] In some embodiments of this disclosure, specifically in the concrete implementation of the aforementioned step 300, Figure 4This document illustrates a flowchart illustrating a process for calculating a quantitative indicator of the editability of news broadcast text based on the average lexical sensitivity of core keywords and the degree of human involvement in the generation of the text. Figure 4 As shown, the specific steps may include the following: Step 410: Obtain the first difference between the maximum word sensitivity and the average word sensitivity of all news broadcast texts to be broadcast, and use the ratio of the first difference to the degree of human participation as the third ratio. In some embodiments, the average word sensitivity refers to the average word sensitivity of the core keywords in the current news broadcast text. The word sensitivity can be determined by pre-grading the sensitivity of words in different fields. The higher the word sensitivity value, the higher the user's sensitivity and attention to the core keyword. The smaller the first difference between the maximum word sensitivity and the average word sensitivity of all news broadcast texts to be broadcast, the more sensitive the core keywords in the current news broadcast text are to the user, and the less suitable it is to make large-scale deletions. In some embodiments, the ratio of the first difference to the degree of human involvement is used as the third ratio, which is also inversely correlated with the degree of human involvement. It can be understood that the higher the degree of human involvement, the more the generated content of the current news broadcast text has been manually screened and reviewed, and the less the AI-generated part is available for deletion and modification. That is, the value of the third ratio is positively correlated with the degree of deletion and modification of the current news broadcast text. The smaller the value of the third ratio, the less suitable the current news broadcast text is for large-scale content deletion and modification.

[0039] Step 420: Obtain the fourth ratio of the total number of characters in the news broadcast text to the maximum number of characters in multiple news broadcast texts. In some embodiments, steps 410 and 420 can be executed synchronously or asynchronously, which is not limited here; in some embodiments, when steps 410 and 420 are executed asynchronously, the execution order of steps 410 and 420 is not limited. In some embodiments, the larger the fourth ratio, the more characters the current news broadcast text has, indicating that it belongs to the range with a relatively large total number of characters among all news broadcast texts to be played, and has more room for deletion and modification compared to other news broadcast texts with a smaller total number of characters.

[0040] Step 430: Calculate a quantitative cutaway index based on the third and fourth ratios, wherein the value of the quantitative cutaway index is positively correlated with the third ratio and positively correlated with the fourth ratio. In some embodiments, the quantitative cutaway index can be represented by the following mathematical expression: ; in, For news broadcast text A quantitative indicator of the degree of quantifiability that can be reduced; For news broadcast text Total number of characters This represents the maximum number of characters in the entire news broadcast text to be broadcast. This is the fourth ratio; The maximum word sensitivity for all news broadcast texts to be broadcast. For news broadcast text The average vocabulary sensitivity, ( The first difference is represented by 0.01, where the increase of 0.01 is the set minimum value, used to avoid errors in news broadcast text. average vocabulary sensitivity This refers to the maximum lexical sensitivity of all news broadcast texts to be broadcast. This leads to an anomaly where other evaluation parameters cannot be reflected in the calculation of the quantitative index of deletability. For news broadcast text The degree of human involvement in the generation This is the third ratio. It's understandable that this applies to news broadcast texts. Quantitative deletability index The larger the value, the better the current news broadcast text. The higher the editability and acceptability of all news broadcast texts to be broadcast, the more priority should be given to quantifying the degree of editability in scenarios where content editing is required within a limited broadcast time. High-value news broadcast text Edit or delete the content. It should be noted that, to avoid a denominator of 0, when... When the value is 0, the minimum value of all non-zero artificially generated participation levels in this embodiment of the invention is used to replace it. The values ​​of the denominators at the corresponding positions are used for calculation.

[0041] In some embodiments of this disclosure, specifically in the concrete implementation of the aforementioned step 400, Figure 5 A flowchart illustrating a process for obtaining broadcast timeliness indicators is shown, such as... Figure 5 As shown, the specific steps may include the following: Step 510: Obtain the fifth ratio of the interval between the current time and the current time of the news broadcast text to the maximum interval of all news broadcast texts to be broadcast. In some embodiments, the interval between the current time and the current time of the news broadcast text refers to the time interval between the occurrence time of the news content or news event corresponding to the news broadcast text and the current time. The closer the interval is to the maximum interval, the longer the current news broadcast text has occurred among all the news broadcast texts to be broadcast. In keeping with the real-time nature of news broadcasting, this news broadcast text has a higher broadcast priority than other news content to be broadcast in the current allowed broadcast time period.

[0042] Step 520: Obtain the second difference between the distance between the news broadcast text and the current broadcast location and the minimum distance between all news broadcast texts to be broadcast. In some embodiments, steps 510 and 520 can be executed synchronously or asynchronously, which is not limited here; in some embodiments, when steps 510 and 520 are executed asynchronously, the execution order of steps 510 and 520 is not limited. It can be understood that the distance between the news broadcast text and the current broadcast location refers to the geographical distance between the location of the news content or news event corresponding to the news broadcast text and the current news broadcast location. The closer the distance is to the minimum distance, the closer the current news broadcast text is to the current news broadcast location among all news broadcast texts to be broadcast. When news broadcasts focus more on broadcasting news content in the current location, the broadcast priority is relatively higher.

[0043] Step 530: Obtain the user interaction behavior growth rate of the news broadcast text, and use the ratio of the user interaction behavior growth rate to the second difference as the sixth ratio. In some embodiments, the user interaction behavior growth rate of the news broadcast text refers to the increase in user interaction behaviors such as forwarding and commenting on the news content or news event corresponding to the news broadcast text on the relevant social media platform from the time the news content or news event occurred to the present moment. The higher the user interaction behavior growth rate of the news broadcast text, the higher the degree of user attention to the news content or news event corresponding to the current news broadcast text, and the higher the broadcast priority of the current news broadcast text among all the news broadcast texts to be broadcast.

[0044] Step 540: Calculate the broadcast timeliness index based on the fifth ratio and the sixth ratio, wherein the value of the broadcast timeliness index is positively correlated with the fifth ratio and the value of the broadcast timeliness index is positively correlated with the sixth ratio. In some embodiments, the broadcast timeliness index can be represented based on the following mathematical expression: ; in, For news broadcast text Corresponding news content The broadcast timeliness indicator, when the same news content Corresponding to multiple news broadcast texts At that time, these news broadcast texts The corresponding broadcast timeliness indicators can all be used The representation is not limited here; For news broadcast text Distance from current time The interval length, This represents the maximum interval between all news broadcasts to be broadcast. This is the fifth ratio; For news broadcast text The growth rate of user interaction behavior; For news broadcast text Distance from the current broadcast location The minimum interval between all news broadcast texts to be broadcast, ( This is used to characterize the second difference, where the increase of 0.01 is a set minimum value used to avoid news broadcast text. Interval distance This is the minimum interval between all the news broadcast texts to be broadcast. This leads to an abnormal calculation situation where the denominator is zero. This is understandable in news broadcast texts. Corresponding news content Broadcast timeliness indicators The larger the value, the better the current news broadcast text. Corresponding news content The higher the demand for real-time broadcasting, the more important it is to prioritize broadcasting within the limited broadcasting time.

[0045] In some embodiments of this disclosure, specifically in the concrete implementation of the aforementioned step 400, Figure 6 This diagram illustrates a process for calculating a quantitative delayability index for news broadcast text based on a quantitative scalability index for editability and a broadcast timeliness index for the news broadcast text. Figure 6 As shown, the specific steps may include the following: Step 610: Obtain the third difference between the quantified editability index and the minimum quantified editability index of all news broadcast texts to be broadcast. It can be understood that the smaller the difference between the quantified editability index and the minimum quantified editability index of all news broadcast texts to be broadcast, the less editable the current news broadcast text is, and the higher its importance relative to all other news broadcast texts to be broadcast.

[0046] Step 620: Calculate the quantified delayability index based on the third difference and the broadcast timeliness index, wherein the value of the quantified delayability index is inversely correlated with the third difference and inversely correlated with the broadcast timeliness index. In some embodiments, the quantified delayability index can be represented by the following mathematical expression: ; in, For news broadcast text Corresponding news content A quantitative indicator of the degree of delay, when the same news content... Corresponding to multiple news broadcast texts At that time, these news broadcast texts The corresponding quantitative indicators of delayability can all be used The representation is not limited here; For news broadcast text A quantitative indicator of the degree of quantifiability that can be reduced; The minimum quantifiable editability index for all news broadcast texts to be broadcast; This is used to characterize the third difference, where the increase of 0.01 is a set minimum value used to avoid errors in news broadcast text. Quantitative deletability index This is the minimum quantifiable reduction index. This can lead to an abnormal calculation situation where the denominator is zero; For news broadcast text Corresponding news content The broadcast timeliness indicator. Understandably, this quantifies the degree of delay. The difference from the third value ( ) and broadcast timeliness indicators Inverse correlation, a quantitative indicator of the degree of delay. The larger the value, the higher the deferability of the current news broadcast text relative to all other news broadcast texts to be broadcast, and the smaller the impact of delaying its broadcast in the broadcast order. It should be noted that, to avoid a denominator of 0, when... When the value is 0, the minimum value of all non-zero broadcast timeliness indicators in this embodiment of the invention is used to replace the value. The values ​​of the denominators at the corresponding positions are used for calculation.

[0047] In some embodiments, the minimum-maximum normalization method can be used to quantify the degree of delay. Normalization is performed, and the normalized quantized delayability index can be located in the range [0,1]. When the normalized quantized delayability index is greater than the preset delayability threshold (e.g., 0.8, 0.9, 0.95, etc., which are not limited here), if it is still impossible to complete the broadcast of all news texts within the limited news broadcast time after reasonable deletion of multiple news broadcast texts, it may be considered to revoke the broadcast of that news content. The corresponding news broadcast text will be postponed, without any restrictions.

[0048] In some embodiments of this disclosure, Figure 7 A schematic diagram of the structure of a news intelligent broadcasting system is shown, such as... Figure 7 As shown, this intelligent news broadcasting system may specifically include a news broadcasting text generation module 710, a core keyword filtering module 720, a deletion assessment module 730, a delay assessment module 740, and a visualization presentation module 750.

[0049] In some embodiments, the news broadcast text generation module 710 can be used to obtain the generated news broadcast text and the keyword set corresponding to the news broadcast text. In some embodiments, the core keyword filtering module 720 can be used to filter out core keywords with a high degree of relevance to the news broadcast text from the keyword set based on preset filtering rules. In some embodiments, the deletability assessment module 730 can be used to calculate and obtain a quantitative deletability index for the news broadcast text based on the average lexical sensitivity of the core keywords and the degree of human participation in the generation of the news broadcast text. In some embodiments, the delayability assessment module 740 can be used to calculate and obtain a quantitative delayability index for the news broadcast text based on the quantitative deletability index and the broadcast timeliness index of the news broadcast text. In some embodiments, the visualization module 750 can be used to visualize and present the quantitative deletability index and the quantitative delayability index. In some embodiments, the functional implementation of the above-mentioned news broadcast text generation module 710 to visualization module 750 can refer to steps 100 to 500 in the foregoing embodiments, and the appendix. Figures 1 to 6 The relevant explanations will not be repeated here.

[0050] In some embodiments of this disclosure, further, such as Figure 7 The intelligent news broadcasting system shown may also include a news broadcasting adjustment module 760, which adjusts the broadcasting order and / or final broadcasting content of the news broadcasting text based on quantitative indicators of the degree of deletion and / or quantitative indicators of the degree of delay.

[0051] In summary, the technical solution provided in this disclosure can identify, judge, and extract core keywords from news broadcast texts partially involving artificial intelligence. Based on the degree of human involvement in the generation process of the news broadcast text, and considering the proportion of core keywords, word sensitivity, and degree of human involvement, it quantifies the editability of the news broadcast text. Furthermore, based on the quantified editability of different news broadcast texts and the timeliness of the corresponding news events, it quantifies the delayability of the news broadcast text. Finally, it visualizes the quantified editability and delayability, enabling news professionals to accurately... This system allows for a direct assessment of the editability and delayability of each news broadcast text. In emergency situations involving a concentration of news events, it enables the use of limited broadcast time to prioritize and focus on news events with higher priority and more core content. It allows for flexible modification of the broadcast order and final content of news broadcast texts, ensuring a more rational arrangement in emergency situations. This effectively avoids negative consequences such as reduced accuracy, public misunderstanding, and delayed reporting of urgent news events due to improper text editing, making it a system with significant potential for wider adoption.

[0052] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0053] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A news intelligent broadcasting method, characterized by, The method comprises: obtaining a generated news report text, and obtaining a keyword set corresponding to the news report text; based on a preset screening rule, screening a core keyword with a high degree of association with the news report text from the keyword set; based on the average value of the keyword sensitivity of the core keyword and the artificial generation participation degree of the news report text, calculating to obtain a quantitative reducible degree index of the news report text; based on the quantitative reducible degree index and the reporting timeliness index of the news report text, calculating to obtain a quantitative delayable degree index of the news report text; visually presenting the quantitative reducible degree index and the quantitative delayable degree index.

2. The news intelligent broadcasting method of claim 1, wherein, The method further comprises: based on the quantitative reducible degree index and / or the quantitative delayable degree index, adjusting the reporting order and / or the final reporting content of the news report text.

3. The news intelligent broadcasting method according to claim 1 or 2, characterized in that, In the process of screening a core keyword with a high degree of association with the news report text from the keyword set based on a preset screening rule, the following steps are included: In the keyword set, obtain the term frequency data and the inverse document frequency data of each keyword in the news report text, and take the product of the term frequency data and the inverse document frequency data as a keyword importance parameter; In the keyword set, take each keyword as a node to construct a text abstract graph corresponding to the news report text, the weight value of each edge in the text abstract graph is positively correlated with the common occurrence frequency of the keywords corresponding to the two end nodes of the edge, and a semantic association parameter is determined according to the weight value distribution of the connected edges of the keyword in the text abstract graph; based on the keyword importance parameter and the semantic association parameter, calculate a core degree parameter, and take the keyword with the core degree parameter greater than a preset threshold as the core keyword.

4. The news intelligent broadcasting method of claim 3, wherein, In the process of calculating the core degree parameter based on the keyword importance parameter and the semantic association parameter, the core degree parameter is positively correlated with the keyword importance parameter, and the core degree parameter is positively correlated with the semantic association parameter.

5. The news intelligent broadcasting method according to claim 1 or 2, characterized in that, In the process of calculating the quantitative reducible degree index of the news report text based on the average value of the keyword sensitivity of the core keyword and the artificial generation participation degree of the news report text, the acquisition of the artificial generation participation degree comprises the following steps: based on the generation process of the news report text, obtaining a first ratio of human-provided reference materials to all reference materials; based on the generation process of the news report text, obtaining the number of non-human decision steps, and taking the ratio of the number of occurrences of the core keyword in the news report text to the number of non-human decision steps as a second ratio; based on the first ratio and the second ratio, calculate the artificial generation participation degree, the value of the artificial generation participation degree is positively correlated with the first ratio, and the value of the artificial generation participation degree is positively correlated with the second ratio.

6. The news intelligent broadcasting method of claim 5, wherein, In the process of calculating the quantitative reducible degree index of the news broadcast text based on the average of the vocabulary sensitivity degree of the core keywords and the artificial generation participation degree of the news broadcast text, the following steps are included: A first difference value between the maximum vocabulary sensitivity degree of all the news broadcast texts to be broadcast and the average of the vocabulary sensitivity degree is obtained, and a third ratio value is obtained by taking the ratio of the first difference value and the artificial generation participation degree; A fourth ratio value between the total number of characters of the news broadcast text and the maximum number of characters of the plurality of news broadcast texts is obtained; The quantitative reducible degree index is calculated based on the third ratio value and the fourth ratio value, the value of the quantitative reducible degree index is positively correlated with the third ratio value, and the value of the quantitative reducible degree index is positively correlated with the fourth ratio value.

7. The news intelligent broadcasting method according to claim 1 or 2, characterized by, In the process of calculating the quantitative delayable degree index of the news broadcast text based on the quantitative reducible degree index and the broadcast timeliness index of the news broadcast text, the acquisition of the broadcast timeliness index includes the following steps: A fifth ratio value between the interval time length of the news broadcast text from the current time and the maximum interval time length of all the news broadcast texts to be broadcast is obtained; A second difference value between the interval distance of the news broadcast text from the current broadcast location and the minimum interval distance of all the news broadcast texts to be broadcast is obtained; A user interaction behavior growth rate of the news broadcast text is obtained, and a sixth ratio value is obtained by taking the ratio of the user interaction behavior growth rate and the second difference value; The broadcast timeliness index is calculated based on the fifth ratio value and the sixth ratio value, the value of the broadcast timeliness index is positively correlated with the fifth ratio value, and the value of the broadcast timeliness index is positively correlated with the sixth ratio value.

8. The news intelligent broadcasting method of claim 7, wherein, In the process of calculating the quantitative delayable degree index of the news broadcast text based on the quantitative reducible degree index and the broadcast timeliness index of the news broadcast text, the following steps are included: A third difference value between the quantitative reducible degree index and the minimum quantitative reducible degree index of all the news broadcast texts to be broadcast is obtained; The quantitative delayable degree index is calculated based on the third difference value and the broadcast timeliness index, the value of the quantitative delayable degree index is inversely correlated with the third difference value, and the value of the quantitative delayable degree index is inversely correlated with the value of the broadcast timeliness index.

9. A news intelligent broadcasting system, characterized by comprising: The system includes: A news broadcast text generation module is configured to obtain a generated news broadcast text and obtain a keyword set corresponding to the news broadcast text; A core keyword screening module is configured to screen out core keywords with high association degree with the news broadcast text from the keyword set based on a preset screening rule; A reducible degree evaluation module is configured to calculate a quantitative reducible degree index of the news broadcast text based on the average of the vocabulary sensitivity degree of the core keywords and the artificial generation participation degree of the news broadcast text. a delayable degree evaluation module configured to calculate a quantified delayable degree indicator of the news report text based on the quantified cuttable degree indicator and a time-sensitive indicator of the news report text; a visual presentation module configured to visually present the quantified cuttable degree indicator and the quantified delayable degree indicator.

10. The news intelligent broadcasting system according to claim 9, wherein, The system further comprises: a news report adjustment module configured to adjust a report sequence and / or a final report content of the news report text based on the quantified cuttable degree indicator and / or the quantified delayable degree indicator.

Images