A method and system for intelligent real-time quality inspection of live broadcasts in the financial industry

Abstract

The present invention provides an intelligent real-time quality inspection method and system for live broadcasting in the financial industry, which belongs to the technical field of live broadcasting data, and specifically comprises: determining the violation risk coefficients of different violation types based on the analysis data of historical violation behaviors of different violation types, determining the audience data and interaction data of the financial service products during the historical live broadcasting process based on the historical live broadcasting data of the financial service products, and determining the matching quality inspection strategy of the financial service products in combination with the violation risk coefficients of the different violation types, performing quality inspection processing during the live broadcasting process of the financial service products by using the matching quality inspection strategy, and determining whether it is necessary to adjust the quality inspection strategy by using the analysis results of the real-time audience data and real-time interaction data during the live broadcasting process, thereby improving the reliability of the quality inspection processing of the violations.

Description

Technical Field

[0001] The present invention belongs to the technical field of live broadcast data, and in particular, relates to an intelligent real-time quality inspection method and system for live broadcast in the financial industry. Background Art

[0002] In order to improve the conversion rate and user reach of financial service products, more and more financial service companies are using live streaming to sell financial service products. However, at the same time, how to ensure compliance during the live streaming process has become a technical problem that needs to be solved urgently.

[0003] In order to realize quality inspection during live broadcast, the abnormal data contained in the live broadcast data stream is monitored by the quality inspection and audit module in the invention patent application CN202210682572.7 "Live broadcast quality inspection system, method, Web server and storage medium", and the abnormal data is displayed by the data display module, thereby realizing efficient multi-dimensional quality inspection and real-time notification of abnormalities. However, there are the following technical problems:

[0004] When conducting quality inspections, when financial service products are recommended and introduced, it is inevitable that violations such as "products guarantee profits" will exist. Therefore, if quality inspections of violations cannot be carried out and timely intervention and processing cannot be carried out, compliance cannot be guaranteed during the live broadcast process.

[0005] In response to the above technical problems, the present invention provides an intelligent real-time quality inspection method and system for live broadcasts in the financial industry. Summary of the invention

[0006] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0007] According to one aspect of the present invention, a method for intelligent real-time quality inspection of live broadcasts in the financial industry is provided.

[0008] An intelligent real-time quality inspection method for live broadcast in the financial industry, specifically comprising:

[0009] S1 obtains the financial service product corresponding to the live broadcast, and uses the historical violations of the financial service product during the live broadcast to determine that the violation risk of the financial service product meets the requirements, and then proceeds to the next step;

[0010] S2 determines the type of violation corresponding to the financial service product based on the type of the financial service product, and determines the violation risk coefficient of the different violation types based on the analysis data of historical violations of the different violation types;

[0011] S3 determines the audience data and interaction data of the financial service product during the historical live broadcast process based on the historical live broadcast data of the financial service product, and determines the matching quality inspection strategy of the financial service product in combination with the violation risk coefficients of different violation types;

[0012] S4 utilizes the matching quality inspection strategy to perform quality inspection during the live broadcast of the financial service product, and utilizes the analysis results of the real-time audience data and real-time interactive data during the live broadcast to determine whether the quality inspection strategy needs to be adjusted.

[0013] The beneficial effects of the present invention are:

[0014] The matching quality inspection strategy for financial service products is determined based on the audience data and interaction data of financial service products during historical live broadcasts and the violation risk coefficients of different violation types. This takes into account the differences in quality inspection requirements for violations of financial service products due to differences in audience data and interaction data during historical live broadcasts, as well as the violation risk situations of different violation types, thereby determining the matching quality inspection strategy for financial service products from multiple angles and avoiding the emergence of technical problems such as untimely discovery and processing of violations due to untimely quality inspection.

[0015] By analyzing the real-time audience data and real-time interactive data during the live broadcast, it is determined whether the quality inspection strategy needs to be adjusted, thereby realizing dynamic adjustment of the quality inspection strategy and avoiding the technical problem that the quality inspection strategy cannot meet the requirements due to excessive or low live broadcast popularity during the live broadcast. This not only ensures the reliability of quality inspection processing, but also improves the real-time matching of quality inspection strategies.

[0016] A further technical solution is that the historical violations of the financial service product during the live broadcast are determined according to the number of historical violations of different violation types of the financial service product during the live broadcast.

[0017] A further technical solution is to determine whether the risk of non-compliance of the financial service product meets the requirements, including:

[0018] Using the historical violations of the financial service product during the live broadcast process, determine the number of live broadcasts of the financial service product with historical violations during the live broadcast process, and use it as the number of live broadcasts with violations;

[0019] The product violation risk of the financial service product is determined based on the number of illegal live broadcasts, and the product violation risk is used to determine whether the violation risk of the financial service product meets the requirements.

[0020] A further technical solution is that the product violation risk of the financial service product is determined based on the ratio of the number of illegal live broadcasts to the number of live broadcasts.

[0021] A further technical solution is that when the product violation risk of the financial service product is greater than a preset risk threshold, it is determined that the violation risk of the financial service product does not meet the requirements.

[0022] A further technical solution is to determine whether the risk of non-compliance of the financial service product meets the requirements, including:

[0023] Using the historical violations of the financial service product during the live broadcast, determine the number of different types of violations of the financial service product during the live broadcast;

[0024] Determine the historical risk coefficients of different types of violations based on the number of different types of violations;

[0025] The product violation risk of the financial service product is determined based on the maximum value of the historical risk coefficients of different violation types, and the product violation risk is used to determine whether the violation risk of the financial service product meets the requirements.

[0026] A further technical solution is to perform quality inspection during the live broadcast of the financial service product, specifically including:

[0027] Based on the matching quality inspection strategy, quality inspection is carried out during the live broadcast of the financial service products.

[0028] A further technical solution is to determine whether the quality inspection strategy needs to be adjusted, including:

[0029] Using the analysis results of the real-time audience data and real-time interaction data during the live broadcast, determine the number of viewers and the number of interactions at different times within the most recent preset duration;

[0030] The live broadcast popularity coefficient at different times is determined by the number of viewers and the number of interactions at different times;

[0031] The live broadcast heat moment is determined based on the live broadcast heat coefficient at different moments, and the number of live broadcast heat moments is used to determine whether the quality inspection strategy needs to be adjusted.

[0032] In a second aspect, the present invention provides a computer system comprising: a memory and a processor that are communicatively connected, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned intelligent real-time quality inspection method for live broadcasts in the financial industry when running the computer program.

[0033] Other features and advantages will be described in the following description. The objects and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the accompanying drawings;

[0036] Figure 1 It is a flow chart of an intelligent real-time quality inspection method for live broadcast in the financial industry;

[0037] Figure 2 It is a flowchart to determine whether the risk of non-compliance of financial services products meets the requirements;

[0038] Figure 3 is a flow chart of a method for determining a violation risk factor for a violation behavior type;

[0039] Figure 4 It is a flow chart of a method for determining a matching quality inspection strategy for a financial services product;

[0040] Figure 5 It is a framework diagram of a computer system. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0042] When conducting quality inspections, when financial service products are recommended and introduced, there will inevitably be violations such as "products guarantee profits". Therefore, it is necessary to conduct quality inspections on violations and intervene in a timely manner to ensure compliance during the live broadcast process.

[0043] Determine whether the risk of violations of financial service products meets the requirements: Use the number of live broadcasts in which financial service products have historical violations during the live broadcast process and count it as the number of illegal live broadcasts. When the number of illegal live broadcasts is less than 5 times, it is determined that the risk of violations of financial service products meets the requirements.

[0044] Violation risk coefficient of violation type: Based on the analysis data of historical violations of the violation type, the number of illegal live broadcasts corresponding to the violation type is determined, and used as the matching live broadcast number. The historical violation number of the violation type in the live broadcast data corresponding to the matching live broadcast number is used to determine the number of screening matches in the matching live broadcast number. The violation risk coefficient of the violation type is determined by the preset risk coefficient corresponding to the screening match number.

[0045] Matching quality inspection strategies: Based on the audience data and interaction data of financial service products during historical live broadcasts, determine the number of interactions and the number of viewers in the historical live broadcasts, and use the number of interactions and the number of viewers to determine the number of popular live broadcasts in the historical live broadcasts. Determine the matching quality inspection strategies in the quality inspection strategies based on the violation risk coefficients and popular live broadcast times of different violation types, and the matching of the violation risk coefficient intervals and popular live broadcast time intervals corresponding to different quality inspection strategies.

[0046] Determine whether the quality inspection strategy needs to be adjusted: Utilize the analysis results of the real-time audience data and real-time interaction data during the live broadcast to determine the number of viewers and the number of interactions at different times within the most recent preset duration. The moments when the number of viewers is greater than the preset number of viewers and the number of interactions is greater than the preset number of interactions are taken as the live broadcast hot moments. When the proportion of the number of live broadcast hot moments is greater than 0.6, it is determined that the quality inspection strategy needs to be adjusted.

[0047] Embodiment 1

[0048] To solve the above problems, according to one aspect of the present invention, Figure 1 As shown, a method for intelligent real-time quality inspection of live broadcasts in the financial industry is provided, which specifically includes:

[0049] S1 obtains the financial service product corresponding to the live broadcast, and uses the historical violations of the financial service product during the live broadcast to determine that the violation risk of the financial service product meets the requirements, and then proceeds to the next step;

[0050] S2 determines the type of violation corresponding to the financial service product based on the type of the financial service product, and determines the violation risk coefficient of the different violation types based on the analysis data of historical violations of the different violation types;

[0051] S3 determines the audience data and interaction data of the financial service product during the historical live broadcast process based on the historical live broadcast data of the financial service product, and determines the matching quality inspection strategy of the financial service product in combination with the violation risk coefficients of different violation types;

[0052] S4 utilizes the matching quality inspection strategy to perform quality inspection during the live broadcast of the financial service product, and utilizes the analysis results of the real-time audience data and real-time interactive data during the live broadcast to determine whether the quality inspection strategy needs to be adjusted.

[0053] Furthermore, the historical violations of the financial service product during the live broadcast are determined according to the number of historical violations of different violation types of the financial service product during the live broadcast.

[0054] Specifically, Figure 2 As shown, it is determined that the risk of non-compliance of the financial service products meets the requirements, including:

[0055] Using the historical violations of the financial service product during the live broadcast process, determine the number of live broadcasts of the financial service product with historical violations during the live broadcast process, and use it as the number of live broadcasts with violations;

[0056] The product violation risk of the financial service product is determined based on the number of illegal live broadcasts, and the product violation risk is used to determine whether the violation risk of the financial service product meets the requirements.

[0057] It should be noted that the product violation risk of the financial service product is determined based on the ratio of the number of illegal live broadcasts to the number of live broadcasts.

[0058] It can be understood that when the product violation risk of the financial service product is greater than a preset risk threshold, it is determined that the violation risk of the financial service product does not meet the requirements.

[0059] Furthermore, it is determined that the risk of non-compliance of the financial service product meets the requirements, including:

[0060] Using the historical violations of the financial service product during the live broadcast, determine the number of different types of violations of the financial service product during the live broadcast;

[0061] Determine the historical risk coefficients of different types of violations based on the number of different types of violations;

[0062] The product violation risk of the financial service product is determined based on the maximum value of the historical risk coefficients of different violation types, and the product violation risk is used to determine whether the violation risk of the financial service product meets the requirements.

[0063] Optionally, determining whether the risk of non-compliance of the financial service product meets the requirements includes:

[0064] S11 uses the historical violations of the financial service product during the live broadcast process to determine the number of live broadcasts in which the financial service product has historical violations during the live broadcast process, and uses it as the number of live broadcasts with violations;

[0065] Optionally, the above step S11 includes steps S111-S113, which are specifically:

[0066] S111 uses the historical violation behaviors of the financial service product during the live broadcast to determine the number of historical violations of the financial service product during the live broadcast. When the number of historical violations of the financial service product during the live broadcast is greater than a preset violation number threshold, it is determined that the violation risk of the financial service product does not meet the requirements. When the number of historical violations of the financial service product during the live broadcast is not greater than the preset violation number threshold, the process proceeds to step S112;

[0067] S112: When the number of historical violations of the financial service product during the live broadcast is within the preset violation number interval, proceed to step S113; when the number of historical violations of the financial service product during the live broadcast is not within the preset violation number interval, proceed to step S12;

[0068] S113 determines the number of live broadcasts in which the financial service product has historical violations during the live broadcast process, and uses it as the number of illegal live broadcasts. When the number of illegal live broadcasts does not meet the requirements, it is determined that the violation risk of the financial service product does not meet the requirements. When the number of illegal live broadcasts meets the requirements, proceed to step S12.

[0069] S12 determines violation coefficients for different numbers of illegal live broadcasts based on the numbers of different types of illegal behaviors for different numbers of illegal live broadcasts;

[0070] Optionally, the above step S12 includes steps S121-S123, specifically:

[0071] S121 determines the number of illegal live broadcasts whose historical violation times are greater than a preset number threshold based on different illegal live broadcast times, and uses it as the number of screened live broadcasts. When the number of screened live broadcasts does not meet the requirements, it is determined that the violation risk of the financial service product does not meet the requirements. When the number of screened live broadcasts meets the requirements, it proceeds to step S122;

[0072] S122 determines the violation coefficients of different illegal live broadcast times based on the number of different illegal behavior types of different illegal live broadcast times, and when there are illegal live broadcast times whose violation coefficients do not meet the requirements, proceeds to step S123; when there are no illegal live broadcast times whose violation coefficients do not meet the requirements, proceeds to step S13;

[0073] S123 uses the number of violation live broadcasts whose violation coefficients do not meet the requirements as the number of screening violation live broadcasts. When the number of screening violation live broadcasts does not meet the requirements, it is determined that the violation risk of the financial service product does not meet the requirements. When the number of screening violation live broadcasts meets the requirements, proceed to step S13.

[0074] S13 determines the product violation risk of the financial service product according to the violation coefficients of different numbers of illegal live broadcasts, and uses the product violation risk to determine whether the violation risk of the financial service product meets the requirements.

[0075] Specifically, Figure 3 As shown, the method for determining the violation risk coefficient of the violation type is:

[0076] Based on the analysis data of historical violations of the violation type, determine the number of illegal live broadcasts corresponding to the violation type, and use it as the matching live broadcast number;

[0077] Determine the screening matching number in the matching live broadcast times by using the historical violation number of the violation behavior type in the live broadcast data corresponding to the matching live broadcast times, the screening matching number being the matching live broadcast times whose historical violation number is not within a preset number interval;

[0078] The violation risk factor of the violation behavior type is determined by screening the matching times.

[0079] It should be noted that the violation risk coefficient of the violation behavior type is determined according to the proportion of the number of screening matches.

[0080] It should also be noted that the audience data during the historical live broadcast process includes the number of viewers at different historical live broadcast times.

[0081] Furthermore, the interaction data during the historical live broadcast process includes the number of interactions for different historical live broadcast times.

[0082] It should be noted that if Figure 4 As shown, the method for determining the matching quality inspection strategy of the financial service product is:

[0083] Based on the audience data and interaction data of the financial service product during the historical live broadcast, determine the number of interactions and the number of viewers in the historical live broadcast times, and use the number of interactions and the number of viewers to determine the number of popular live broadcasts in the historical live broadcast times;

[0084] The matching coefficients with different quality inspection strategies are determined based on the violation risk coefficients of different violation types and the number of popular live broadcasts;

[0085] A matching quality inspection strategy in the quality inspection strategy is determined according to the matching coefficient.

[0086] Furthermore, the matching coefficient with different quality inspection strategies is determined according to the violation risk coefficient of different violation types, the number of hot live broadcasts and the matching number of quality inspection strategies.

[0087] Optionally, the matching quality inspection strategy is a quality inspection strategy with the largest matching coefficient.

[0088] Optionally, the method for determining the matching quality inspection strategy of the financial service product is:

[0089] Based on the audience data and interaction data of the financial service product during the historical live broadcast, determine the number of interactions and the number of viewers in the historical live broadcast times, and use the number of interactions and the number of viewers to determine the number of popular live broadcasts in the historical live broadcast times;

[0090] The matching coefficients with different quality inspection strategies are determined based on the violation risk coefficients of different violation types and the number of popular live broadcasts;

[0091] A matching quality inspection strategy in the quality inspection strategy is determined according to the matching coefficient.

[0092] Specifically, the method for determining the matching quality inspection strategy of the financial service product is:

[0093] Based on the audience data and interaction data of the financial service product during the historical live broadcast, determine the number of interactions and the number of viewers in the historical live broadcast times, and determine the matching quality inspection strategy for different historical live broadcast times according to the number of interactions and the number of viewers;

[0094] Determine the matching quality inspection strategies for different violation types based on the violation risk coefficients of different violation types;

[0095] The matching times of different quality inspection strategies are determined according to the matching quality inspection strategies of different violation types and the matching quality inspection strategies of different historical live broadcast times, and the matching quality inspection strategies in the quality inspection strategies are determined using the matching times.

[0096] Furthermore, the matching quality inspection strategy for different historical live broadcast times is determined according to the matching results of the interaction times and audience numbers of different historical live broadcast times and the preset interaction times and audience numbers corresponding to different quality inspection strategies.

[0097] Optionally, the method for determining the matching quality inspection strategy of the financial service product is:

[0098] S31: determining the number of interactions and the number of viewers in the historical live broadcast times based on the audience data and interaction data of the financial service product in the historical live broadcast process, determining the heat coefficients of different historical live broadcast times according to the number of interactions and the number of viewers, and determining the comprehensive heat coefficient of the financial service product using the heat coefficients of different historical live broadcast times;

[0099] Optionally, the above step S31 includes steps S311-S313, which are specifically:

[0100] S311 determines the number of interactions and the number of viewers in the historical live broadcast times based on the audience data and interaction data of the financial service product in the historical live broadcast process, and determines the heat coefficients of different historical live broadcast times according to the number of interactions and the number of viewers. When there is a historical live broadcast time with a heat coefficient greater than a preset heat coefficient, the process proceeds to step S312; when there is no historical live broadcast time with a heat coefficient greater than the preset heat coefficient, the process proceeds to step S313;

[0101] S312: when the number of historical live broadcasts whose heat coefficient is greater than the preset heat coefficient is greater than the preset number of live broadcasts, a matching quality inspection strategy in the quality inspection strategy is determined based on the preset quality inspection strategy; when the number of historical live broadcasts whose heat coefficient is greater than the preset heat coefficient is not greater than the preset number of live broadcasts, the process proceeds to step S313;

[0102] S313 uses the heat coefficients of different historical live broadcast times to determine the comprehensive heat coefficient of the financial service product. When the comprehensive heat coefficient of the financial service product is greater than the preset comprehensive heat coefficient, the matching quality inspection strategy in the quality inspection strategy is determined based on the preset quality inspection strategy. When the comprehensive heat coefficient of the financial service product is not greater than the preset comprehensive heat coefficient, proceed to step S32.

[0103] S32 determines the product violation coefficient of the financial service product based on the violation risk coefficients of different violation types;

[0104] S33 determines the quality inspection requirement coefficient of the financial service product according to the comprehensive heat coefficient and the product violation coefficient of the financial service product, and uses the quality inspection requirement coefficient to determine a matching quality inspection strategy in the quality inspection strategy.

[0105] A further technical solution is to perform quality inspection during the live broadcast of the financial service product, specifically including:

[0106] Based on the matching quality inspection strategy, quality inspection is carried out during the live broadcast of the financial service products.

[0107] Furthermore, determine whether the quality inspection strategy needs to be adjusted, including:

[0108] Using the analysis results of the real-time audience data and real-time interaction data during the live broadcast, determine the number of viewers and the number of interactions at different times within the most recent preset duration;

[0109] The live broadcast popularity coefficient at different times is determined by the number of viewers and the number of interactions at different times;

[0110] The live broadcast heat moment is determined based on the live broadcast heat coefficient at different moments, and the number of live broadcast heat moments is used to determine whether the quality inspection strategy needs to be adjusted.

[0111] Embodiment 2

[0112] Second, as Figure 5 As shown, the present invention provides a computer system, comprising: a memory and a processor that are communicatively connected, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned intelligent real-time quality inspection method for live broadcasts in the financial industry when running the computer program.

[0113] Optionally, the method for determining the violation risk coefficient of the violation behavior type is:

[0114] Based on the analysis data of historical violations of the violation type, determine the number of historical violations corresponding to the violation type, and use it as the number of matching violations;

[0115] When the number of matching violations is greater than a preset matching number threshold, a preset risk coefficient is used to determine a violation risk coefficient of the violation behavior type;

[0116] When the number of matching violations is not greater than the preset matching number threshold:

[0117] Determine the number of illegal live broadcasts corresponding to the illegal behavior type and use it as the number of matching live broadcasts. When the number of matching live broadcasts is greater than the preset matching index number threshold:

[0118] Determine an average value of the number of matching violations in different matching live broadcast times according to the number of matching violations in different matching live broadcast times; when the average value of the number of matching violations in different matching live broadcast times is greater than a preset matching violation number threshold, determine a violation risk coefficient of the violation behavior type according to a preset risk coefficient;

[0119] When the average number of match violations among different match live broadcast times is not greater than the preset match violation threshold:

[0120] When the number of matching violations is within the corresponding preset matching number interval, the violation risk coefficient of the violation behavior type is determined using the preset risk coefficient;

[0121] When the number of matching live broadcasts is not greater than the preset matching index number threshold or the number of matching violations is not within the corresponding preset matching number interval:

[0122] Determine the number of screening matches in the matching live broadcast times by using the number of historical violations of the violation behavior type in the live broadcast data corresponding to the matching live broadcast times, and when the screening match times are not within a preset screening times interval, determine the violation risk coefficient of the violation behavior type based on a preset risk coefficient;

[0123] When the number of screening matches is within the preset screening number range:

[0124] The violation risk coefficient of the violation behavior type is determined by the number of historical violations in different matching index times.

[0125] Optionally, the above step S32 includes steps S321-S323, specifically:

[0126] When it is determined in step S321 that there is a violation type with a violation risk coefficient greater than the preset violation coefficient based on the violation risk coefficients of different violation types, the process proceeds to step S322; when there is no violation type with a violation risk coefficient greater than the preset violation coefficient, the process proceeds to step S323;

[0127] S322: The violation behavior type whose violation risk coefficient is greater than the preset violation coefficient is used as the screening violation type. When the number of the screened violation types does not meet the requirement, a matching quality inspection strategy in the quality inspection strategy is determined based on the preset quality inspection strategy. When the number of the screened violation types meets the requirement, the process proceeds to step S323.

[0128] S323 determines the product violation coefficient of the financial service product based on the violation risk coefficients of different violation types. When the product violation coefficient of the financial service product does not meet the requirements, a matching quality inspection strategy in the quality inspection strategy is determined based on a preset quality inspection strategy. When the product violation coefficient of the financial service product meets the requirements, proceed to step S33.

[0129] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0130] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0131] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A method for intelligent real-time quality inspection of live broadcasts in the financial industry, characterized in that: Specifically include: Obtain the financial service product corresponding to the live broadcast, and use the historical violations of the financial service product during the live broadcast to determine that the violation risk of the financial service product meets the requirements, and then proceed to the next step; Based on the type of the financial service product, determine the type of violation corresponding to the financial service product, and based on the analysis data of historical violations of different violation types, determine the violation risk coefficients of different violation types; Based on the historical live broadcast data of the financial service product, determine the audience data and interaction data of the financial service product during the historical live broadcast, and determine the matching quality inspection strategy of the financial service product in combination with the violation risk coefficients of different types of violation behaviors; The matching quality inspection strategy is used to perform quality inspection during the live broadcast of the financial service product, and the analysis results of the real-time audience data and real-time interactive data during the live broadcast are used to determine whether the quality inspection strategy needs to be adjusted.

2. The intelligent real-time quality inspection method for live broadcast in the financial industry as claimed in claim 1, characterized in that: Determine that the risk of non-compliance of the financial service products meets the requirements, including: Using the historical violations of the financial service product during the live broadcast process, determine the number of live broadcasts of the financial service product with historical violations during the live broadcast process, and use it as the number of live broadcasts with violations; The product violation risk of the financial service product is determined based on the number of illegal live broadcasts, and the product violation risk is used to determine whether the violation risk of the financial service product meets the requirements.

3. The intelligent real-time quality inspection method for live broadcast in the financial industry as claimed in claim 2, characterized in that: The product violation risk of the financial service product is determined based on the ratio of the number of live broadcasts of the violation to the number of live broadcasts.

4. The intelligent real-time quality inspection method for live broadcast in the financial industry as claimed in claim 2, characterized in that: When the product violation risk of the financial service product is greater than a preset risk threshold, it is determined that the violation risk of the financial service product does not meet the requirements.

5. The intelligent real-time quality inspection method for live broadcast in the financial industry as claimed in claim 1, characterized in that: The method for determining the violation risk coefficient of the violation type is: Based on the analysis data of historical violations of the violation type, determine the number of illegal live broadcasts corresponding to the violation type, and use it as the matching live broadcast number; Determine the screening matching number in the matching live broadcast number by using the historical violation number of the violation behavior type in the live broadcast data corresponding to the matching live broadcast number, wherein the screening matching number is the matching live broadcast number whose historical violation number is not within a preset number interval; The violation risk factor of the violation behavior type is determined by screening the matching times.

6. The intelligent real-time quality inspection method for live broadcast in the financial industry as claimed in claim 1, characterized in that: Conduct quality inspection during the live broadcast of the financial service products, including: Based on the matching quality inspection strategy, quality inspection is carried out during the live broadcast of the financial service products.

7. A computer system comprising: A memory and a processor that are communicatively connected, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor runs the computer program, a method for intelligent real-time quality inspection of live broadcasts in the financial industry as described in any one of claims 1-6 is executed.

Images