Video marketing data analysis method and system based on big data

Abstract

The present invention relates to the technical field of marketing data analysis, and discloses a video marketing data analysis method based on big data, the method comprising the following steps: S1, classifying the marketing videos of self-media platforms based on big data; S2, collecting feedback information data of consumers who purchase products through purchase links provided by marketing videos; S3, respectively counting the relevant parameter data of various marketing videos after classification; S4, comprehensively analyzing the relevant parameter data of various marketing videos and the feedback information data of the consumers, and establishing an evaluation strategy for the brand of video marketing products and the publisher of marketing videos based on the comprehensive analysis. The present invention can not only promote the protection of the legitimate rights and interests of consumers, but also improve the fairness of competition in the video marketing market by analyzing and evaluating the feedback information of consumers of video marketing products and the relevant parameter data of marketing videos.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing data analysis, and in particular to a video marketing data analysis method and system based on big data. Background Art

[0002] The marketing effect of video marketing is also very significant. Video marketing can achieve rapid dissemination and wide coverage of brand information with the help of the influence and follower base of video publishers. By choosing the right video publisher partners and content formats, brands can better reach the target audience and improve brand awareness and reputation. In addition, video marketing can also stimulate the purchasing desire and loyalty of followers, promote product sales and performance growth. Through the recommendation and sharing of video publishers, followers are more likely to have a willingness to buy and trust and rely on the brand. The disadvantage of video marketing is that there are unstable factors: since the influence of video publishers mainly comes from their personal image and follower base, once the video publisher has negative news or loses the attention of followers, its marketing effect will be greatly reduced; high cost investment: it is necessary to pay higher fees to cooperate with well-known video publishers, including product gifts, profit sharing, sponsorship, etc., which may be unbearable for small and medium-sized enterprises.

[0003] For consumers, the final focus is on the quality of the product itself. However, based on the current status of video marketing, it is usually difficult to meet this demand of consumers. The publisher of marketing videos attracts a lot of attention from followers because of their identity as video publishers. The disadvantage is that groups with dual identities as followers and consumers often find it difficult to participate in consumption rationally. The above marketing operations are beneficial to video platforms, product brands and video publishers, but they invisibly destroy the fair competition market environment and bring many adverse effects to consumers. Summary of the invention

[0004] The purpose of the present invention is to provide a video marketing data analysis method and system based on big data to solve the above technical problems.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A video marketing data analysis method based on big data, the method comprising the following steps: S1, categorizing the marketing videos of self-media platforms based on big data; S2, collecting feedback information data of consumers who purchase products through purchase links provided in marketing videos; S3, respectively counting relevant parameter data of each type of marketing videos after categorization; S4, conducting a comprehensive analysis of the relevant parameter data of the various types of marketing videos and the feedback information data of the consumers, and establishing an evaluation strategy for video marketing product brands and marketing video publishers based on the comprehensive analysis.

[0007] As a further technical solution, the process of categorizing the marketing videos on the self-media platform includes:

[0008] S11. Preliminarily classify the marketing videos according to the brand types of the video marketing products;

[0009] S12, marking the various videos that have been initially classified for the first time, marking them as , , … ;

[0010] S13, secondary classification of the marketing videos based on the initial classification according to the product types of the video marketing products;

[0011] S14, marking the videos that have been classified twice for the second time, marking them as , , … .

[0012] As a further technical solution, feedback information data includes the number of consumers who have not rated and the number of negative reviews for each marketing product; relevant parameter data of marketing videos include: video playback volume, likes volume and forwarding volume.

[0013] As a further technical solution, the process of comprehensively analyzing the relevant parameter data of the various marketing videos and the consumer feedback information data includes:

[0014] S41. Evaluate and analyze the number of unevaluated and negative reviews of video marketing products of each brand;

[0015] S42, judging whether each brand meets the preset evaluation criteria according to the evaluation analysis results;

[0016] S43, performing risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet the preset evaluation criteria, and the result of the risk assessment indicates whether there is a risk of information fraud in the marketing videos;

[0017] S44. Issue warnings to publishers of marketing videos that pose a risk of falsifying information.

[0018] As a further technical solution, the process of evaluating and analyzing the number of unevaluated and negative reviews of each brand's video marketing products includes:

[0019] By formula Calculate the negative evaluation coefficient of the video marketing product of the i-th brand ;

[0020] in, is the number of video marketing product categories for the i-th brand; is the first preset weight coefficient corresponding to the jth video marketing product of the i-th brand; is the number of negative reviews for the jth video marketing product of the i-th brand; is the number of unevaluated products of the jth video marketing product of the i-th brand;

[0021] Negative evaluation coefficient With the preset negative review threshold To compare:

[0022] like , then conduct risk assessment on the relevant parameter data of the marketing video corresponding to the i-th brand;

[0023] like , then no risk assessment will be conducted on the relevant parameter data of the marketing video corresponding to the i-th brand.

[0024] As a further technical solution, the process of conducting risk assessment on the relevant parameter data of the marketing videos corresponding to the brands that do not meet the preset evaluation criteria includes:

[0025]

[0026]

[0027]

[0028] By formula Calculate the fraud risk value of the relevant parameter data of the marketing video ;

[0029] in, ; is the amount of data with k as the first place in a set of data; The total amount of a set of data; is the data deviation value; , , They are the data deviation values ​​of video playback volume, like volume and forwarding volume respectively; , , are second preset weight coefficients corresponding to the video playback volume, the number of likes, and the number of reposts, respectively;

[0030] The risk of falsifying With the preset standard value To compare:

[0031] like , it is judged that there is a risk of fraud in the relevant parameter data of the marketing video, and a warning message is issued to the publisher of the corresponding marketing video;

[0032] like , then it is judged that there is no risk of falsification in the relevant parameter data of the marketing video.

[0033] A video marketing data analysis system based on big data, the system comprising:

[0034] The classification module is used to classify the marketing videos on self-media platforms according to the brand and product type of the video marketing products;

[0035] Data collection module, used to collect relevant parameter data of various marketing videos;

[0036] Feedback information management terminal, used to collect consumer feedback information data;

[0037] Data analysis module, used to conduct comprehensive analysis of relevant parameter data of various marketing videos and consumer feedback information data;

[0038] The prompt module is used to establish a warning prompt strategy for marketing video publishers based on comprehensive analysis results.

[0039] As a further technical solution, the data analysis module includes:

[0040] The evaluation analysis unit is used to evaluate and analyze the number of unevaluated and negative reviews of each brand's video marketing products;

[0041] A screening unit, used to screen out marketing videos corresponding to brands that do not meet the preset evaluation criteria according to the evaluation analysis results;

[0042] The risk assessment unit is used to conduct risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet preset evaluation standards.

[0043] Beneficial effects of the present invention:

[0044] The present invention indirectly reflects the quality of products marketed through videos by analyzing and evaluating the feedback information of consumers of video marketing products and relevant parameter data of marketing videos, and sends prompt information to publishers of marketing videos corresponding to products of substandard quality, thereby timely blocking the spread of bad marketing information. Therefore, the present invention can not only promote the protection of the legitimate rights and interests of consumers, but also improve the fairness of competition in the video marketing market. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1It is a flow chart of the video marketing data analysis method based on big data in the present invention;

[0047] Figure 2 This is a block diagram of the content summary of the video marketing data analysis system based on big data in the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, a video marketing data analysis method based on big data includes the following steps: S1, based on big data, the marketing videos of the self-media platform are classified into categories, and the classification distinguishes different brands related to the marketing videos and different products under the same brand. The reason for classifying the marketing videos according to the product types is that the costs, audience groups and content introduction lengths of the corresponding videos of different products of the same brand are different. The result of the classification defines the marketing video as: the marketing video corresponding to the b-type product of the a-brand party; S2, collecting the feedback information data of consumers who purchase the product through the purchase link provided by the marketing video, and the purchase link in the marketing video is traceable, such as using a specific coupon code, tracking link (UTM parameter), etc., so as to distinguish which purchases are made through the video link; S3, respectively counting the relevant parameter data of each type of marketing video after the classification, and the relevant parameter data of the marketing video includes: video playback volume, like volume and forwarding volume, and the relevant parameters can reflect the popularity and influence of the marketing video; S4, comprehensively analyzing the relevant parameter data of the various types of marketing videos and the feedback information data of the consumers, and establishing an evaluation strategy for the brand of the video marketing product and the publisher of the marketing video based on the comprehensive analysis.

[0050] Through the above technical scheme, this embodiment provides a basic process of a video marketing data analysis method based on big data. Specifically, first, the video data of the self-media platform is called from the big data framework, and the marketing videos of the self-media platform are classified. The classification is specifically used to distinguish different brands related to the marketing video and different products under the same brand. Then, the feedback information data of consumers who purchase products through the purchase link provided by the marketing video is collected, and the relevant parameter data of various marketing videos after classification are respectively counted, and then the relevant parameter data of various marketing videos and the feedback information data of the consumers are comprehensively analyzed. On the one hand, it is judged whether the product marketed through the video meets the preset evaluation criteria according to the feedback information of the consumers, and on the other hand, it is analyzed whether there is fraud in the relevant parameter data of the marketing video, and finally, the evaluation strategy for the brand party of the video marketing product and the publisher of the marketing video is established according to the comprehensive analysis. Therefore, this embodiment can indirectly reflect the quality of the product marketed through the video by analyzing and evaluating the feedback information of consumers of the video marketing product and the relevant parameter data of the marketing video, and timely send prompt information to the publisher of the marketing video corresponding to the product with substandard quality, which can not only promote the protection of the legitimate rights and interests of consumers, but also improve the fairness of competition in the video marketing market.

[0051] The process of categorizing marketing videos on self-media platforms includes:

[0052] S11. Preliminarily classify the marketing videos according to the brand types of the video marketing products;

[0053] S12, marking the various videos that have been initially classified for the first time, marking them as , , … ;

[0054] S13, secondary classification of the marketing videos based on the initial classification according to the product types of the video marketing products;

[0055] S14, marking the videos that have been classified twice for the second time, marking them as , , … .

[0056] Through the above technical solution, this embodiment provides a process for categorizing marketing videos on self-media platforms. Specifically, marketing videos are firstly categorized according to the brand types of video marketing products. , , … ,For example This type of video means that the video content is a marketing video of the product provided by the first brand. is the total number of brand types involved in the marketing videos on the video platform. Then, based on the initial classification, the marketing videos are secondary classified according to the product types of the video marketing products. , , … ,For example, Category video means the video content is the marketing video of the first product provided by the first brand. The reason for classifying marketing videos according to product type is that different types of products of the same brand have different costs, target audiences and content introduction lengths of the corresponding videos.

[0057] Feedback information data includes the number of consumers who have not rated and the number of negative reviews for each marketing product; relevant parameter data of marketing videos include: video playback volume, likes volume and forwarding volume.

[0058] Through the above technical solution, this embodiment provides the specific content of feedback information data and marketing video related parameter data. It should be noted that based on the possible existence of fake positive reviews in the market, the number of positive reviews has a certain degree of unreality, so only the number of unevaluated and negative reviews of marketing products are collected.

[0059] The process of comprehensively analyzing the relevant parameter data of the various marketing videos and the consumer feedback information data includes:

[0060] S41. Evaluate and analyze the number of unevaluated and negative reviews of video marketing products of each brand;

[0061] S42, judging whether each brand meets the preset evaluation criteria according to the evaluation analysis results;

[0062] S43, performing risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet the preset evaluation criteria, and the result of the risk assessment indicates whether there is a risk of information fraud in the marketing videos;

[0063] S44. Issue warnings to publishers of marketing videos that pose a risk of falsifying information.

[0064] Through the above technical solution, this embodiment provides a process for comprehensively analyzing the relevant parameter data of various marketing videos and the feedback information data of the consumers. Specifically, firstly, the number of unevaluated and negative reviews of the video marketing products of each brand is evaluated and analyzed, and then the results of the evaluation and analysis are used to determine whether each brand meets the preset evaluation criteria, where the preset evaluation criteria are obtained by fitting experimental data based on market research, which will not be described in detail here. Then, a risk assessment is performed on the relevant parameter data of the marketing videos corresponding to the brands that do not meet the preset evaluation criteria. The results of the risk assessment indicate whether the marketing videos have the risk of information fraud, and finally a warning is issued to the publishers of the marketing videos that have the risk of information fraud.

[0065] The process of evaluating and analyzing the number of unrated and negative reviews of each brand’s video marketing products includes:

[0066] By formula Calculate the negative evaluation coefficient of the video marketing product of the i-th brand ;

[0067] in, is the number of video marketing product categories for the i-th brand; is the first preset weight coefficient corresponding to the jth video marketing product of the i-th brand; is the number of negative reviews for the jth video marketing product of the i-th brand; is the number of unevaluated products of the jth video marketing product of the i-th brand;

[0068] Negative evaluation coefficient With the preset negative review threshold To compare:

[0069] like , then conduct risk assessment on the relevant parameter data of the marketing video corresponding to the i-th brand;

[0070] like , then no risk assessment will be conducted on the relevant parameter data of the marketing video corresponding to the i-th brand.

[0071] Through the above technical solution, this embodiment provides a process for evaluating and analyzing the number of unevaluated and negative reviews of video marketing products of each brand. Specifically, firstly, the formula Calculate the negative evaluation coefficient of the video marketing product of the i-th brand , and then the negative evaluation coefficient With the preset negative review threshold For comparison, when , which reflects that consumers have more negative feedback on the i-th brand products, so risk assessment is conducted on the relevant parameter data of the marketing video corresponding to the i-th brand. It should be noted that is the first preset weight coefficient corresponding to the j-th video marketing product of the i-th brand, which is related to the cost of various products, the target audience and the length of the content introduction of the product corresponding video, and can be obtained by fitting the experimental data, which will not be described in detail here.

[0072] The process of conducting risk assessment on the relevant parameter data of the marketing videos corresponding to brands that do not meet the preset evaluation criteria includes:

[0073]

[0074]

[0075]

[0076] By formula Calculate the fraud risk value of the relevant parameter data of the marketing video ;

[0077] in, ; is the amount of data with k as the first place in a set of data; The total amount of a set of data; is the data deviation value; , , They are the data deviation values ​​of video playback volume, like volume and forwarding volume respectively; , , are second preset weight coefficients corresponding to the video playback volume, the number of likes, and the number of reposts, respectively;

[0078] The risk of falsifying With the preset standard value To compare:

[0079] like , it is judged that there is a risk of fraud in the relevant parameter data of the marketing video, and a warning message is issued to the publisher of the corresponding marketing video;

[0080] like , then it is judged that there is no risk of falsification in the relevant parameter data of the marketing video.

[0081] Through the above technical solution, this embodiment provides a process for risk assessment of relevant parameter data of marketing videos corresponding to brands that do not meet the preset evaluation criteria. Specifically, firstly, the formula Calculate the fraud risk value of the relevant parameter data of the marketing video , and then the fraud risk value With the preset standard value Compare: When When , it is judged that the relevant parameter data of the marketing video has the risk of fraud, and a warning message is issued to the publisher of the corresponding marketing video. , it is judged that there is no risk of fraud in the relevant parameter data of the marketing video. Formula In the specific calculation process, This is the Benford law formula, which is used to calculate the probability standard value of the occurrence of a set of data headed by number k without human intervention , and taking the video playback volume data as an example, the formula It can be concluded that ,in It indicates the ratio of the number of videos with the playback volume starting with k to the total number of videos. It should also be noted that , , It is the second preset weight coefficient corresponding to the video playback volume, the number of likes and the number of reposts respectively, which is related to the ease of falsification of the above data and can be obtained by fitting the experimental data.

[0082] See also Figure 2 As shown, a video marketing data analysis system based on big data, the system includes:

[0083] The classification module is used to classify the marketing videos on the self-media platform according to the brand and product type of the video marketing product. The classification process can be achieved by training a convolutional neural network model, which automatically learns the feature representation in the video. Commonly used models include 3DCNN and temporal convolution, which can consider the information of the time dimension and extract key spatiotemporal features from continuous video frames;

[0084] The data collection module is used to collect relevant parameter data of various marketing videos. Relevant parameter data such as video playback volume, like volume and forwarding volume can be obtained through the analysis tools provided by the video platform;

[0085] Feedback information management terminal, used to collect consumer feedback information data;

[0086] Data analysis module, used to conduct comprehensive analysis of relevant parameter data of various marketing videos and consumer feedback information data;

[0087] The prompt module is used to establish a warning prompt strategy for marketing video publishers based on comprehensive analysis results.

[0088] The data analysis module includes:

[0089] The evaluation analysis unit is used to evaluate and analyze the number of unevaluated and negative reviews of each brand's video marketing products;

[0090] A screening unit, used to screen out marketing videos corresponding to brands that do not meet the preset evaluation criteria according to the evaluation analysis results;

[0091] The risk assessment unit is used to conduct risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet preset evaluation standards.

[0092] Through the above technical solution, the analysis and evaluation of consumers' feedback information on video marketing products and relevant parameter data of marketing videos can indirectly reflect the quality of products marketed through videos, and promptly issue reminder information to publishers of marketing videos corresponding to products of substandard quality, which can not only promote the protection of consumers' legitimate rights and interests, but also improve the fairness of competition in the video marketing market.

[0093] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A video marketing data analysis method based on big data, characterized in that: The method comprises the following steps: S1. Categorize marketing videos on self-media platforms based on big data; S2. Collecting feedback information data from consumers who purchase products through the purchase link provided in the marketing video; S3, respectively counting relevant parameter data of various marketing videos after classification; S4. Comprehensively analyzing the relevant parameter data of the various marketing videos and the consumer feedback information data, and establishing an evaluation strategy for the video marketing product brand and the marketing video publisher based on the comprehensive analysis; The process of categorizing marketing videos on self-media platforms includes: S11. Preliminarily classify the marketing videos according to the brand types of the video marketing products; S12, marking the various videos that have been initially classified for the first time, marking them as , , … ; S13, secondary classification of the marketing videos based on the initial classification according to the product types of the video marketing products; S14, marking the videos that have been classified twice for the second time, marking them as , , … ; Feedback information data includes the number of consumers who have not reviewed and the number of negative reviews for each marketing product; relevant parameter data of marketing videos include: video playback volume, number of likes, and number of reposts; The process of comprehensively analyzing the relevant parameter data of the various marketing videos and the consumer feedback information data includes: S41. Evaluate and analyze the number of unevaluated and negative reviews of video marketing products of each brand; S42, judging whether each brand meets the preset evaluation criteria according to the evaluation analysis results; S43, performing risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet the preset evaluation criteria, and the result of the risk assessment indicates whether there is a risk of information fraud in the marketing videos; S44. Issue warnings to publishers of marketing videos that carry the risk of falsifying information; The process of evaluating and analyzing the number of unrated and negative reviews of each brand’s video marketing products includes: By formula Calculate the negative evaluation coefficient of the video marketing product of the i-th brand ; in, is the number of video marketing product categories for the i-th brand; is the first preset weight coefficient corresponding to the jth video marketing product of the i-th brand; is the number of negative reviews for the jth video marketing product of the i-th brand; is the number of unevaluated products of the jth video marketing product of the i-th brand; Negative evaluation coefficient With the preset negative review threshold To compare: like , then conduct risk assessment on the relevant parameter data of the marketing video corresponding to the i-th brand; like , then no risk assessment will be conducted on the relevant parameter data of the marketing video corresponding to the i-th brand; The process of conducting risk assessment on the relevant parameter data of the marketing videos corresponding to brands that do not meet the preset evaluation criteria includes: ； ； ； By formula ~ Calculate the fraud risk value of the relevant parameter data of the marketing video ; in, ; is the amount of data with k as the first place in a set of data; The total amount of a set of data; is the data deviation value; , , They are the data deviation values ​​of video playback volume, like volume and forwarding volume respectively; , , are second preset weight coefficients corresponding to the video playback volume, the number of likes, and the number of reposts, respectively; The risk of falsifying With the preset standard value To compare: like , it is judged that there is a risk of fraud in the relevant parameter data of the marketing video, and a warning message is issued to the publisher of the corresponding marketing video; like , then it is judged that there is no risk of falsification in the relevant parameter data of the marketing video.

2. Video marketing data analysis system based on big data, characterized by: The system is used to execute the video marketing data analysis method based on big data as described in claim 1, and the system includes: The classification module is used to classify the marketing videos on self-media platforms according to the brand and product type of the video marketing products; Data collection module, used to collect relevant parameter data of various marketing videos; Feedback information management terminal, used to collect consumer feedback information data; Data analysis module, used to conduct comprehensive analysis of relevant parameter data of various marketing videos and consumer feedback information data; The prompt module is used to establish a warning prompt strategy for marketing video publishers based on comprehensive analysis results.

3. The video marketing data analysis system based on big data according to claim 2 is characterized in that: The data analysis module includes: The evaluation analysis unit is used to evaluate and analyze the number of unevaluated and negative reviews of each brand's video marketing products; A screening unit, used to screen out marketing videos corresponding to brands that do not meet the preset evaluation criteria according to the evaluation analysis results; The risk assessment unit is used to conduct risk assessment on relevant parameter data of marketing videos corresponding to brands that do not meet preset evaluation standards.

Images