Brand operation method based on gold nail principle

Abstract

The invention provides a brand operation method based on a gold nail principle. The brand operation method comprises the following steps: acquiring a mainstream network propagation medium of a product corresponding to a brand needing to be created; according to a mainstream network propagation medium, obtaining a development history and a traditional culture story of a product corresponding to the brand needing to be created; a deep learning algorithm is adopted to score all development history and traditional culture stories of the product, and according to the development history and the traditional culture stories with the highest score, age-fitting re-creation is carried out to generate brand symbols; and the brand symbols are propagated through a mainstream network propagation medium to complete brand operation. According to the invention, a deep learning algorithm is adopted to mine the attention degree of a user on a certain product development history and a traditional culture story in mainstream media, a brand theme is extracted according to the most representative product development history and the traditional culture story, and a brand is propagated through mainstream network media. Therefore, the user can quickly and deeply impress the brand and better understand the new meaning released by the brand.

Description

Technical Field

[0001] The present invention relates to the technical field of brand operation, and in particular to a brand operation method based on the golden nail principle. Background Art

[0002] At present, the communication technology in the world is highly developed, and the information circulation mode of the previous world communication order has undergone subversive changes. The relationship between communication and production, consumption and wealth growth is getting closer and closer. Along this thinking path, there can be many complex academic implications and intertwined discourses that can be expressed, but no matter how it is expressed, it is just the rationalization and practical development of the basic concept context, that is, the current communication operation method cannot make the image represented by the brand deeply rooted in the hearts of the people. If we change the logic and use brand discourse to analyze, there will be a "freshness" of conceptual experience. To achieve better practice and development results, it is still necessary to make use of the overall operation rules of the brand.

[0003] It can be said that the globalization approach is a way of operating the "whole world" as a big brand, which goes beyond the national brand approach. Therefore, as long as the complex operating rules of the whole world are clearly projected into the brand rules, brands have become the great discovery and reinvention of mankind, and have become people's way of thinking, survival and lifestyle to grasp the world; therefore, the brand approach that is common in our daily life needs more theoretical thinking. Summary of the invention

[0004] In view of this, the present invention proposes a brand operation method based on the golden nail principle to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention proposes a brand operation method based on the golden nail principle, which is characterized by comprising:

[0006] Obtain the mainstream online media for the products corresponding to the brand to be created;

[0007] Obtain the development history and traditional cultural stories of the products corresponding to the brand to be created based on the mainstream online communication media;

[0008] Use deep learning algorithms to score all the product's development history and traditional cultural stories, and generate brand symbols based on the highest-scoring development history and traditional cultural stories;

[0009] The brand symbol is spread through the mainstream network communication media to complete the brand operation.

[0010] Furthermore, the process of obtaining the mainstream online communication media corresponding to the brand to be created includes:

[0011] Obtain the operating region of the product brand and obtain the user activity of all media platforms in the operating region;

[0012] An activity threshold is constructed, and the media platform whose user activity exceeds the activity threshold is used as the mainstream network communication medium.

[0013] Furthermore, the process of obtaining the development history and traditional cultural stories of the products corresponding to the brand to be created according to the mainstream network communication media includes:

[0014] Obtain the historical overview and cultural significance of the product through the academic paper platform, and obtain the historical nodes from the product's creation to its development based on the historical overview;

[0015] Generate search keywords according to the historical nodes and product names;

[0016] Search the search keyword through the mainstream network communication media to obtain relevant articles at each historical node of the product development, and obtain the corresponding historical stories at each historical node based on the relevant articles;

[0017] The development history of the product is generated according to the historical story and the historical nodes, and the traditional cultural story is generated according to the historical story.

[0018] Furthermore, the process of using a deep learning algorithm to score all of the product's development history and traditional cultural stories includes:

[0019] Identify the historical stories corresponding to each historical node of the product, obtain the importance of each historical story to the product development, assign a value to the importance, and encode to generate first coded data;

[0020] Count the number of comments and views under the relevant articles at each development history node of the product, and encode the statistical results to generate second coded data;

[0021] Respectively converting the first encoded data and the second encoded data into a first data set and a second data set, and respectively setting weights for the first data set and the second data set;

[0022] A deep learning model is trained according to the first data set and the second data set, and the trained deep learning model is used for scoring.

[0023] Furthermore, in the process of counting the number of comments, a sentiment analysis algorithm is used to realize intelligent identification and distinction between positive and negative comments under relevant articles, and the statistics are completed after the negative comments are eliminated.

[0024] Furthermore, the process of using sentiment analysis algorithm to realize intelligent identification and distinction of positive and negative comments under relevant articles includes:

[0025] Construct sentiment terms representing positive and negative sentiment terms, extract features of comments under the article based on the sentiment terms, and obtain sentiment tendency features;

[0026] A deep learning model is trained according to the sentiment tendency features, and the trained deep learning model is used for intelligent identification and distinction of positive comments and negative comments under relevant articles.

[0027] Furthermore, the encoding process includes:

[0028] Constructing a relationship between the assignment result of the importance and the coding number, arranging each group of numbers according to the size of the assignment result, filling in the coding sequence, and generating the first coding data;

[0029] After sorting the number of comments and the number of views according to the statistical results, a relationship between the sorting results and the coding labels is constructed, each group of labels is arranged according to the size of the assignment results, the coding sequence is filled in, and the first coding data is generated.

[0030] Furthermore, the process of disseminating the brand symbol through the mainstream network communication media includes:

[0031] A number of articles are generated by combining the brand symbol with the development history and traditional cultural stories of the corresponding products, and the articles are published and pushed to users through the mainstream network communication media. The number of comments and the number of views under the articles are counted, and a user response threshold is constructed. When the number of comments and the number of views both exceed the user response threshold, the articles are pushed in a loop. When the number of comments and the number of views both exceed the user response threshold, it is determined that the brand operation is completed.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention adopts deep learning algorithms to mine the degree of attention paid by users in mainstream media to the development history of a product and traditional cultural stories, extracts brand themes based on the most representative product development history and traditional cultural stories, generates brand symbols, and disseminates the brand through mainstream network media, so that users can quickly form a deep impression of the brand and better understand the new ideas released by the brand, effectively solving the technical problem that existing brand operations cannot make the image represented by the brand deeply rooted in the hearts of the people. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0035] Figure 1 It is a flow chart of the brand operation method based on the golden nail principle of the present invention;

[0036] Figure 2 This is a flow chart of sentiment analysis and feature extraction performed by the sentiment analysis algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] This embodiment proposes a brand operation method based on the golden nail principle, wherein the historical background and culture of the "golden nail" originate from the completion of the Pacific Railway, the world's first railway connecting the Atlantic and Pacific Oceans and crossing the American continent. In order to permanently commemorate this achievement, the last four special spikes with special significance were nailed into the last joint of the railway. The last spike is an 18k gold spike weighing 17.5 carats, which marks the successful completion of the great project of the Pacific Railway. Later, the term "golden nail" was used to refer to the last spike with symbolic significance in railway construction, and is also often used to refer to the completion of other significant projects or events. It can be seen that, under the premise of being familiar with the historical and cultural background, users often think of the great project of the Pacific Railway and the story behind it when they see or hear the symbol of the "golden nail".

[0039] This embodiment applies the "Golden Nail" principle to the field of brand communication technology, and makes the brand image deeply rooted in the hearts of the people through a structural communication method. Furthermore, the Golden Nail Principle includes an upper structure, a middle structure and a lower structure. The upper structure represents the brand to be created and the ideas contained in the brand, the middle structure represents various network communication media of the brand, and the lower structure includes social groups that understand and are familiar with the brand through a variety of network communication media. By obtaining the brand development history and traditional cultural stories related to the products corresponding to the brand to be created and with the highest public response in the communication media, and refining the brand theme based on the brand development history and traditional cultural stories, creating brand symbols, and finally circulating the brand symbols through a variety of network communication media, the social groups will have a deep impression of the brand symbols, thus completing the brand operation.

[0040] like Figure 1 As shown, the method described in this embodiment specifically includes the following steps:

[0041] Obtain the mainstream online media for the products corresponding to the brand to be created;

[0042] Obtain the development history and traditional cultural stories of the products corresponding to the brand to be created based on the mainstream online communication media;

[0043] Use deep learning algorithms to score all the product's development history and traditional cultural stories, and generate brand symbols based on the highest-scoring development history and traditional cultural stories;

[0044] The brand symbol is spread through the mainstream network communication media to complete the brand operation.

[0045] As a preferred embodiment, the process of obtaining the mainstream network communication media corresponding to the product of the brand to be created includes:

[0046] Obtain the operating region of the product brand and obtain the user activity of all media platforms in the operating region;

[0047] An activity threshold is constructed, and the media platform whose user activity exceeds the activity threshold is used as the mainstream network communication medium.

[0048] As a preferred embodiment, the process of obtaining the development history and traditional cultural stories of the products corresponding to the brand to be created according to the mainstream network communication media includes:

[0049] Obtain the historical overview and cultural significance of the product through the academic paper platform, and obtain the historical nodes from the product's creation to its development based on the historical overview;

[0050] Generate search keywords according to the historical nodes and product names;

[0051] Search the search keyword through the mainstream network communication media to obtain relevant articles at each historical node of the product development, and obtain the corresponding historical stories at each historical node based on the relevant articles;

[0052] The development history of the product is generated according to the historical story and the historical nodes, and the traditional cultural story is generated according to the historical story.

[0053] For example, when you need to create a car brand, you can use academic paper-related platforms to obtain the historical process of the car from its birth to its development and perfection, the cultural traditions represented by the car, and the related traditional stories in the development process. You can count the nodes that have an important impact on the car in the overall development history of the car, generate search keywords based on the development nodes and the car name, and then obtain the historical stories and traditional culture contained in the car at different development nodes.

[0054] As a preferred embodiment, the process of using a deep learning algorithm to score all the development history and traditional cultural stories of the product includes:

[0055] Identify the historical stories corresponding to each historical node of the product, obtain the importance of each historical story to the product development, assign a value to the importance, and encode to generate first coded data;

[0056] Count the number of comments and views under the relevant articles at each development history node of the product, and encode the statistical results to generate second coded data;

[0057] Respectively converting the first encoded data and the second encoded data into a first data set and a second data set, and respectively setting weights for the first data set and the second data set;

[0058] A deep learning model is trained according to the first data set and the second data set, and the trained deep learning model is used for scoring.

[0059] As a preferred embodiment, in the process of counting the number of comments, a sentiment analysis algorithm is used to realize intelligent identification and distinction between positive and negative comments under relevant articles, and the statistics are completed after the negative comments are eliminated.

[0060] As a preferred embodiment, the process of using sentiment analysis algorithm to realize intelligent identification and distinction of positive comments and negative comments under relevant articles includes:

[0061] Construct sentiment terms representing positive and negative sentiment terms, extract features of comments under the article based on the sentiment terms, and obtain sentiment tendency features;

[0062] A deep learning model is trained according to the sentiment tendency features, and the trained deep learning model is used for intelligent identification and distinction of positive comments and negative comments under relevant articles.

[0063] Specifically, sentiment analysis algorithms mainly rely on established sentiment terms. The establishment of sentiment terms is the premise and basis of sentiment classification. In actual use, they can be divided into four categories: general sentiment terms, degree adverbs, negative terms, and domain terms. The synonymy and near-synonymy relationships between words in the sentiment dictionary are used to judge the sentiment tendency of the user's sentiment terms, and the sentiment polarity of the opinion is judged on this basis. Chinese sentiment analysis is mainly an expansion of Hownet. The semantic similarity calculation method is used to calculate the semantic similarity between words and the benchmark sentiment word set, so as to infer the sentiment tendency of the word. In addition, a special domain dictionary can be established to improve the accuracy of sentiment classification, such as establishing a new network vocabulary dictionary to more accurately grasp the sentiment tendency of new words. Based on the method of sentiment terms, the text is first pre-processed by word segmentation, and then the sentiment dictionary constructed in advance is used to match the text string to mine positive and negative information.

[0064] After completing the construction of sentiment terms, sentiment analysis needs to be performed using the sentiment term text matching algorithm. Figure 2 As shown in the figure, the text matching algorithm based on terms is relatively simple. The words in the sentence after word segmentation are traversed one by one. If the word hits the dictionary, the corresponding weight is processed. The weight of positive words is addition, the weight of negative words is subtraction, the weight of negation words is the opposite, and the weight of degree adverbs is multiplied by the weight of the words it modifies. Using the final output weight value, it is possible to distinguish whether it is positive, negative or neutral sentiment. The sentiment analysis model based on sentiment terms is simple and easy to implement, and has universal and generalization. The dictionary needs to be constantly refreshed to add new words. In the current era of the continuous emergence of network vocabulary, if the refresh speed of the dictionary cannot keep up with the speed of the emergence of new words, then the sentiment analysis will be far from the expectation in actual use. At the same time, the sentiment segmentation of the comments under the article is also weighted, so as to determine the emotional tendency of the user's comment.

[0065] Sentiment analysis is calculated as follows:

[0066]

[0067] Where T n is the frequency of each word in the text, δ is the sentiment value corresponding to different types of articles, which can be set in the sentiment item, v nThe sentiment coefficient corresponding to each part of speech of each word segment can be set according to the above model to train the model. V is the text sentiment analysis value. The sentiment coefficient of a single word needs to fully consider the part of speech. For positive, negative, and negative words, the part of speech of the previous word of the word needs to be considered. The calculation formula is as follows:

[0068] V n =β(v n-1 *v n )+K,V n =β(v n-1 *v n )-K,V n =-βv n

[0069] If the previous word of a positive word is a negative or negative word, the sentiment coefficient value of the word will decrease. If it is a positive or positive word, the sentiment coefficient value will increase. The weight of positive words is addition, the weight of negative words is subtraction, the weight of negative words is the opposite, and the weight of degree adverbs is multiplied by the weight of the words it modifies. The above formulas correspond to the calculation of sentiment coefficients when the previous word is a positive word, a negative word, and a negative word, respectively, where β is the weight of the degree adverb and K is the corresponding weighted value.

[0070] The sentiment scores of positive and negative comments of the article are realized according to the calculation results of the sentiment coefficients of positive words, negative words and negation words in the article comments. After feature extraction of relevant sentiment words, a data set is constructed for model training.

[0071] The deep learning model is trained with the dataset generated by sentiment analysis. The training set trains the variational autoencoder, i.e., the deep learning model. The deep learning model is trained using the Adam optimizer. The loss function of the VAE model training is expressed as follows:

[0072] min L=λ 1 NMSE ( x ) +λ 2 KL ( x)

[0073]

[0074] Where NMSE is the normalized root mean square error, which is used to measure the difference between the slice matrix generated by the decoder, i.e., the reconstructed sample x, and the original slice matrix, i.e., the original sample x. 0 The gap between them; KL represents the Kullback-Leibler divergence, which is used as a regular term in the loss function to ensure the generation ability of the VAE model. 1 and 2represents the corresponding item weight coefficient, μ and σ represent the mean and standard deviation of each component of the latent variable in the latent space, Represents the expected function of the encoder input data operation result, Represents the expected function of the encoder output data operation result.

[0075] As a preferred embodiment, the process of encoding the importance of the product, the number of article comments, and the number of page views includes:

[0076] Constructing a relationship between the assignment result of the importance and the coding number, arranging each group of numbers according to the size of the assignment result, filling in the coding sequence, and generating the first coding data;

[0077] After sorting the number of comments and the number of views according to the statistical results, a relationship between the sorting results and the coding labels is constructed, each group of labels is arranged according to the size of the assignment results, the coding sequence is filled in, and the first coding data is generated.

[0078] As a preferred embodiment, the process of spreading the brand symbol through the mainstream network communication media includes:

[0079] A number of articles are generated by combining the brand symbol with the development history and traditional cultural stories of the corresponding products, and the articles are published and pushed to users through the mainstream network communication media. The number of comments and the number of views under the articles are counted, and a user response threshold is constructed. When the number of comments and the number of views both exceed the user response threshold, the articles are pushed in a loop. When the number of comments and the number of views both exceed the user response threshold, it is determined that the brand operation is completed.

[0080] For example, after generating a brand symbol, you can introduce the main ideas and products contained in the brand symbol, and then you can disassemble the structure of the brand symbol, and explain the product development history and traditional cultural stories behind each component symbol (the development history and traditional cultural stories refer to the development history and traditional cultural stories with the highest scores obtained by the deep learning algorithm). When the explanation of each part of the structure of the brand symbol is completed, the article corresponding to the brand symbol is generated, and the article is circulated and pushed through the main online media platforms of regional users. At the same time, the number of comments and views under the article is counted, and a user response threshold is constructed. When the threshold is met, the brand operation is judged to be completed.

[0081] This embodiment uses a deep learning algorithm to mine the degree of attention paid by users in mainstream media to the development history and traditional cultural stories of a product, extracts brand themes based on the most representative product development history and traditional cultural stories, generates brand symbols, and disseminates the brand through mainstream online media, so that users can quickly form a deep impression of the brand and better understand the new ideas released by the brand, effectively solving the technical problem that existing brand operations cannot make the image represented by the brand deeply rooted in people's hearts.

[0082] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or perform equivalent replacements on some of the technical features thereof; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A brand operation method based on the golden nail principle, characterized in that: include: Obtain the mainstream online media for the products corresponding to the brand to be created; Obtain the development history and traditional cultural stories of the products corresponding to the brand to be created based on the mainstream online communication media; Use deep learning algorithms to score all the product's development history and traditional cultural stories, and generate brand symbols based on the highest-scoring development history and traditional cultural stories; The brand symbol is spread through the mainstream network communication media to complete the brand operation.

2. The brand operation method based on the golden nail principle according to claim 1 is characterized in that: The process of obtaining the mainstream online media for the products that need to be created includes: Obtain the operating region of the product brand and obtain the user activity of all media platforms in the operating region; An activity threshold is constructed, and the media platform whose user activity exceeds the activity threshold is used as the mainstream network communication medium.

3. The brand operation method based on the golden nail principle according to claim 1 is characterized in that: The process of obtaining the development history and traditional cultural stories of the products corresponding to the brand to be created based on the mainstream network communication media includes: Obtain the historical overview and cultural significance of the product through the academic paper platform, and obtain the historical nodes from the product's creation to its development based on the historical overview; Generate search keywords according to the historical nodes and product names; Search the search keyword through the mainstream network communication media to obtain relevant articles at each historical node of the product development, and obtain the corresponding historical stories at each historical node based on the relevant articles; The development history of the product is generated according to the historical story and the historical nodes, and the traditional cultural story is generated according to the historical story.

4. The brand operation method based on the golden nail principle according to claim 1 is characterized in that: The process of using deep learning algorithms to score all the product's development history and traditional cultural stories includes: Identify the historical stories corresponding to each historical node of the product, obtain the importance of each historical story to the product development, assign a value to the importance, and encode to generate first coded data; Count the number of comments and views under the relevant articles at each development history node of the product, and encode the statistical results to generate second coded data; Respectively converting the first encoded data and the second encoded data into a first data set and a second data set, and respectively setting weights for the first data set and the second data set; A deep learning model is trained according to the first data set and the second data set, and the trained deep learning model is used for scoring.

5. The brand operation method based on the golden nail principle according to claim 4 is characterized in that: In the process of counting the number of comments, a sentiment analysis algorithm is used to realize intelligent identification and distinction between positive and negative comments under relevant articles, and the statistics are completed after the negative comments are removed.

6. The brand operation method based on the golden nail principle according to claim 5 is characterized in that: The process of using sentiment analysis algorithms to intelligently identify and distinguish positive and negative comments under relevant articles includes: Construct sentiment terms that represent positive and negative sentiments respectively, extract features of comments under the article based on the sentiment terms, and obtain sentiment tendency features; A deep learning model is trained according to the sentiment tendency features, and the trained deep learning model is used for intelligent identification and distinction of positive comments and negative comments under relevant articles.

7. The brand operation method based on the golden nail principle according to claim 4 is characterized in that: The encoding process includes: Constructing a relationship between the assignment result of the importance and the coding number, arranging each group of numbers according to the size of the assignment result, filling in the coding sequence, and generating the first coding data; After sorting the number of comments and the number of views according to the statistical results, a relationship between the sorting results and the coding labels is constructed, each group of labels is arranged according to the size of the assignment results, the coding sequence is filled in, and the first coding data is generated.

8. The brand operation method based on the golden nail principle according to claim 1 is characterized in that: The process of spreading the brand symbol through the mainstream network communication media includes: A number of articles are generated by combining the brand symbol with the development history and traditional cultural stories of the corresponding products, and the articles are published and pushed to users through the mainstream network communication media. The number of comments and the number of views under the articles are counted, and a user response threshold is constructed. When the number of comments and the number of views both exceed the user response threshold, the articles are pushed in a loop. When the number of comments and the number of views both exceed the user response threshold, it is determined that the brand operation is completed.

Images