Method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology

Abstract

The present invention proposes a method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology. The method collects Baidu Index consumer social network search query data, organizes and updates the data, and constructs a social network search query dynamic data set to form an initial domain ontology of cigar consumer demand. On this basis, network text analysis is used to construct a cigar consumer demand evolution ontology corpus to form a cigar consumer evolution demand domain ontology. Finally, a label indicator statistical method and a high-value hot tag recognition method are used to identify consumer hot topics, providing decision support for brand content marketing theme selection, and providing an important reference basis for brand building and market segmentation in digital marketing in the cigar industry.

Description

Technical Field

[0001] The present invention relates to computer technology, and in particular to a method and device for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology. Background Art

[0002] With the continuous expansion of high-end consumer groups and the continuous upgrading of consumption concepts, Chinese cigars have shown a sustained growth trend, but objectively speaking, the market potential and consumption momentum of cigar consumption have not been tapped.

[0003] Content marketing is a key means of promoting domestic cigar brands. Some potential Chinese cigar consumers come from cigarette consumers. Through effective marketing strategies, brand building, and relationship management, we can encourage potential cigar consumers to switch to different categories and brands, thereby increasing sales and brand loyalty. Social network search query data (social network data) has become a crucial data source for research on cigar consumer behavior.

[0004] To this end, how to mine consumer demand themes in social network data, obtain consumer demand data, screen out target consumers, and provide decision-making for cigar brand content marketing becomes a technical issue. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a method and device for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology.

[0006] The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology of the present invention uses a computer device composed of a processor and a memory, and the processor calls a calculation program in the memory to complete the calculation steps. The steps are as follows:

[0007] S1: Collect Baidu Index consumer social network search query data.

[0008] S2: By organizing the search query data, a social network search query dynamic dataset is constructed to form an initial demand ontology of cigar consumers; the demand ontology refers to a demand term set.

[0009] S3: Use network text analysis to identify the implicit needs of potential cigar consumers and transform the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers.

[0010] S4: Based on the evolutionary demand ontology of cigar consumers, a labeling system that adapts to market changes is constructed, and consumer hot topics are identified using label indicator statistics and high-value hot label identification methods.

[0011] The above-mentioned method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology collects Baidu Index consumer social network search query data. The specific steps are as follows:

[0012] Establish candidate keywords related to cigars, determine network search terms based on the candidate keywords, crawl the demand graph data of the previous week on the Internet through the network search terms, and store it in a local document according to the corresponding data dictionary specifications; wherein, the data file is the tabular data formed by reconstructing the data, and the document is saved in the name format of the search term and date.

[0013] The above-mentioned method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, wherein the search query data is collated to construct a social network search query dynamic dataset to form the initial demand ontology of cigar consumers, specifically:

[0014] (1) The search query data is merged, and the searches of the same type of keywords at different times and with different inputs are merged to construct the initial demand ontology of cigar consumers. The initial demand ontology of cigar consumers is normalized to form a dynamic data set of social network search queries.

[0015] (2) Automatically update the search keywords related to the demand graph every week. During the update process, if the initial table contains updated data, it will be directly processed according to the standardized reference table and merged into the search query dynamic data set; if not, the associated keywords will be re-determined and standardized, and the standardized reference terms will be updated to the search query dynamic data set.

[0016] As mentioned above, network text analysis is used to identify the implicit needs of potential cigar consumers and transform the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers, specifically:

[0017] S41: Preprocessing the data of the initial demand ontology of cigar consumers, cleaning, converting and organizing the original data; standardizing the initial demand ontology of cigar consumers; the standardization processing includes three steps: screening out available data and standardizing related keywords, and normalizing.

[0018] S42: Classify the initial demand ontology of cigar consumers and construct a hierarchical structure of the evolutionary demand ontology of cigar consumers. Specifically, the evolutionary demand ontology of cigar consumers is divided into 8 categories, including brand, knowledge, channel, consumption attribute, matching category, life preference, culture, and innovation; and then divided into 53 subcategories at the lower level; complete the construction of the corpus and form the evolutionary demand ontology of cigar consumers.

[0019] The above-mentioned method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology includes:

[0020] Said brand is defined as cigar brand information, tobacco company, and category information;

[0021] Said knowledge is defined as relevant information that consumers want to know through search;

[0022] The channels are defined as the channels through which consumers learn relevant information and purchase products;

[0023] The consumption attributes are defined as the scenarios of cigar consumption recognized by consumers;

[0024] The said pairing category is defined as the correlation between cigars and other categories;

[0025] The life preference is defined as the distribution of interest in crowd attributes;

[0026] Said culture is defined as the cigar cultural concept recognized by consumers;

[0027] The innovation is defined as the taste of Chinese cigars based on consumers' individual brand choices.

[0028] The above-mentioned method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, wherein, as described above, a label system that adapts to market changes is constructed through the cigar consumer evolution demand ontology, and consumer hot topics are identified using label indicator statistics and high-value hot label identification methods, specifically:

[0029] Normalize the ontology of cigar consumer evolutionary demand and divide the data into cigar demand labels.

[0030] In the cigar demand tags, high-value hot topics are identified through tag indicator statistics and high-value hot tag identification methods.

[0031] As mentioned above, the method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, wherein the tag index statistics method and high-value hot tag identification method specifically adopt the RFM model to score cigar demand tags and identify high-value hot topics based on the scoring results.

[0032] The RFM model specifically refers to the weights of the three indicators of proximity, frequency, and value, realizing dynamic weights, and using the hierarchical analysis method to determine the optimal weight parameters, thereby achieving keyword value segmentation of cigar demand tags, obtaining keywords with high-value tags, and using high-value keywords as high-value hot topics for content marketing.

[0033] As mentioned above, in the method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, the RFM model is to establish a cigar demand tag, calculate the RFM value of the keywords associated with the cigar demand tag, and obtain the RFM value of the cigar demand tag. The RFM model is as follows:

[0034]

[0035] Among them, R represents the time distance between the most recent appearance of the associated keyword and the data cutoff, that is, the proximity; F represents the total frequency of the associated keyword appearing within the time span, that is, the frequency; M represents the SIM value of the associated keyword and the network search term, that is, the value; is the weight of R, is the weight of F, is the weight of M, the weight range is (0,1), and satisfies .

[0036] In the above-mentioned method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, the RFM model evaluation results are divided into five result levels, specifically:

[0037] Assign high-value labels with 10≤RFM;

[0038] Assign a general value label with 4≤RFM<10;

[0039] The development label is assigned as important if 2≤RFM<4;

[0040] Assign a general development label when 1≤RFM<2;

[0041] Assign low-value labels if RFM < 1.

[0042] The present invention's device for identifying high-value hot topics in domestic cigar content marketing based on a social network data demand ontology comprises a computer device comprised of a processor and a memory, wherein the processor invokes a computational program in the memory to perform computational steps. The computational program is used to execute the aforementioned method for identifying high-value hot topics in domestic cigar content marketing based on a social network data demand ontology. The computer device comprises:

[0043] The crawler module is used to collect Baidu Index consumer social network search query data.

[0044] The processing module is used to organize the search query data, construct a social network search query dynamic data set, and form an initial demand ontology of cigar consumers; the demand ontology refers to a demand term set.

[0045] The conversion module is used to identify the implicit needs of potential cigar consumers and convert the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers.

[0046] The output module is used to build a labeling system that adapts to market changes through the evolutionary demand ontology of cigar consumers, and to identify consumer hot topics using label indicator statistics and high-value hot label identification methods.

[0047] The effective technical effects of the present invention are as follows: by collecting Baidu Index consumer social network search query data, and through data collation and updating, a social network search query dynamic data set is constructed to form an initial domain ontology of cigar consumer demand; on this basis, network text analysis is used to identify the implicit needs of potential cigar consumers, and a cigar consumer demand evolution ontology corpus is constructed to form a cigar consumer evolution demand domain ontology; finally, based on the cigar demand evolution domain ontology engineering, a label system that adapts to market changes is constructed, and a label indicator statistical method and a high-value hot label identification method are used to identify consumer hot topics, providing decision support for brand content marketing theme selection.

[0048] The data obtained by the method of the present invention is accurate, the algorithm is simple, and the computer system resources are less occupied. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is the Baidu Index web crawler process;

[0050] Figure 2 It is the ontological structure diagram of the evolutionary needs of cigar consumers. DETAILED DESCRIPTION

[0051] In the present invention:

[0052] The "demand ontology" is a special type of terminology that expresses demand. The "cigar consumer initial demand ontology" is a special type of terminology that expresses initially established consumer demand; the "cigar consumer evolved demand ontology" is a special type of terminology that is derived from optimizing the "cigar consumer initial demand ontology."

[0053] "Tag" refers to the business encapsulation of physical layer data information items and is the logical definition of data.

[0054] "Content marketing" refers to a strategy that achieves online marketing goals by creating, publishing, and disseminating valuable content to attract and influence target customers.

[0055] "Keywords" are words that reflect the decisive role of "actors" in social networks. They are essential terms for searching online. They specifically refer to the terms used by individual media outlets when creating and using indexes, such as "cigars" and "long X cigars." "Online search terms" are words or phrases that users enter to describe the information or content they are looking for when searching on search engines.

[0056] Example 1:

[0057] The implementation of the method and / or system of an embodiment of the present invention may involve performing or completing the selected task manually, automatically, or a combination thereof. For example, the hardware for performing the selected task according to an embodiment of the present invention may be implemented as a chip or circuit. The selected task according to an embodiment of the present invention may be implemented as multiple software instructions executed by a computer using any appropriate operating system. In an exemplary embodiment of the present invention, one or more tasks according to the exemplary embodiments of the method and / or system as described herein are performed by a data processor, such as a computing platform for executing multiple instructions. Optionally, the data processor includes volatile storage for storing instructions and / or data and / or non-volatile storage for storing instructions and / or data, for example, a magnetic hard disk and / or removable media. A display and / or user input device, such as a keyboard or mouse, is also provided.

[0058] The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, including the following: an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0059] Computer-readable signal media carries computer-readable program code. These computer program instructions are stored in the computer-readable medium. These instructions cause a computer, other programmable data processing device, or other device to operate in a specific manner. Thus, the code stored in the computer-readable medium generates instructions for the algorithm of the present invention.

[0060] The device of the present invention is a computer device composed of a processor and a memory, wherein the processor calls a calculation program in the memory to complete the calculation steps, and the computer device has:

[0061] Crawler module: used to collect Baidu Index consumer social network search query data.

[0062] Processing module: used to organize the search query data, construct a social network search query dynamic data set, and form an initial demand ontology of cigar consumers; the demand ontology refers to a demand term set.

[0063] Conversion module: used to identify the implicit needs of potential cigar consumers and convert the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers.

[0064] Output module: used to build a labeling system that adapts to market changes through the evolution of cigar consumers' demand ontology, and identify consumer hot topics using label indicator statistics and high-value hot label recognition methods.

[0065] Example 2:

[0066] The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology of the present invention is implemented by executing the following steps 1 to 4 on a computer.

[0067] Step 1:

[0068] Collect Baidu Index consumer social network search query data; establish candidate keywords related to cigars, and determine network search terms through candidate keywords.

[0069] The data source for this invention is the Baidu Index search engine. Baidu search engine data includes Baidu Index data, demand graph data, and demographic data. The demand graph data reflects the demand for related search terms as consumers' search behavior changes related to the search term, and can better reflect the other needs of users who pay attention to the keyword. It belongs to the social network data type studied in this invention. In the Baidu Index demand graph interface, three data points for related keywords can be obtained: the search change rate (Ratio), the related keyword relevance (SIM), and the related keyword search popularity (PV).

[0070] Search Change Rate (RATIO): This is calculated by combining the relevance of the keyword with related terms, as well as the search demand for the related terms themselves. The distance of the related terms from the center of the circle represents the strength of the related terms' relevance; the size of the related terms themselves represents their own search index.

[0071] Related keyword relevance (SIM): The relevance index between the search term and the related term, which refers to the degree of relevance between the search term and the related term. It is directly provided by the distance value of the Baidu Index Demand Map-Demand Related Words interface. The closer the distance, the higher the relevance.

[0072] Search popularity (PV): Page views. Related words (PV) refer to the page views of related words, indicating the popularity of related words.

[0073] Combining the CNKI database, industry terminology and Baidu search candidate words, a total of candidate keywords were selected and shown in Table 1 below.

[0074] Table 1 Candidate keywords

[0075]

[0076] The demand graph data of the previous week is crawled on the Internet through online search terms, and stored in local documents according to the corresponding data dictionary specifications; Baidu Index data is collected through web crawlers, and the Baidu data obtained are all structured data. The Baidu Index web crawler process is as follows, reference Figure 1 :

[0077] (1) After the crawler program is started, it automatically reads and enters the Baidu Index homepage.

[0078] (2) Read the local configuration table and load user data. Load candidate keyword data and corresponding time and space parameters.

[0079] (3) Read the current date and use the date function to calculate the day of the week. The file is saved as "search term + date" to facilitate management and later data processing.

[0080] (4) Crawl the demand graph data of the previous week according to the current date and store it in a local document according to the corresponding data dictionary specification. The data file is the tabular data formed by reconstructing the data, and the document is saved in the name format of the search term and date.

[0081] Step 2: Organize the search query data: Since the data table obtained by the crawler is in weekly units and the same type of network search terms contains multiple words, based on this, the data is merged. Different time and different input searches of the same type of keywords are merged to construct the initial demand ontology of cigar consumers. The initial demand ontology of cigar consumers is normalized to form a dynamic data set of social network search queries.

[0082] Automatic updates of demand graph-related search keywords are performed every week. During the update process, if the initial table contains updated data, it is directly processed according to the standardized reference table and merged into the search query dynamic data set; if not, the associated keywords are re-determined and standardized, and the standardized reference terms are updated to the search query dynamic data set.

[0083] Step 3: Use network text analysis to identify the implicit needs of potential cigar consumers and transform the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers, specifically:

[0084] Step 3.1: Preprocess the data on the initial cigar consumer demand ontology by cleaning, transforming, and organizing the raw data. This involves cleaning, transforming, and organizing the raw data to ensure data quality and usability. Raw data often contains missing values, outliers, and duplicate values. Data preprocessing improves data quality.

[0085] By pre-processing the data for transformation, such as logarithmic transformation, normalization, standardization, etc., it is possible to ensure that the data meets the assumptions and requirements of the model. The present invention performs outlier removal, text standardization and normalization according to the particularity of the acquired data.

[0086] The standardization of the initial demand ontology of cigar consumers includes three steps: screening available data, standardizing related keywords, and normalizing them. Specifically, it includes:

[0087] Standardization of cigar consumers’ initial demand ontology:

[0088] After the initial ontology of the cigar demand domain is constructed, the original data of cigar social network search queries needs to be standardized:

[0089] (1) Cigar and cigarette brand specifications usually appear in the initial table of demand map data. Tobacco industry enterprises have multiple brands, and each brand has multiple specifications. Due to the diversity of search terms, in order to facilitate the calculation of various data labels and the analysis of market changes based on specifications, the same specification of cigars and cigarettes is standardized by adding prefixes. The standardization method is shown in Table 2.

[0090] Table 2 Standardized reference table

[0091]

[0092] (2) Normalization is performed to compress all data into the range [0, 1], so that the mathematical units of the data remain consistent. When a data point is exactly the minimum value, it is normalized to 0; if the data point is exactly the maximum value, it is normalized to 1. For example, the processed data of cigars are shown in Table 3.

[0093] Table 3 Normalization results of the online search term “cigar”

[0094]

[0095] Step 3.2: Classify the initial demand ontology of cigar consumers, construct the hierarchical structure of the cigar consumer evolutionary demand ontology, complete the construction of the corpus, and form the cigar consumer evolutionary demand ontology.

[0096] The initial demand ontology of cigar consumers is transformed into the evolutionary demand ontology of cigar consumers. The evolutionary demand ontology of cigar consumers is divided into three layers. The first layer is divided into 8 categories, and there are 53 subcategories under the first layer. Figure 2 .

[0097] The eight categories in the first layer of the cigar consumer evolutionary demand ontology are specifically defined as follows:

[0098] Brand: cigar and cigarette related brand information, tobacco companies, category information, etc.

[0099] Knowledge: The relevant information that consumers want to know through search;

[0100] Channel: the channel through which consumers learn relevant information and purchase products;

[0101] Consumption attributes: In what scenarios are cigars consumed and recognized?

[0102] Pairing categories: Which categories are cigars related to?

[0103] Lifestyle Preferences: Combined with Baidu Index’s demographic attributes – interest distribution, this provides insights into current consumer lifestyle preferences, serving as crucial reference data for cigar content marketing.

[0104] Culture: cultural concepts.

[0105] Innovation: Consumers are most concerned about health-related issues when it comes to cigars, with ingredients being a key factor. Taste, driven by individual needs, is a key factor in consumer brand selection.

[0106] Through the above social network data processing, a corpus was constructed and the ontology of cigar consumers' evolutionary needs was formed, as shown in Table 4.

[0107] Table 4 Ontology of the evolving needs of cigar consumers (term set or corpus)

[0108]

[0109] Step 4:

[0110] Through the evolutionary demand ontology of cigar consumers, a labeling system that adapts to market changes is constructed, and consumer hot topics are identified using label indicator statistics and high-value hot label recognition methods.

[0111] The evolutionary demand ontology of cigar consumers is standardized and the data is divided into cigar demand labels.

[0112] The cigar demand tagging system is based on the cigar consumer demand evolution ontology corpus, combined with a tag category system construction method. Using the cigar demand evolution domain ontology as a key reference, it standardizes the description of tobacco and social search domain concepts, terms, and their relationships, and divides the data into demand tags. Because this system is targeted at brand content marketing needs, tagging construction varies depending on the application scenario. Ultimately, 53 tags were constructed for each online search term, as shown in Table 5.

[0113] Table 5 Label construction results

[0114]

[0115] Based on the cigar demand tag system, the tag indicator statistics method and the high-value hot tag identification method are used to identify high-value hot topics. The tag indicator statistics method is based on the construction of the cigar demand tag system. By counting the number of tag samples, PV index and SIM index, high-value tags are identified. The tag indicator statistics method and the high-value hot tag identification method use an improved RFM model, introduce the RFM model into the demand tag to build a "high-value model", score the demand tag based on the RFM "high-value model", and identify hot topics based on the scoring results. The details are as follows:

[0116] 1. High-value tag identification based on tag indicator statistics.

[0117] The statistical method, based on the construction of a cigar demand tag system, identifies high-value tags by analyzing the number of tag samples, PV index, and SIM index. The tag results for each online search term are shown in Table 6 below.

[0118] Table 6 Statistics of cigar three-level labels

[0119]

[0120] 2. High-value tag identification based on high-value hotspot tag identification method.

[0121] Through the cigar demand labeling system, the RFM model is adopted to identify the high-value labels of cigar consumers within a specified time period.

[0122] Based on the construction of the keyword RFM model, the present invention fully considers the weights of the three indicators of proximity, frequency and value, realizes dynamic weights, combines the hierarchical analysis method, determines the optimal weight parameters, realizes keyword value segmentation, and identifies high-value keywords.

[0123] (2.1) Construct RFM model.

[0124] The present invention introduces the RFM model into the cigar demand tag to construct a "high-value model". The demand tag is used as the research object, and the RFM value of the associated keywords in the tag is calculated to obtain the RFM value of the demand tag, thereby identifying its value and predicting the development trend of the corresponding tag. In the conventional RFM model, R represents the time distance from the most recent appearance of the associated keyword to the data cutoff; F represents the total frequency of the associated keywords appearing within the time span; and M represents the SIM value of the associated keywords and the network search terms. In the conventional RFM model, the present invention introduces a weighted average to comprehensively measure the value of each tag, and provides the RFM calculation method of the present invention. The weights of each indicator are obtained according to the hierarchical analysis method, and then the RFM values ​​of each tag are calculated by adding them up, that is:

[0125]

[0126] In the above formula, R, F, and M are the proximity, frequency, and value of the keyword respectively; are the weights of R, F, and M, ranging from (0,1) and satisfying .

[0127] (2.2) Calculate the RFM value. Through data calculation, the RFM results of each online search term are shown in Table 7 below.

[0128] Table 7 Statistics of improved RFM indicators for cigar three-level labels

[0129]

[0130] (2.3) Definition of label value hierarchy.

[0131] After classifying the tags, we define the tag value hierarchy as eight value types: important and average value customers, important development and retention customers, average development and retention customers, and important and average retention customers. The tag value hierarchy is divided into five levels, as defined in Table 8. Substituting the calculated RFM index into the table yields a high-value tag. This indicates that "Film, TV, and Entertainment" is a high-value tag.

[0132] Table 8 Definition of label value levels

[0133]

[0134] Marketing content theme selection. Tag-based content marketing theme selection relies on identifying high-value tags. Through the aforementioned research methods, we can identify high-value, hot consumer tags to help optimize brand marketing strategies. By selecting content marketing themes based on the tag system, we can discover which tags resonate with and interest our target consumers, thereby optimizing content marketing strategies and increasing the relevance and conversion rates of marketing campaigns. Tag-based theme selection provides data-driven decision-making and enables the development of more forward-looking market strategies.

[0135] The above are specific examples of the present invention, which further explain the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention. In the above examples, each example is not a single example. If there is a part that is not fully explained in a certain example, other examples can be referred to. The various examples can be reasonably combined to form a new example that can achieve the purpose of the present invention.

Claims

1. A method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, wherein a computer device composed of a processor and a memory is used, and the processor calls a calculation program in the memory to complete the calculation steps, characterized in that: The steps are as follows: S1: Collect Baidu Index consumer social network search query data; S2: By organizing the search query data, a dynamic data set of social network search queries is constructed to form an initial demand ontology of cigar consumers; The demand ontology refers to a set of demand terms. The steps include: merging search query data, merging searches of the same keyword at different times and with different inputs to construct an initial demand ontology for cigar consumers, normalizing the initial demand ontology for cigar consumers to form a dynamic data set of social network search queries; and storing the data in an initial table. S3: Using network text analysis, identify the implicit needs of potential cigar consumers and transform the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers; including the following steps: S31: Preprocessing the cigar consumer initial demand ontology data, cleaning, converting and organizing the original data; standardizing the cigar consumer initial demand ontology; the standardization process includes three steps: screening available data and related keyword standardization, and normalization; S32: Classify the initial demand ontology of cigar consumers and construct a hierarchical structure of the cigar consumer evolutionary demand ontology. Specifically, the cigar consumer evolutionary demand ontology is divided into 8 categories, including brand, knowledge, channel, consumption attribute, pairing category, life preference, culture, and innovation; and then divided into 53 subcategories at the lower level; The corpus construction was completed, and the ontology of cigar consumers’ evolving needs was formed; S4: Based on the evolutionary demand ontology of cigar consumers, a labeling system that adapts to market changes is constructed, and consumer hot topics are identified using label indicator statistics and high-value hot label identification methods.

2. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 1, characterized in that: Collect Baidu Index consumer social network search query data. The specific steps are as follows: Establish candidate keywords related to cigars, determine network search terms based on the candidate keywords, crawl demand graph data for the previous time period on the Internet through the network search terms, and store them in a local document after sorting according to the specifications; wherein the data file is tabular data formed by reconstructing the data, and the document is saved in the name format of the search term and date.

3. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 1, characterized in that: The method of organizing the search query data to construct a dynamic data set of social network search queries and forming an ontology of initial demand of cigar consumers specifically includes: Automatic updates of demand graph-related search keywords are performed every week. During the update process, if the initial table contains updated data, it is directly processed according to the standardized reference table and merged into the search query dynamic data set; if not, the keyword is re-determined and standardized, and the standardized reference terms are updated to the search query dynamic data set.

4. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 1, characterized in that: Said brand is defined as cigar brand information, tobacco company, and category information; Said knowledge is defined as relevant information that consumers want to know through search; The channels are defined as the channels through which consumers learn relevant information and purchase products; The consumption attributes are defined as the scenarios of cigar consumption recognized by consumers; The said pairing category is defined as the correlation between cigars and other categories; The life preference is defined as the distribution of interest in crowd attributes; Said culture is defined as the cigar cultural concept recognized by consumers; The innovation is defined as the taste of Chinese cigars based on consumers' individual brand choices.

5. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 1, characterized in that: The above mentioned method builds a labeling system that adapts to market changes through the ontology of cigar consumer evolutionary needs, and uses label indicator statistics and high-value hot label identification methods to identify consumer hot topics, specifically: Normalize the ontology of cigar consumer evolutionary demand and divide the data into cigar demand labels. In the cigar demand tags, high-value hot topics are identified through tag indicator statistics and high-value hot tag identification methods.

6. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 5, characterized in that: in, The tag index statistics method and high-value hot tag identification method specifically use the RFM model to score cigar demand tags and identify high-value hot topics based on the scoring results. The RFM model specifically refers to the weights of the three indicators of proximity, frequency, and value, realizing dynamic weights, and using the hierarchical analysis method to determine the optimal weight parameters, thereby achieving keyword value segmentation of cigar demand tags, obtaining keywords with high-value tags, and using high-value keywords as high-value hot topics for content marketing.

7. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 6, characterized in that: The RFM model is to establish a cigar demand tag, calculate the RFM value of the keywords associated with the cigar demand tag, and obtain the RFM value of the cigar demand tag. The RFM model is as follows: Among them, R represents the time distance between the most recent appearance of the associated keyword and the data cutoff, that is, the proximity; F represents the total frequency of the associated keyword appearing within the time span, that is, the frequency; M represents the SIM value of the associated keyword and the network search term, that is, the value; is the weight of R, is the weight of F, is the weight of M, the weight range is (0,1), and satisfies .

8. The method for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology according to claim 7, characterized in that: The RFM model evaluation results are divided into five result levels, specifically: Assign high-value labels with 10≤RFM; Assign a general value label with 4≤RFM<10; The development label is assigned as important if 2≤RFM<4; Assign a general development label when 1≤RFM<2; Assign low-value labels if RFM < 1.

9. A device for identifying high-value hot topics in domestic cigar content marketing based on social network data demand ontology, comprising a computer device composed of a processor and a memory, wherein the processor calls a calculation program in the memory to complete the calculation steps, characterized in that: The computer program is used to execute the method according to any one of claims 1 to 8; The computer device comprises: The crawler module is used to collect Baidu Index consumer social network search query data; a processing module for collating the search query data, constructing a dynamic data set of social network search queries, and forming an initial demand ontology of cigar consumers; the demand ontology refers to a set of demand terms; The transformation module is used to identify the implicit needs of potential cigar consumers and transform the initial demand ontology of cigar consumers into the evolved demand ontology of cigar consumers; The output module is used to build a labeling system that adapts to market changes through the evolutionary demand ontology of cigar consumers, and to identify consumer hot topics using the high-value hot label recognition method.

Images