Feedback analysis method and device for cigarette user

Abstract

The invention relates to the technical field of data information analysis, in particular to a feedback analysis method and device for cigarette users. The method comprises the following steps: collecting feedback data of a plurality of cigarette users for each cigarette brand and a user portrait of each cigarette user; the user portrait comprises the smoking age, the smoking frequency and the income capacity; analyzing the feedback data according to a preset rule to obtain a scoring result of the corresponding cigarette brand, and mining a corresponding relationship between the user portrait and the scoring result based on an association rule algorithm; dividing each cigarette user into a plurality of user groups according to the corresponding relationship to determine the cigarette preference of each user group; the problem of inaccurate prediction of the cigarette market caused by neglecting the feedback information of the consumer on the cigarette can be effectively solved.

Description

Technical Field

[0001] The present application relates to the technical field of data information analysis, and in particular to a method and device for analyzing feedback from cigarette users. Background Art

[0002] Cigarettes are a special commodity circulating in the market, and their production and sales need to be strictly managed and restricted; however, as a commodity, user feedback analysis can effectively grasp consumer demand, improve product quality, and predict market trends, so that the sales process can better serve consumers.

[0003] At present, the prediction of cigarette market behavior mostly stays at the prediction of sales data direction, but ignores the feedback information of consumers on cigarettes, which makes the prediction of cigarette market inaccurate. Summary of the invention

[0004] Based on this, it is necessary to provide a cigarette user feedback analysis method and device to address the above technical issues.

[0005] In a first aspect, the present application provides a method for analyzing feedback from cigarette users, the method comprising:

[0006] Collect feedback data from multiple cigarette users on each cigarette brand, as well as the user profile of each cigarette user; the user profile includes smoking age, smoking frequency and income ability;

[0007] The feedback data is parsed according to preset rules to obtain the scoring results of the corresponding cigarette brands, and the corresponding relationship between user portraits and scoring results is mined based on the association rule algorithm;

[0008] According to the corresponding relationship, each cigarette user is divided into multiple user groups to determine the cigarette preference of each user group.

[0009] In one embodiment, the step of parsing the feedback data according to a preset rule to obtain a scoring result of the corresponding cigarette brand includes:

[0010] Identify the type of feedback data; the types of feedback data include cigarette ratings and cigarette reviews;

[0011] In the case where the type of feedback data is a cigarette score, normalizing the cigarette score corresponding to the score result, and converting the normalized cigarette score into the score result;

[0012] In the case where the type of feedback data is cigarette evaluation, executing the evaluation analysis strategy includes: performing feature comparison between the cigarette evaluation and the preset keyword text to determine whether the keyword feature exists; if so, extracting the context structure in the keyword feature, and evaluating and identifying the context structure through a pre-established recurrent neural network; the evaluation and identification results include favorable reviews with adjectives, favorable reviews, favorable reviews with transitions, unfavorable reviews, and unfavorable reviews with adjectives, with decreasing probability of favorable reviews; establishing a mapping relationship between the evaluation and identification results and the scoring results, and obtaining the scoring results based on the cigarette evaluation based on the mapping relationship.

[0013] In one embodiment, the cigarette brand corresponds to a plurality of cigarette ingredients; the method further comprises:

[0014] Determine the corresponding cigarette ingredients based on the cigarette brands preferred by each user group and use them as target cigarette ingredients respectively;

[0015] According to the target cigarette ingredients, the discreteness analysis of the cigarette ingredients of the cigarette brands on sale is carried out respectively, and the preset thresholds corresponding to the cigarette ingredients are set;

[0016] In response to the dispersion of each cigarette component of any cigarette brand on sale exceeding the corresponding preset threshold compared with the dispersion of the target cigarette component, it is marked as an unpopular cigarette brand.

[0017] In one embodiment, the cigarette ingredients include tar and nicotine; the method further comprises:

[0018] Taking the lowest tar and nicotine content as the goal, the dispersion of the cigarette ingredients of each brand of cigarettes on sale compared with the target cigarette ingredients is traversed;

[0019] According to the traversal results, the cigarette brand on sale with the lowest tar content and the lowest nicotine content is selected as the most suitable brand for sale to replace the brand corresponding to the target cigarette ingredients.

[0020] In one embodiment, the method further comprises:

[0021] Obtain the geographical distribution of sales terminals of each cigarette brand on sale, and randomly select multiple pre-collected user portraits of the geographical distribution as user portraits to be identified;

[0022] Obtaining multiple feature vectors based on the portraits of each user in the user group, and extracting a first preset number of feature vectors as support vectors;

[0023] Calculate the distance between each support vector and the hyperplane according to multiple feature vectors, the user group corresponding to each feature vector and the preset hyperplane parameters;

[0024] If the minimum value of the distance between each support vector and the hyperplane is less than the preset value, the hyperplane parameters are updated; if the minimum value of the distance between each support vector and the hyperplane is not less than the preset value, the support vector machine is set to the trained support vector machine;

[0025] The user portraits to be identified are classified according to the trained support vector machine.

[0026] In one embodiment, the method further comprises:

[0027] If there is any user profile to be identified that cannot find a decision boundary in the support vector machine, then the user profile to be identified is marked;

[0028] Obtain feedback data corresponding to the marked user portrait to be identified, obtain a scoring result based on the corresponding feedback data analysis, determine the correspondence between the user portrait to be identified and the obtained scoring result, and divide the user portrait to be identified into user groups based on the correspondence.

[0029] In a second aspect, the present application provides a cigarette user feedback analysis device, the device comprising:

[0030] The collection module is used to collect feedback data from multiple cigarette users on each cigarette brand, as well as the user profile of each cigarette user; the user profile includes smoking age, smoking frequency and income ability;

[0031] The parsing module is used to parse the feedback data according to preset rules to obtain the scoring results of the corresponding cigarette brands, and to mine the corresponding relationship between the user portrait and the scoring results based on the association rule algorithm;

[0032] The division module is used to divide each cigarette user into multiple user groups according to the corresponding relationship to determine the cigarette preference of each user group.

[0033] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method provided in the first aspect of the present application are implemented.

[0034] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method provided in the first aspect of the present application.

[0035] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method provided in the first aspect of the present application.

[0036] The above-mentioned cigarette user feedback analysis method and device can collect feedback data of multiple cigarette users on various cigarette brands, as well as user portraits of each cigarette user, so as to accurately identify the preferences of users with different smoking ages, smoking frequencies and income abilities for cigarette brands, and quantify and segment the preference levels through preset rules. Since each user may be the same or similar in one dimension of the user portrait, and the same and similar to other users in other dimensions, the user groups are re-divided based on the correspondence between the segmented preference levels and smoking ages, smoking frequencies and income abilities, respectively, so as to accurately identify the real preferences of user groups with different user portraits for cigarette brands, and effectively solve the problem of inaccurate predictions of the cigarette market caused by ignoring consumers' feedback information on cigarettes. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 A flowchart of the steps of dividing cigarette user groups in one embodiment;

[0039] Figure 2 A flowchart of steps for obtaining a scoring result based on feedback data in one embodiment;

[0040] Figure 3 A flowchart of the steps of performing a dispersion analysis on cigarette brands on sale in one embodiment;

[0041] Figure 4 A flowchart of steps for replacing a brand of a target cigarette component in one embodiment;

[0042] Figure 5 A flowchart of the steps of classifying a user portrait to be identified in one embodiment;

[0043] Figure 6 A flowchart of the steps of dividing user groups into identified user portraits in one embodiment;

[0044] Figure 7 FIG. 1 is a schematic diagram of a frame structure of a cigarette user feedback analysis device in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0047] In an exemplary embodiment, a method for analyzing cigarette user feedback is provided, such as Figure 1 As shown, the method includes the following steps S102 to S106. Among them:

[0048] S102, collecting feedback data from multiple cigarette users on each cigarette brand, as well as a user portrait of each cigarette user; the user portrait includes smoking age, smoking frequency, and income ability.

[0049] Specifically, since different cigarette brands have different tastes, and the demand for strong-tasting cigarettes and light-tasting cigarettes varies depending on the age and frequency of smoking, and each cigarette brand corresponds to a different price, when the price exceeds the user's economic income and expectations, the user will give different feedback due to the cost-effectiveness of the cigarette. Therefore, the true feedback of users with different cigarette demands on any brand of cigarettes can be determined by analyzing the user portraits of cigarette users.

[0050] Specifically, the collected users may be representative volunteer users who meet the conditions in the user portrait.

[0051] S104, parsing the feedback data according to preset rules to obtain the scoring results of the corresponding cigarette brands, and mining the corresponding relationship between the user portrait and the scoring results based on the association rule algorithm.

[0052] Specifically, the preset rules are used to normalize different feedback types and convert them into scoring results that can quantify user experience.

[0053] On the other hand, since the user portrait includes characteristics of multiple dimensions such as smoking age, smoking frequency and income ability, and the characteristics of each dimension are actually related to the user's rating results for multiple different brands of cigarettes, while giving full marks to some brands of cigarettes, other brands of cigarettes may also be given high scores second only to full marks. This is actually the impact of the user portraits of each dimension as the user's focus on the rating results. Therefore, the association rule algorithm can be used to analyze which brands users with different user portraits focus on more.

[0054] S106, dividing each cigarette user into a plurality of user groups according to the corresponding relationship, so as to determine the cigarette preference of each user group.

[0055] Furthermore, by analyzing which brands users with different user portraits focus on, we can divide the user groups to determine which user groups favor each brand of cigarettes.

[0056] The embodiment of the present application aims to provide a feedback analysis method for cigarette users, which can collect feedback data of multiple cigarette users on various cigarette brands, as well as user portraits of various cigarette users, so as to accurately identify the preferences of users with different smoking ages, smoking frequencies and income abilities for cigarette brands, and quantify and segment the preference levels through preset rules. Since each user may be the same or similar in one dimension of the user portrait, and the same and similar to other users in other dimensions, the user groups are re-divided based on the correspondence between the segmented preference levels and smoking ages, smoking frequencies and income abilities, so as to accurately identify the true preferences of user groups with different user portraits for cigarette brands.

[0057] In an exemplary embodiment, Figure 2 As shown, a method for analyzing cigarette user feedback further includes the following steps S1042 to S1046. Among them:

[0058] S1042, identifying the type of feedback data; the types of feedback data include cigarette ratings and cigarette evaluations.

[0059] Specifically, the cigarette rating may be a direct rating given by a user to the cigarette, and the cigarette evaluation may be a language-based review.

[0060] S1044: When the type of the feedback data is a cigarette score, normalization processing is performed on the cigarette score corresponding to the score result, and the normalized cigarette score is converted into a score result.

[0061] Specifically, in the scoring process, since sales terminals are dispersed, and questionnaires or scoring sheets may not have uniform formats or scoring standards, the cigarette scores can be normalized to correspond to the scoring results.

[0062] S1046, when the type of feedback data is cigarette evaluation, executing the evaluation analysis strategy includes: performing feature comparison between the cigarette evaluation and the preset keyword text to determine whether the keyword feature exists; if so, extracting the context structure in the keyword feature, and evaluating and identifying the context structure through a pre-established recurrent neural network; the evaluation and identification results include favorable reviews with adjectives, favorable reviews, favorable reviews with transitions, unfavorable reviews, and unfavorable reviews with adjectives, with decreasing probability of favorable reviews; establishing a mapping relationship between the evaluation and identification results and the scoring results, and obtaining the scoring results based on the cigarette evaluation based on the mapping relationship.

[0063] Specifically, by pre-setting keyword feature templates, the praise rates are logically sorted and judged according to praise with adjectives, praise, praise with transitions, bad reviews, and bad reviews with adjectives, and further mapped to the scoring results.

[0064] In an exemplary embodiment, Figure 3 As shown, the method also includes the following steps S202 to S206.

[0065] in:

[0066] S202: Determine corresponding cigarette ingredients based on the cigarette brands preferred by each user group and use them as target cigarette ingredients respectively.

[0067] It is understandable that cigarette brands correspond to cigarette ingredients. When a user has a preference for a certain cigarette brand, it is actually a preference for the cigarette ingredients. Therefore, the user preference is further quantified through the cigarette ingredients.

[0068] S204: performing a discrete degree analysis on the cigarette components of the cigarette brands on sale according to the target cigarette components, and setting preset thresholds corresponding to the cigarette components.

[0069] Specifically, the discreteness of each cigarette component is analyzed to quantify the difference between the taste of the cigarette and the taste of the target cigarette.

[0070] S206: In response to the dispersion of each cigarette component of any cigarette brand on sale exceeding the corresponding preset threshold compared with the dispersion of the target cigarette component, mark it as an unpopular cigarette brand.

[0071] Specifically, the preset threshold can be used to measure the obvious difference in the taste of cigarettes that users can feel. User opinions can be collected in the form of decreasing component proportions during the specific implementation process, and the preset threshold can be given after reference to a large amount of data.

[0072] Exemplarily, the rating results of multiple users may be extracted respectively according to cigarette brands, and the ratings may be preset, and the preset ratings may be mapped to preset thresholds regarding cigarette ingredients.

[0073] In an exemplary embodiment, the cigarette ingredients include tar amount and nicotine amount; Figure 4 As shown, the method further includes the following steps S302 to S304. Among them:

[0074] S302, with the lowest tar and nicotine content as the goal, iterate through the dispersion of cigarette ingredients of each cigarette brand on sale compared with the target cigarette ingredients.

[0075] Specifically, tar and nicotine levels, as the two chemical components that are most harmful to the body, have always been required to be controlled compulsorily. Therefore, when the discreteness does not exceed the preset threshold, indicating that the user can accept the taste of the cigarette, the lowest tar and nicotine levels should be selected to ensure the health of the user.

[0076] S304: According to the traversal result, the cigarette brand on sale with the lowest tar content and the lowest nicotine content is selected as the most suitable brand to replace the brand corresponding to the target cigarette composition.

[0077] Furthermore, when the tar and nicotine levels are at their lowest, they can be appropriately promoted to user groups to replace high-harm cigarette brands.

[0078] In an exemplary embodiment, Figure 5 As shown, the method further includes the following steps S402 to S410. Among them:

[0079] S402, obtaining the distribution areas of sales terminals of each cigarette brand on sale, and randomly selecting a plurality of pre-collected user portraits of the distribution areas as user portraits to be identified.

[0080] Specifically, in the process of launching cigarette brands, people in each region have relatively different tastes for cigarettes, so user portraits of the local population are randomly sampled to obtain the average smoking age, smoking frequency and income of users.

[0081] S404, obtaining a plurality of feature vectors based on the portraits of each user in the user group, and extracting a first preset number of feature vectors as support vectors.

[0082] Specifically, before classification is performed using a support vector machine, a portion of support vectors is selected from the feature vectors to train the support vector machine.

[0083] Furthermore, the feature vectors correspond to user groups

[0084] S406, calculating the distance between each support vector and the hyperplane according to the multiple feature vectors, the user group corresponding to each feature vector and the preset hyperplane parameters.

[0085] S408, if the minimum value of the distance between each support vector and the hyperplane is less than a preset value, the hyperplane parameters are updated; if the minimum value of the distance between each support vector and the hyperplane is not less than the preset value, the support vector machine is set to the trained support vector machine.

[0086] S410, classifying the user portrait to be identified according to the trained support vector machine.

[0087] Specifically, the support vector machine can effectively process high-dimensional data and has good generalization ability by constructing the optimal hyperplane, thus performing well in the classification task of user portraits.

[0088] In an exemplary embodiment, Figure 6 As shown, the method further includes the following steps S502 to S504.

[0089] in:

[0090] S502: If there is any user portrait to be identified for which a decision boundary cannot be found in the support vector machine, the user portrait to be identified is marked.

[0091] Specifically, when the decision boundary cannot be found, it means that the user portraits extracted in the distribution area exceed the user portraits of cigarette users collected during the user group segmentation process, and therefore the user portraits need to be re-divided into user groups.

[0092] S504, obtaining feedback data corresponding to the marked user portrait to be identified, obtaining a scoring result based on the corresponding feedback data analysis, determining a corresponding relationship between the user portrait to be identified and the obtained scoring result, and dividing the user portrait to be identified into user groups based on the corresponding relationship.

[0093] Specifically, feedback data generated by the user portraits in the distribution area are collected, and the feedback data generated by the user portraits in the distribution area are re-analyzed according to the method adopted in steps S102 to S106 to obtain a scoring result. Since the scoring result is consistent with the scoring result of the aforementioned method, the correspondence between the feedback data generated by the user portraits in the distribution area and the user portraits in the distribution area is obtained again, and they are divided into corresponding user groups to finally determine the user group of each user portrait in the distribution area, and the user group of each user portrait in the distribution area is analyzed based on the aforementioned analysis of the user group.

[0094] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0095] Based on the same inventive concept, the embodiment of the present application also provides a cigarette user feedback analysis device for implementing the above-mentioned cigarette user feedback analysis method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations of one or more embodiments of the cigarette user feedback analysis device provided below can refer to the limitations of the cigarette user feedback analysis method above, and will not be repeated here.

[0096] Second, as Figure 7 As shown, the present application provides a cigarette user feedback analysis device 700, the device comprising:

[0097] The collection module 701 is used to collect feedback data of multiple cigarette users on each cigarette brand, as well as a user profile of each cigarette user; the user profile includes smoking age, smoking frequency and income capacity;

[0098] The parsing module 702 is used to parse the feedback data according to preset rules to obtain the scoring results of the corresponding cigarette brands, and mine the corresponding relationship between the user portrait and the scoring results based on the association rule algorithm;

[0099] The division module 703 is used to divide each cigarette user into multiple user groups according to the corresponding relationship to determine the cigarette preference of each user group.

[0100] Each module in the above-mentioned cigarette user feedback analysis device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0101] In an exemplary embodiment, the parsing module further includes:

[0102] An identification unit, used for identifying the type of feedback data; the type of feedback data includes cigarette rating and cigarette evaluation;

[0103] A normalization processing unit, for performing normalization processing on the cigarette score corresponding to the score result when the type of the feedback data is a cigarette score, and converting the normalized cigarette score into a score result;

[0104] The evaluation analysis strategy unit is used to execute the evaluation analysis strategy when the type of feedback data is cigarette evaluation, including: comparing the cigarette evaluation with the preset keyword text to determine whether the keyword feature exists; if so, extracting the context structure in the keyword feature, and evaluating and identifying the context structure through a pre-established recurrent neural network; the evaluation and identification results include good reviews with adjectives, good reviews, good reviews with transitions, bad reviews, and bad reviews with adjectives, with the probability of good reviews decreasing in order; establishing a mapping relationship between the evaluation and identification results and the scoring results, and obtaining the scoring results based on the cigarette evaluation based on the mapping relationship.

[0105] In an exemplary embodiment, a cigarette user feedback analysis device further includes:

[0106] An ingredient determination module is used to determine the corresponding cigarette ingredients based on the cigarette brands preferred by each user group and use them as target cigarette ingredients respectively;

[0107] An analysis module, used to perform discrete degree analysis on the cigarette components of the cigarette brands on sale according to the target cigarette components, and to set preset thresholds corresponding to the respective cigarette components;

[0108] The marking module is used to mark any cigarette brand on sale as an unpopular cigarette brand in response to the discreteness of each cigarette component of the cigarette brand exceeding the corresponding preset threshold compared with the discreteness of the target cigarette component.

[0109] In an exemplary embodiment, the cigarette ingredients include tar and nicotine, and a cigarette user feedback analysis device further includes:

[0110] A traversal module is used to traverse the discreteness of the cigarette components of each cigarette brand on sale compared with the target cigarette components with the lowest tar and nicotine content as the goal;

[0111] The promotion module is used to select the cigarette brand with the lowest tar and nicotine content on sale as the most suitable promotion brand according to the traversal results, and promote it to the user group corresponding to the target cigarette ingredients.

[0112] In an exemplary embodiment, a cigarette user feedback analysis device further includes:

[0113] An acquisition module is used to acquire the distribution areas of sales terminals of various cigarette brands on sale, and randomly select multiple pre-collected user portraits of the distribution areas as user portraits to be identified;

[0114] An extraction module, configured to obtain a plurality of feature vectors based on the portraits of each user in the user group, and extract a first preset number of feature vectors as support vectors;

[0115] A calculation module, used to calculate the distance between each support vector and the hyperplane according to multiple feature vectors, the user group corresponding to each feature vector and preset hyperplane parameters;

[0116] An updating module, configured to update the hyperplane parameters if the minimum value of the distances between each support vector and the hyperplane is less than a preset value; and to set the support vector machine to a trained support vector machine if the minimum value of the distances between each support vector and the hyperplane is not less than a preset value;

[0117] The classification module is used to classify the user portraits to be identified based on the trained support vector machine.

[0118] In an exemplary embodiment, the marking module is further used to mark the user portrait to be identified if there is any user portrait to be identified for which a decision boundary cannot be found in the support vector machine;

[0119] The acquisition module is also used to obtain feedback data corresponding to the marked user portrait to be identified, obtain scoring results based on the corresponding feedback data analysis, determine the correspondence between the user portrait to be identified and the obtained scoring results, and divide the user portrait to be identified into user groups based on the correspondence.

[0120] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the remote update method of the above application when executing the computer program.

[0121] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the automatic deployment and recovery method of the buoy as described above.

[0122] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the automatic deployment and recovery method of the buoy as described above.

[0123] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0124] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0125] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for analyzing cigarette user feedback, characterized in that: The method comprises: Collect feedback data from multiple cigarette users on each cigarette brand, as well as a user profile of each cigarette user; the user profile includes smoking age, smoking frequency, and income capacity; Parsing the feedback data according to preset rules to obtain a scoring result corresponding to the cigarette brand, and mining the corresponding relationship between the user portrait and the scoring result based on an association rule algorithm; The cigarette users are divided into a plurality of user groups according to the corresponding relationship to determine the cigarette preference of each user group.

2. The method according to claim 1, characterized in that The step of parsing the feedback data according to preset rules to obtain a scoring result corresponding to the cigarette brand includes: Identifying the type of the feedback data; the type of the feedback data includes cigarette ratings and cigarette evaluations; In the case where the type of the feedback data is a cigarette score, performing normalization processing on the cigarette score corresponding to the score result, and converting the normalized cigarette score into the score result; In the case where the type of feedback data is cigarette evaluation, executing the evaluation analysis strategy includes: performing feature comparison between the cigarette evaluation and a preset keyword text to determine whether a keyword feature exists; if so, extracting a context structure in the keyword feature, and evaluating and identifying the context structure through a pre-established recurrent neural network; the evaluation and identification results include favorable reviews with adjectives, favorable reviews, favorable reviews with transitions, unfavorable reviews, and unfavorable reviews with adjectives, with decreasing probability of favorable reviews; establishing a mapping relationship between the evaluation and identification results and the scoring results, and obtaining a scoring result based on the cigarette evaluation based on the mapping relationship.

3. The method according to claim 1, characterized in that The cigarette brand corresponds to a plurality of cigarette ingredients; the method further comprises: Determine corresponding cigarette ingredients based on the cigarette brands preferred by each user group and use them as target cigarette ingredients respectively; According to the target cigarette components, discrete degree analysis is performed on the cigarette components of the cigarette brands on sale, and preset thresholds corresponding to the cigarette components are set; In response to the dispersion of each cigarette component of any cigarette brand on sale exceeding the corresponding preset threshold compared with the dispersion of the target cigarette component, it is marked as an unpopular cigarette brand.

4. The method according to claim 3, characterized in that The cigarette ingredients include tar and nicotine; the method further includes: Taking the minimum tar amount and nicotine amount as the target, traverse the dispersion of the cigarette components of each cigarette brand on sale compared with the target cigarette components; According to the traversal results, the cigarette brand on sale with the lowest tar content and the lowest nicotine content is used as the most suitable brand for sale to replace the brand corresponding to the target cigarette composition.

5. The method according to claim 1, characterized in that: The method further comprises: Obtain the distribution area of ​​sales terminals of each cigarette brand on sale, and randomly select multiple pre-collected user portraits of the distribution area as user portraits to be identified; Obtaining a plurality of feature vectors based on the portraits of each user in the user group, and extracting a first preset number of feature vectors as support vectors; Calculating the distance between each support vector and the hyperplane according to a plurality of feature vectors, the user group corresponding to each feature vector and preset hyperplane parameters; If the minimum value of the distance between each support vector and the hyperplane is less than a preset value, the hyperplane parameters are updated; if the minimum value of the distance between each support vector and the hyperplane is not less than the preset value, the support vector machine is set to a trained support vector machine; The user portraits to be identified are classified according to the trained support vector machine.

6. The method according to claim 5, characterized in that The method further comprises: If there is any user profile to be identified that cannot find a decision boundary in the support vector machine, mark the user profile to be identified; Acquire feedback data corresponding to the marked user portrait to be identified, obtain a scoring result based on the corresponding feedback data analysis, determine the corresponding relationship between the user portrait to be identified and the obtained scoring result, and divide the user portrait to be identified into the user group based on the corresponding relationship.

7. A cigarette user feedback analysis device, characterized in that: The device comprises: A collection module, used to collect feedback data of multiple cigarette users on each cigarette brand, as well as a user portrait of each cigarette user; the user portrait includes smoking age, smoking frequency and income ability; An analysis module, used to analyze the feedback data according to preset rules to obtain a scoring result corresponding to the cigarette brand, and mine the corresponding relationship between the user portrait and the scoring result based on an association rule algorithm; The division module is used to divide each of the cigarette users into a plurality of user groups according to the corresponding relationship, so as to determine the cigarette preference of each user group.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images