Methods, apparatus, computer equipment and storage media for cigarette taste analysis

Abstract

This application provides a method, apparatus, computer device, and storage medium for cigarette taste analysis. The method includes: when the taste analysis results of the cigarette do not meet preset taste qualification conditions, acquiring taste characteristic data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke; analyzing the correlation between each taste and each compound among the multiple compounds based on the taste characteristic data and compound information; selecting key compounds corresponding to each taste from the multiple compounds based on the taste characteristic data, compound information, and preset key compound screening conditions; and visually outputting the key compounds corresponding to each taste and the correlation between each taste and the key compounds to indicate optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions. This method improves the efficiency and accuracy of cigarette taste analysis.

Description

Technical Field

[0001] This application relates to the field of cigarette technology. Specifically, this application relates to a method, apparatus, computer device, and storage medium for analyzing the taste of cigarettes. Background Technology

[0002] With the increasing diversification of cigarette consumption demands and the ever-improving requirements for cigarette consumption experience, analyzing the differences in taste among cigarettes has become an important means for cigarette companies to enhance their product competitiveness and a key link in improving the quality of cigarette products.

[0003] In traditional techniques, the analysis of cigarette taste usually relies on manual tasting, which is a highly subjective and time-consuming process. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for cigarette taste analysis that can improve the efficiency and accuracy of cigarette taste analysis, in order to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for analyzing the taste of cigarettes. The method includes:

[0006] If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, taste feature data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke are obtained; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument.

[0007] Based on the taste characteristic data and the compound information, analyze the correlation between each taste and each compound among the multiple compounds;

[0008] Based on the taste characteristic data, the compound information, and the preset key compound screening conditions, key compounds corresponding to each taste are screened from the multiple compounds; wherein, the key compounds are compounds that play a key role in each taste.

[0009] The key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0010] In one embodiment, the step of analyzing the correlation between each taste and each of the multiple compounds based on the taste feature data and the compound information includes:

[0011] Calculate the Spearman correlation coefficient between the taste characteristic data of each of the multiple tastes and the compound information of each of the multiple compounds;

[0012] Based on the Spearman correlation coefficient, the correlation between each taste and each of the multiple compounds is determined; wherein the correlation includes positive and negative correlations.

[0013] In one embodiment, the step of selecting key compounds corresponding to each taste from the plurality of compounds based on the taste feature data, the compound information, and preset key compound screening conditions includes:

[0014] Based on the taste characteristic data of the various tastes and the compound information of the various compounds, calculate the gray correlation level between each taste and each compound among the various compounds;

[0015] Based on the taste characteristic data of the various tastes and the compound information of the various compounds, calculate the variable importance projection value between each taste and each compound in the various compounds;

[0016] For each taste, compounds whose gray relational degree level is greater than or equal to a preset level threshold and whose variable importance projection value is greater than or equal to a preset projection threshold are identified as key compounds corresponding to each taste.

[0017] In one embodiment, the taste analysis result is taste similarity; the method further includes:

[0018] Principal component analysis was performed on the taste feature data to obtain taste dimensionality reduction features;

[0019] Calculate the spatial distribution distance between the taste dimension reduction features and the pre-stored taste dimension reduction features of standard cigarettes;

[0020] The similarity of taste is determined based on the spatial distribution distance.

[0021] In one embodiment, the taste analysis result is a taste evaluation result; the method further includes:

[0022] The taste feature data is input into a pre-trained cigarette taste evaluation model to obtain the taste evaluation results of the cigarette.

[0023] In the cigarette taste evaluation model, the taste feature data is upscaled through a first dot product operation layer to obtain a first upscaled result. The first upscaled result is then processed by a first SD module to obtain a first processed result. The sum of the first upscaled result and the first processed result is processed by a second SD module to obtain a second processed result. The first upscaled result, the sum of the first processed result and the second processed result is then upscaled through a second dot product operation layer to obtain a second upscaled result. The second upscaled result is then pooled through a pooling layer to obtain a pooled result. The pooled result is then processed by a linear layer and input into a regression layer for regression prediction to obtain the cigarette taste evaluation result.

[0024] In one embodiment, in the first SD module, the first dimensionality-upgrading result is segmented by a channel segmentation module to obtain a first part of data and a second part of data. The first part of data is sequentially processed by a first group convolutional layer, a depth separation convolutional layer, an attention mechanism module, and a second group convolutional layer to obtain a third part of data. The third part of data and the second part of data are concatenated and then input into a channel shuffling module for channel shuffling to obtain a first processing result.

[0025] In one embodiment, in the attention mechanism module, the output of the depth separation convolutional layer is sequentially processed by a global pooling layer, a first linear layer, and a second linear layer to obtain a processing result. The sum of the output of the depth separation convolutional layer and the processing result is scaled to obtain the output of the attention mechanism module.

[0026] Secondly, this application provides a cigarette taste analysis device. The device includes:

[0027] The data acquisition module is used to acquire olfactory detection data of cigarettes collected by the electronic nose and gustatory detection data of cigarettes collected by the electronic tongue.

[0028] The olfactory analysis module is used to input the olfactory detection data into the olfactory neural network model in the pre-trained olfactory-gustatory synesthesia model for olfactory analysis and to obtain olfactory perception results.

[0029] The taste analysis module is used to input the taste detection data into the taste pathway calculation model in the olfactory-gustatory synesthesia model for taste analysis and to obtain the taste perception results.

[0030] The cigarette recognition module is used to input the olfactory perception results and the gustatory perception results into the orbitofrontal cortex module in the olfactory-gustatory synesthesia model for recognition, so as to obtain the cigarette recognition result.

[0031] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0032] If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, taste feature data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke are obtained; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument.

[0033] Based on the taste characteristic data and the compound information, analyze the correlation between each taste and each compound among the multiple compounds;

[0034] Based on the taste characteristic data, the compound information, and the preset key compound screening conditions, key compounds corresponding to each taste are screened from the multiple compounds; wherein, the key compounds are compounds that play a key role in each taste.

[0035] The key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0036] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0037] If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, taste feature data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke are obtained; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument.

[0038] Based on the taste characteristic data and the compound information, analyze the correlation between each taste and each compound among the multiple compounds;

[0039] Based on the taste characteristic data, the compound information, and the preset key compound screening conditions, key compounds corresponding to each taste are screened from the multiple compounds; wherein, the key compounds are compounds that play a key role in each taste.

[0040] The key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0041] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0042] If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, taste feature data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke are obtained; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument.

[0043] Based on the taste characteristic data and the compound information, analyze the correlation between each taste and each compound among the multiple compounds;

[0044] Based on the taste characteristic data, the compound information, and the preset key compound screening conditions, key compounds corresponding to each taste are screened from the multiple compounds; wherein, the key compounds are compounds that play a key role in each taste.

[0045] The key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0046] The aforementioned method, apparatus, computer equipment, computer-readable storage medium, and computer program product for cigarette taste analysis integrate taste characteristic data and compound information collected by an electronic tongue and gas chromatography-mass spectrometry, respectively. By analyzing the correlation between each taste and each compound, and combining the screening analysis results of compounds that play a key role in each taste, the method guides the optimization of cigarette formulation, thereby obtaining cigarettes that meet taste requirements. The advantage of this application is that it avoids the influence of subjective human factors on the taste analysis results, and can automatically and objectively analyze and evaluate the taste of cigarettes, which is beneficial to improving the efficiency and accuracy of cigarette taste analysis. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a method for analyzing the taste of cigarettes in one embodiment;

[0048] Figure 2 This is a structural block diagram of a cigarette taste analysis device in one embodiment;

[0049] Figure 3 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0051] In one embodiment, such as Figure 1 As shown, a method for analyzing the taste of cigarettes is provided. This embodiment illustrates the application of this method to a terminal. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc.

[0052] Understandably, this method can also be applied to servers, and to systems that include both terminals and servers, and can be implemented through the interaction between terminals and servers.

[0053] In this embodiment, the method includes the following steps:

[0054] Step S102: If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, obtain the taste characteristic data of multiple tastes of the cigarette and the compound information of multiple compounds in the cigarette smoke.

[0055] Specifically, the taste analysis results of cigarettes can include taste similarity and taste evaluation results (e.g., taste evaluation value). The taste analysis results of cigarettes may not meet preset taste qualification conditions, such as a taste similarity lower than a preset similarity threshold or a taste evaluation value lower than a preset evaluation threshold. Taste can be, but is not limited to, umami, salty, sour, bitter, and astringent. Taste characteristic data is collected through an electronic tongue, which can simulate the human tongue's perception of different chemical components, such as bitterness, sourness, and sweetness, and can assess the taste characteristics of cigarettes. Compound information is collected using gas chromatography-mass spectrometry (GC-MS). Compound information can include compound content.

[0056] Step S104: Based on taste characteristic data and compound information, analyze the correlation between each taste and each compound among multiple compounds.

[0057] Specifically, the correlation coefficient (positive or negative) between the taste characteristic data of each taste in a variety of tastes and the compound information of each compound in a variety of compounds is calculated. Based on this correlation coefficient, the correlation between each taste and each compound in the variety of compounds is determined. The correlation includes positive and negative correlations.

[0058] Step S106: Based on taste characteristic data, compound information, and preset key compound screening conditions, key compounds corresponding to each taste are screened from a variety of compounds.

[0059] Specifically, based on taste characteristic data and compound information, the contribution of each compound to each taste is calculated. For each taste, based on this contribution, compounds that meet the preset key compound screening conditions are screened from a variety of compounds. These compounds are the key compounds that play a key role in each taste and are identified as the key compounds corresponding to each taste.

[0060] Step S108: Visualize the key compounds corresponding to each taste and the correlation between each taste and the key compounds to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0061] Specifically, the key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to instruct humans to optimize the cigarette formula based on the visualized information. The taste analysis results of the optimized cigarette are then used to determine whether they meet the preset taste qualification conditions. If not, step S102 is continued; if yes, the method is terminated.

[0062] The aforementioned method for analyzing cigarette taste integrates taste characteristic data and compound information collected by electronic tongue and gas chromatography-mass spectrometry, respectively. It then analyzes the correlation between each taste and each compound, and combines this with the screening results of compounds that play a key role in each taste, to guide the optimization of cigarette formulation, thereby obtaining cigarettes that meet taste requirements. The advantage of this method is that it avoids the influence of subjective human factors on the taste analysis results, and can automatically and objectively analyze and evaluate cigarette taste, thus improving the efficiency and accuracy of cigarette taste analysis.

[0063] In one embodiment, step S104, "analyzing the correlation between each taste and each compound among multiple compounds based on taste characteristic data and compound information," is specifically implemented through the following steps:

[0064] Step S1042: Calculate the Spearman correlation coefficient (positive or negative) between the taste characteristic data of each taste in multiple tastes and the compound information of each compound in multiple compounds;

[0065] Step S1044: Determine the correlation between each taste and each compound among multiple compounds based on the Spearman correlation coefficient.

[0066] The correlations include positive correlation, negative correlation, and correlation strength. Correlation strength characterizes the degree of positive or negative contribution of different compounds to the effect of taste. Correlation strength is determined based on the Spearman correlation coefficient. A higher correlation strength indicates a stronger positive or negative correlation between the corresponding taste and the compound.

[0067] In one embodiment, step S106, "screening out key compounds corresponding to each taste from a variety of compounds based on taste feature data, compound information, and preset key compound screening conditions," is specifically implemented through the following steps:

[0068] Step S1062: Based on the taste characteristic data of multiple tastes and the compound information of multiple compounds, calculate the gray correlation level between each taste and each compound in the multiple compounds;

[0069] Step S1064: Based on the taste characteristic data of multiple tastes and the compound information of multiple compounds, calculate the variable importance projection value of each taste and each compound in multiple compounds;

[0070] Step S1066: For each taste, compounds whose gray relational degree level is greater than or equal to a preset level threshold and whose variable importance projection value is greater than or equal to a preset projection threshold are identified as key compounds corresponding to each taste.

[0071] Grey Relational Analysis (GRA) offers several advantages: It considers the mutual influence and correlation between multiple indicators. It doesn't just focus on individual indicator values, but calculates correlation coefficients to comprehensively consider their interrelationships, resulting in more comprehensive and accurate evaluation results. GRA doesn't impose requirements on the specific distribution of the data, thus maintaining reliable evaluation results even with highly variable data. It is insensitive to outliers and better adapts to data variations and noise. Furthermore, by transforming observed indicator values ​​into dimensionless correlation coefficients, it eliminates dimensional differences between indicators, allowing for comparison and comprehensive analysis. This effectively avoids significant biases in the evaluation results caused by individual indicators.

[0072] In general, grey relational analysis can calculate and rank the independent and dependent variables according to their degree of correlation, identifying the independent variable with a higher degree of correlation to the dependent variable. Based on this principle, the dependent variable is set as the taste characteristics of cigarettes. Since the electronic tongue sensor of cigarettes is related to umami, saltiness, bitterness, astringency, and sourness, the dependent variable becomes umami, saltiness, bitterness, astringency, and sourness. The independent variable is set as the content of different compounds in cigarettes, thus constructing the correlation between umami, saltiness, bitterness, astringency, and sourness in cigarettes and the content of different compounds in cigarettes. Therefore, the grey relational analysis method can be used to obtain the grey relational degree level (GRA level) between umami, saltiness, bitterness, astringency, and sourness in cigarettes and the content of different compounds in cigarettes. The higher the GRA level, the greater the influence of the compound on the corresponding taste.

[0073] On the other hand, in Orthogonal Partial Least Squares-Discriminant Analysis (OPLS-DA), the Variable Importance in Projection (VIP) value is an indicator used to assess the importance of a variable in distinguishing between categories. A VIP value greater than 1 indicates a variable with a significant impact and can be considered as a priority feature. A higher VIP value indicates that the variable contributes significantly to distinguishing samples from different categories, while a lower VIP value indicates that the variable contributes less. It measures the degree of contribution of each variable to the model; a higher value indicates a greater contribution of the variable to the classification of samples. Since OPLS-DA is a classification method that focuses more on differences between cigarette categories, a VIP value greater than 1 can be used to determine whether a compound is a compound that causes significant differences between cigarette categories.

[0074] For each taste, compounds with a gray relational degree level greater than or equal to a preset level threshold (e.g., 0.6) and a variable importance projection value greater than or equal to a preset projection threshold (e.g., 1) are identified as key compounds that affect the specific taste between different categories of cigarettes.

[0075] In one embodiment, the taste analysis result is a taste similarity; based on this, the method further includes the following steps:

[0076] Step S112: Perform principal component analysis on the taste feature data to obtain taste dimensionality reduction features;

[0077] Step S114: Calculate the spatial distribution distance between the taste dimension reduction features and the pre-stored taste dimension reduction features of standard cigarettes;

[0078] Step S116: Determine the taste similarity based on the spatial distribution distance.

[0079] Specifically, firstly, for the taste feature data samples, based on the principle that the principal components after dimensionality reduction using Principal Component Analysis (PCA) are linear combinations of the original features, the resulting 2D feature space can be transformed into a cigarette taste feature space. Both coordinate axes compress a portion of the original cigarette taste features. Therefore, the position of the dimensionality-reduced cigarette taste features in the cigarette taste feature space can characterize the overall taste features in the cigarette taste feature data. Thus, based on the relationship between the position of the dimensionality-reduced taste features in the cigarette taste feature space and cigarette taste, the similarity between different cigarette tastes can be obtained by considering the spatial distribution distance. Since the spatial distribution distance between cigarettes and standard cigarettes is inversely proportional to the cigarette taste similarity, the degree of similarity between cigarettes and standard cigarettes can be obtained.

[0080] Furthermore, LoA analysis was performed on the principal component results (taste dimensionality reduction features). Each principal component has a set of weights with the original variables, called loadings. Loadings represent the degree of contribution of the original taste features (taste feature data) to the two principal components after dimensionality reduction. The loading matrix shows the relationship between the original taste features and the principal components. By observing the loadings on each principal component, it is possible to determine which original variables contribute significantly to that principal component. Therefore, the meaning of the principal components and the interpretation of the results can be better understood. LoA analysis can identify which original taste features contribute significantly to a principal component formed by a linear combination of different original taste features after dimensionality reduction. Therefore, LoA can be used to obtain the cigarette taste that causes differences between different cigarette categories.

[0081] In one embodiment, the taste analysis result is a taste evaluation result. Based on this, the method further includes the following steps:

[0082] Step S122: Input the taste feature data into the pre-trained cigarette taste evaluation model to obtain the taste evaluation results of the cigarette.

[0083] In the cigarette taste evaluation model, the taste feature data is upscaled through a first dot product operation layer to obtain a first upscaled result. The first upscaled result is then processed by a first SD module to obtain a first processed result. The sum of the first upscaled result and the first processed result is processed by a second SD module to obtain a second processed result. The first upscaled result, the sum of the first processed result and the second processed result is then upscaled through a second dot product operation layer to obtain a second upscaled result. The second upscaled result is then pooled through a pooling layer to obtain a pooled result. The pooled result is then processed by a linear layer and input into a regression layer for regression prediction to obtain the cigarette taste evaluation result.

[0084] In one embodiment, in the first SD module, the first dimensionality increase result is segmented by the channel segmentation module to obtain a first part of data and a second part of data. The first part of data is sequentially processed by a grouped convolutional layer, a depth separation convolutional layer, an attention mechanism module, and a second grouped convolutional layer to obtain a third part of data. The third part of data and the second part of data are concatenated and then input into the channel shuffling module for channel shuffling to obtain the first processing result.

[0085] In one embodiment, in the attention mechanism module, the output of the depth separation convolutional layer is sequentially processed by a global pooling layer, a first linear layer, and a second linear layer to obtain the processing result. The sum of the output of the depth separation convolutional layer and the processing result is scaled to obtain the output of the attention mechanism module.

[0086] Specifically, the taste feature data is input into the cigarette taste evaluation model. First, the data undergoes 1x1 dot product operations in the first dot product layer to increase its dimensionality. The result of this dimensionality increase is then used as the input to the first SD module. The output of the first SD module is added to the output of the first dot product layer to serve as the input to the second SD module. The output of the second SD module, combined with the outputs of the first and first dot product layers, becomes the input to the second dot product layer. The output of the second dot product layer is then used as the input to the average pooling layer. After the average pooling layer output is expanded, a linear layer is obtained, which is then input to the regression layer to obtain the regression result, i.e., the taste evaluation result.

[0087] The SD module input first undergoes channel segmentation. The right half (first part of the data) after segmentation is subjected to group convolution, depthwise splitting convolution, and attention mechanism. The output of the SE module is then concatenated with the left half (second part of the data) before segmentation, and channel shuffling is performed to obtain the output of the SD module.

[0088] The structural parameters of each component in the cigarette taste evaluation model are shown in the table below:

[0089]

[0090]

[0091] Table 1 Detailed parameters of the cigarette taste evaluation model

[0092]

[0093] Table 2 Detailed Parameters of SD Module Structure

[0094]

[0095]

[0096] Table 3. Detailed Parameters of the SE Module Structure

[0097] In one embodiment, the cigarette taste evaluation model can be a Long Short-Term Memory (LSTM) network model. LSTM has advantages in regression analysis, such as better modeling nonlinear relationships, handling time series data, handling long-term dependencies, handling asynchronous inputs, and handling missing data. These characteristics make LSTM a powerful modeling tool for regression problems, which can improve the predictive accuracy and generalization ability of regression models. Therefore, an LSTM network was chosen to construct the cigarette taste evaluation model.

[0098] In one embodiment, the cigarette taste evaluation model can be a Gaussian Process Regression and Particle Swarm Optimization (GPR-PSO) model. PSO-GPR combines the particle swarm optimization algorithm and the Gaussian process regression model. The particle swarm optimization algorithm has global search capabilities, enabling it to find the global optimum in the search space. PSO iteratively updates the position and velocity of particles to effectively explore the solution space. Combining PSO with Gaussian process regression fully utilizes the global optimization characteristics of the PSO algorithm to find a more accurate regression model. The PSO-GPR algorithm can adaptively adjust the parameters of the Gaussian process regression model. Through PSO optimization, the optimal hyperparameter settings can be automatically searched, resulting in a better regression model. This adaptability improves the fitting ability and prediction accuracy of the regression model. The Gaussian process regression model can provide an estimate of the uncertainty of the prediction results. The PSO-GPR algorithm, through the Bayesian framework of Gaussian process regression, combined with prior information and training samples, can better handle small-sample regression problems and provide reasonable prediction results.

[0099] In summary, the PSO-GPR algorithm offers advantages in regression analysis, including global optimization, parameter adaptation, uncertainty estimation, handling of nonlinear relationships, and effective processing of small samples. These characteristics make PSO-GPR a powerful tool that can improve the accuracy, generalization ability, and robustness of regression models. Therefore, PSO-GPR was chosen to construct the cigarette taste evaluation model.

[0100] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0101] Based on the same inventive concept, this application also provides a cigarette taste analysis device for implementing the cigarette taste analysis method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more cigarette taste analysis device embodiments provided below can be found in the limitations of the cigarette taste analysis method described above, and will not be repeated here.

[0102] In one embodiment, such as Figure 2 As shown, a cigarette taste analysis device is provided, comprising:

[0103] The data acquisition module 202 is used to acquire taste feature data of multiple tastes of cigarettes and compound information of multiple compounds in cigarette smoke when the taste analysis results of cigarettes do not meet the preset taste qualification conditions; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument.

[0104] The relationship analysis module 204 is used to analyze the correlation between each taste and each compound among multiple compounds based on taste characteristic data and compound information;

[0105] The compound screening module 206 is used to screen key compounds corresponding to each taste from a variety of compounds based on taste characteristic data, compound information and preset key compound screening conditions; wherein, key compounds are compounds that play a key role in each taste.

[0106] The visualization output module 208 is used to visualize the key compounds corresponding to each taste and the correlation between each taste and the key compounds, so as to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions.

[0107] The aforementioned cigarette taste analysis device integrates taste characteristic data and compound information collected by an electronic tongue and gas chromatography-mass spectrometry (GC-MS). It then analyzes the correlation between each taste and each compound, and combines this with the screening results of compounds that play a key role in each taste, to guide the optimization of cigarette formulation, thereby obtaining cigarettes that meet taste requirements. The advantage of this device is that it avoids the influence of subjective human factors on the taste analysis results, and can automatically and objectively analyze and evaluate the taste of cigarettes, thus improving the efficiency and accuracy of cigarette taste analysis.

[0108] In one embodiment, the relationship analysis module 204 is specifically used to calculate the Spearman correlation coefficient between the taste feature data of each taste in a variety of tastes and the compound information of each compound in a variety of compounds; and to determine the correlation between each taste and each compound in a variety of compounds based on the Spearman correlation coefficient; wherein the correlation includes positive correlation and negative correlation.

[0109] In one embodiment, the compound screening module 206 is specifically used to calculate the gray relational degree level between each taste and each compound in the multiple compounds based on the taste feature data of multiple tastes and the compound information of multiple compounds; calculate the variable importance projection value between each taste and each compound in the multiple compounds based on the taste feature data of multiple tastes and the compound information of multiple compounds; and for each taste, identify compounds whose gray relational degree level is greater than or equal to a preset level threshold and whose variable importance projection value is greater than or equal to a preset projection threshold as key compounds corresponding to each taste.

[0110] It should be noted that the cigarette taste analysis device provided in the above embodiments is only illustrated by the division of the above functional modules when realizing the corresponding functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the cigarette taste analysis device and the cigarette taste analysis method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0111] According to one aspect of this application, embodiments of the present invention also provide a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component. When the computer program is executed by a processor, the cigarette taste analysis method provided in embodiments of this application is performed.

[0112] Furthermore, embodiments of the present invention also provide a computer device, which includes a processor and a memory. The memory stores a computer program, and the processor is capable of executing the computer program stored in the memory. When the computer program is executed by the processor, it can implement the cigarette taste analysis method provided in any of the above embodiments.

[0113] For example, Figure 3 An embodiment of the present invention provides a computer device, which includes a bus 1110, a processor 1120, a transceiver 1130, a bus interface 1140, a memory 1150, and a user interface 1160.

[0114] In this embodiment of the invention, the device further includes a computer program stored in a memory 1150 and executable on a processor 1120, which, when executed by the processor 1120, implements the various processes of the above-described cigarette taste analysis method embodiment.

[0115] Transceiver 1130 is used to receive and send data under the control of processor 1120.

[0116] In this embodiment of the invention, a bus architecture (represented by bus 1110) is used. Bus 1110 may include any number of interconnected buses and bridges. Bus 1110 connects various circuits, including one or more processors represented by processor 1120 and memory represented by memory 1150.

[0117] Bus 1110 represents one or more of several types of bus architectures, including memory buses and memory controllers, peripheral buses, Accelerated Graphics Port (AGP), processors, or local buses using any bus architecture from various bus architectures. As an example and not a limitation, such architectures include: Industry Standard Architecture (ISA) buses, Micro Channel Architecture (MCA) buses, Enhanced ISA (EISA) buses, Video Electronics Standards Association (VESA) buses, and Peripheral Component Interconnect (PCI) buses.

[0118] The processor 1120 can be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor hardware or by instructions in software form. The processors mentioned above include: general-purpose processors, central processing units (CPUs), network processors (NPs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs), programmable logic arrays (PLAs), microcontroller units (MCUs) or other programmable logic devices, discrete gates, transistor logic devices, and discrete hardware components. They can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. For example, the processor can be a single-core processor or a multi-core processor, and the processor can be integrated on a single chip or located on multiple different chips.

[0119] Processor 1120 can be a microprocessor or any conventional processor. The method steps disclosed in the embodiments of the present invention can be directly executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in readable storage media known in the art, such as Random Access Memory (RAM), Flash Memory, Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), registers, etc. The readable storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0120] Bus 1110 can also connect various other circuits, such as peripheral devices, voltage regulators, or power management circuits. Bus interface 1140 provides an interface between bus 1110 and transceiver 1130, all of which are well known in the art. Therefore, embodiments of the present invention will not be described further.

[0121] Transceiver 1130 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. For example, transceiver 1130 receives external data from other devices, and transceiver 1130 is used to send data processed by processor 1120 to other devices. Depending on the nature of the computer system, a user interface 1160 may also be provided, such as a touchscreen, physical keyboard, monitor, mouse, speaker, microphone, trackball, joystick, or stylus.

[0122] It should be understood that, in embodiments of the present invention, memory 1150 may further include memory remotely configured relative to processor 1120, and such remotely configured memory can be connected to a server via a network. One or more portions of the aforementioned network may be an ad hoc network, intranet, extranet, virtual private network (VPN), local area network (LAN), wireless local area network (WLAN), wide area network (WAN), wireless wide area network (WWAN), metropolitan area network (MAN), Internet, public switched telephone network (PSTN), ordinary old-style telephone service (POTS), cellular telephone network, wireless network, Wi-Fi network, and combinations of two or more of the aforementioned networks. For example, cellular telephone networks and wireless networks can be Global System for Mobile Communications (GSM), Code Division Multiple Access (CDMA), WiMAX, General Packet Radio Service (GPRS), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), LTE Frequency Division Duplex (FDD), LTE Time Division Duplex (TDD), Advanced Long Term Evolution (LTE-A), Universal Mobile Telecommunications System (UMTS), Enhanced Mobile Broadband (eMBB), Massive Machine Type Communication (mMTC), Ultra Reliable Low Latency Communications (uRLLC), etc.

[0123] It should be understood that the memory 1150 in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile memory and non-volatile memory. Non-volatile memory includes: read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0124] Volatile memory includes random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 1150 described in this embodiment includes, but is not limited to, the above and any other suitable types of memory.

[0125] In this embodiment of the invention, the memory 1150 stores the following elements of the operating system 1151 and the application 1152: executable modules, data structures, or subsets thereof, or extended sets thereof.

[0126] Specifically, the operating system 1151 includes various system programs, such as a framework layer, a core library layer, and a driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 1152 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this embodiment of the invention can be included in the application program 1152. The application program 1152 includes applets, objects, components, logic, data structures, and other computer system executable instructions that perform specific tasks or implement specific abstract data types.

[0127] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the various processes of the above-described cigarette taste analysis method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0128] Computer-readable storage media include: permanent and non-permanent, removable and non-removable media, which are tangible devices capable of retaining and storing instructions for use by an instruction execution device. Computer-readable storage media include: electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, and any suitable combination thereof. Computer-readable storage media include: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, optical disc read-only memory (CD-ROM), digital versatile optical disc (DVD) or other optical storage, magnetic tape storage, magnetic disk storage or other magnetic storage devices, memory sticks, mechanical encoding devices (e.g., punched cards or raised structures in grooves on which instructions are recorded), or any other non-transfer medium that can be used to store information accessible by a computing device. As defined in the embodiments of the present invention, computer-readable storage media do not include temporary signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0129] In the several embodiments provided in this application, it should be understood that the disclosed apparatus, devices, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, or it may be an electrical, mechanical, or other form of connection.

[0130] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to solve the problems addressed by the embodiments of the present invention, depending on actual needs.

[0131] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0132] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (including: a personal computer, a server, a data center, or other network device) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media listed above that can store program code.

[0133] In the description of the embodiments of the present invention, those skilled in the art should understand that the embodiments of the present invention can be implemented as methods, apparatuses, devices, and storage media. Therefore, the embodiments of the present invention can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software. Furthermore, in some embodiments, the embodiments of the present invention can also be implemented as a computer program product in one or more computer-readable storage media, the computer-readable storage media containing computer program code.

[0134] The aforementioned computer-readable storage medium may be any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, optical disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any combination thereof. In embodiments of the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0135] The computer program code contained in the aforementioned computer-readable storage medium may be transmitted using any suitable medium, including wireless, wire, optical fiber, radio frequency (RF), or any suitable combination thereof.

[0136] Computer program code for performing the operations of the embodiments of the present invention can be written in assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or in one or more programming languages ​​or combinations thereof. The programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The computer program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer or an external computer via any type of network, including a local area network (LAN) or a wide area network (WAN).

[0137] The embodiments of the present invention describe the provided methods, apparatus, and devices through flowcharts and / or block diagrams.

[0138] It should be understood that each block of a flowchart and / or block diagram, as well as combinations of blocks in a flowchart and / or block diagram, can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine that, when executed by a computer or other programmable data processing apparatus, creates means for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0139] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to function in a particular manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction apparatus product that includes the functions / operations specified in the blocks of a flowchart and / or block diagram.

[0140] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus or other device to produce a computer-implemented process, such that the instructions that execute on the computer or other programmable data processing apparatus provide a process for implementing the functions / operations specified in the blocks of the flowchart and / or block diagram.

[0141] The above description is merely a specific implementation of the embodiments of the present invention, but the protection scope of the embodiments of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of the present invention should be included within the protection scope of the embodiments of the present invention. Therefore, the protection scope of the embodiments of the present invention should be determined by the protection scope of the claims.

Claims

1. A method for analyzing the taste of cigarettes, characterized in that, include: If the taste analysis results of the cigarette do not meet the preset taste qualification conditions, taste characteristic data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke are acquired; wherein, the taste characteristic data is collected by an electronic tongue, and the compound information is collected by gas chromatography-mass spectrometry; based on the taste characteristic data and the compound information, the correlation between each taste and each compound among the multiple compounds is analyzed; the analysis of the correlation between each taste and each compound among the multiple compounds includes: calculating the Spearman correlation coefficient between the taste characteristic data of each taste among the multiple tastes and the compound information of each compound among the multiple compounds; determining the correlation between each taste and each compound among the multiple compounds based on the Spearman correlation coefficient; wherein, the correlation includes positive correlation and negative correlation; Based on the taste feature data, the compound information, and preset key compound screening conditions, key compounds corresponding to each taste are screened from the multiple compounds; wherein, the key compounds are compounds that play a key role in each taste; the process of screening key compounds corresponding to each taste from the multiple compounds includes: calculating the gray correlation degree level between each taste and each compound in the multiple compounds based on the taste feature data of the multiple tastes and the compound information of the multiple compounds; calculating the variable importance projection value between each taste and each compound in the multiple compounds based on the taste feature data of the multiple tastes and the compound information of the multiple compounds; for each taste, compounds whose gray correlation degree level is greater than or equal to a preset level threshold and whose variable importance projection value is greater than or equal to a preset projection threshold are identified as key compounds corresponding to each taste; The key compounds corresponding to each taste and the correlation between each taste and the key compounds are visualized and output to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions; wherein, if the taste analysis results of the optimized cigarette do not meet the preset taste qualification conditions, the process returns to the step of obtaining the taste feature data of multiple tastes of the cigarette and the compound information of multiple compounds of the cigarette smoke; The taste analysis results include taste similarity or taste evaluation results. When the taste analysis results are taste evaluation results, the method further includes: inputting the taste feature data into a pre-trained cigarette taste evaluation model to obtain the cigarette taste evaluation results. In the cigarette taste evaluation model, the taste feature data is upscaled through a first dot product operation layer to obtain a first upscaled result. The first upscaled result is processed through a first SD module to obtain a first processed result. The sum of the first upscaled result and the first processed result is processed through a second SD module to obtain a second processed result. The first upscaled result, the sum of the first processed result and the second processed result is upscaled through a second dot product operation layer to obtain a second upscaled result. The second upscaled result is pooled through a pooling layer to obtain a pooling result. The pooling result is processed through a linear layer and then input into a regression layer for regression prediction to obtain the cigarette taste evaluation results.

2. The method according to claim 1, characterized in that, The taste analysis result is a taste similarity; the method further includes: Principal component analysis was performed on the taste feature data to obtain taste dimensionality reduction features; Calculate the spatial distribution distance between the taste dimension reduction features and the pre-stored taste dimension reduction features of standard cigarettes; The similarity of taste is determined based on the spatial distribution distance.

3. The method according to claim 1, characterized in that, In the first SD module, the first dimensionality increase result is segmented by the channel segmentation module to obtain a first part of data and a second part of data. The first part of data is sequentially processed by the first group convolutional layer, the depth separation convolutional layer, the attention mechanism module, and the second group convolutional layer to obtain a third part of data. The third part of data and the second part of data are concatenated and then input into the channel shuffling module for channel shuffling to obtain the first processing result.

4. The method according to claim 3, characterized in that, In the attention mechanism module, the output of the depth separation convolutional layer is sequentially processed by a global pooling layer, a first linear layer, and a second linear layer to obtain the processing result. The sum of the output of the depth separation convolutional layer and the processing result is scaled to obtain the output of the attention mechanism module.

5. A cigarette taste analysis device, characterized in that, include: The data acquisition module is used to acquire taste feature data of multiple tastes of the cigarette and compound information of multiple compounds in the cigarette smoke when the taste analysis results of the cigarette do not meet the preset taste qualification conditions; wherein, the taste feature data is collected by an electronic tongue and the compound information is collected by a gas chromatography-mass spectrometry instrument. The relationship analysis module is used to analyze the correlation between each taste and each compound among the multiple compounds based on the taste feature data and the compound information; The relationship analysis module is specifically used to calculate the Spearman correlation coefficient between the taste feature data of each taste among the multiple tastes and the compound information of each compound among the multiple compounds; and to determine the correlation between each taste and each compound among the multiple compounds based on the Spearman correlation coefficient; wherein the correlation includes positive correlation and negative correlation. The compound screening module is used to screen out key compounds corresponding to each taste from the multiple compounds based on the taste feature data, the compound information, and preset key compound screening conditions; wherein, the key compound is a compound that plays a key role in each taste. The compound screening module is specifically used to calculate the gray correlation level between each taste and each compound in the multiple compounds based on the taste feature data of the multiple tastes and the compound information of the multiple compounds; calculate the variable importance projection value between each taste and each compound in the multiple compounds based on the taste feature data of the multiple tastes and the compound information of the multiple compounds; and for each taste, identify compounds whose gray correlation level is greater than or equal to a preset level threshold and whose variable importance projection value is greater than or equal to a preset projection threshold as key compounds corresponding to each taste. The visualization output module is used to visualize the key compounds corresponding to each taste and the correlation between each taste and the key compounds, so as to indicate the optimization of the cigarette formula until the taste analysis results of the optimized cigarette meet the preset taste qualification conditions. The judgment module is used to return to the step of obtaining the taste feature data of multiple tastes of the cigarette and the compound information of multiple compounds of the cigarette smoke if the taste analysis result of the optimized cigarette does not meet the preset taste qualification conditions. The taste analysis results include taste similarity or taste evaluation results; when the taste analysis results are taste evaluation results, the device further includes: The taste evaluation module is used to input the taste feature data into a pre-trained cigarette taste evaluation model to obtain the taste evaluation result of the cigarette. Specifically, in the cigarette taste evaluation model, the taste feature data is upscaled through a first dot product operation layer to obtain a first upscaled result. This first upscaled result is then processed by a first SD module to obtain a first processed result. The sum of the first upscaled result and the first processed result is processed by a second SD module to obtain a second processed result. The first upscaled result, the sum of the first processed result and the second processed result are then upscaled through a second dot product operation layer to obtain a second upscaled result. The second upscaled result is then pooled through a pooling layer to obtain a pooled result. Finally, the pooled result is processed by a linear layer and input into a regression layer for regression prediction to obtain the taste evaluation result of the cigarette.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images