Odor feature interaction analysis method and application

Abstract

The invention relates to the technical field of smell analysis, and discloses a smell characteristic interaction analysis method and application, a smell sensor is used to collect smell data of a target smell, the smell data is converted into a smell display map, experience data of a user for the target smell is acquired and bound with the smell display map, and analysis data is formed; analyzing the mapping relation between the user experience data and the smell display atlas, generating experience mapping features, obtaining mapping feature distribution of the user in combination with analysis, performing simulation analysis on the smell display atlas of any smell according to the mapping feature distribution of the user, and generating a simulation result of a user experience effect; the association between the smell data and the subjective experience of the user is accurately described, personalized smell experience feedback is provided, the perception ability of the user for different smells is comprehensively known, simulation analysis is performed by using a network platform, and efficient experience effect simulation is realized. The problem that in the prior art, the feeling effect of an individual on the smell cannot be accurately fed back only through smell data is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of odor analysis, and in particular to an odor feature interactive analysis method and application. Background Art

[0002] With the development of odor sensing technology, odor sensors have been able to accurately detect a variety of odor molecules and convert them into usable odor data. However, different users may have completely different experience reactions to the same odor. This is due to the influence of various factors such as individual olfactory perception, past experience, cultural background, etc. Odor data alone cannot accurately reflect an individual's perception of odor. As a result, when purchasing odor-related products such as perfumes on online platforms, customers cannot accurately know the degree of suitability of the smell of these products for themselves. Summary of the invention

[0003] The purpose of the present invention is to provide an odor feature interactive analysis method and application, aiming to solve the problem in the prior art that the individual's perception effect of the odor cannot be accurately fed back only through odor data.

[0004] The present invention is implemented in this way. In a first aspect, the present invention provides an odor feature interactive analysis method, comprising: The odor data of the target odor is collected by an odor sensor to obtain odor data corresponding to the target odor, and a graph conversion process is performed according to the odor data to generate an odor display graph for graphical expression; Acquire user experience data of the target odor from the user, and bind the user experience data with the odor display map to obtain a set of analysis data, and summarize the groups of analysis data belonging to the same user into a user data set; According to each group of the analysis data in the user data set, the mapping relationship between the user experience data and the odor display map is analyzed and processed respectively to obtain the experience mapping features corresponding to each group of the analysis data, and the experience mapping features corresponding to each group of the analysis data are combined and analyzed to obtain the mapping feature distribution of the user; wherein the experience mapping feature is used to describe the correlation relationship between the odor data and the user experience, and the mapping feature distribution is used to describe the odor experience ability of the user itself; Upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulation experience effect of the simulation object; wherein the simulation experience effect is a simulation of the user's experience of the simulation object.

[0005] In a second aspect, the present invention provides an application of an odor feature interactive analysis method, which is used to implement an odor feature interactive analysis method as described in any one of the first aspects, including: An odor collection module is used to collect odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and perform a graph conversion process based on the odor data to generate an odor display graph for graphical expression; An odor processing module is used to obtain the user's user experience data of the target odor, and bind the user experience data with the odor display map to obtain a set of analysis data, and summarize the groups of analysis data belonging to the same user into a user data set; A mapping analysis module, used to analyze the mapping relationship between the user experience data and the odor display map according to each group of analysis data in the user data set, so as to obtain experience mapping features corresponding to each group of analysis data, and combine and analyze the experience mapping features corresponding to each group of analysis data to obtain the user's mapping feature distribution; wherein the experience mapping feature is used to describe the correlation between the odor data and the user experience, and the mapping feature distribution is used to describe the user's own odor experience ability; The experience simulation module is used to upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulated experience effect of the simulation object; wherein the simulated experience effect is a simulation of the user's experience of the simulation object.

[0006] The present invention provides an odor feature interactive analysis method, which has the following beneficial effects: The present invention uses an odor sensor to collect odor data of a target odor and converts it into an odor display map, obtains user experience data of the target odor and binds it to the odor display map to form analysis data, analyzes the mapping relationship between the user experience data and the odor display map, generates experience mapping features, and obtains the user's mapping feature distribution in combination with the analysis, performs simulation analysis on the odor display map of any odor according to the user's mapping feature distribution, generates simulation results of user experience effects, accurately depicts the relationship between odor data and user subjective experience, provides personalized odor experience feedback, comprehensively understands the user's perception ability of different odors, uses a network platform to perform simulation analysis, and realizes efficient experience effect simulation, thereby solving the problem in the prior art that the individual's perception effect of odors cannot be accurately fed back only through odor data. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1It is a schematic diagram of the steps of an odor feature interactive analysis method provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of an application of an odor feature interactive analysis method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0008] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention 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 invention and are not intended to limit the present invention.

[0009] The implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0010] Reference Figure 1 , Figure 2 As shown, a preferred embodiment of the present invention is provided.

[0011] In a first aspect, the present invention provides an odor feature interactive analysis method, comprising: S1: collecting odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and performing a graph conversion process according to the odor data to generate an odor display graph for graphical expression; S2: obtaining the user's user experience data of the target odor, and binding the user experience data with the odor display map to obtain a set of analysis data, and summarizing the groups of analysis data belonging to the same user into one user data set; S3: performing parsing processing on the mapping relationship between the user experience data and the odor display map according to each group of the analysis data in the user data set, so as to obtain experience mapping features corresponding to each group of the analysis data, and combining and analyzing the experience mapping features corresponding to each group of the analysis data, so as to obtain the mapping feature distribution of the user; wherein the experience mapping feature is used to describe the correlation relationship between the odor data and the user experience, and the mapping feature distribution is used to describe the odor experience ability of the user itself; S4: Upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulation experience effect of the simulation object; wherein the simulation experience effect is a simulation of the user's experience of the simulation object.

[0012] Specifically, in step S1 of the embodiment provided by the present invention, an odor sensor, such as an electronic nose or a gas sensor, is prepared, which can detect and record the chemical composition and concentration of the odor, collect target odor samples under a controlled environment, ensure that the environment is stable to avoid external interference, and use the odor sensor to record sensor data of the target odor, including parameters such as gas concentration and chemical composition.

[0013] More specifically, there may be multiple odor sensors for odor collection, and they may have cross-selectivity with each other. Through the different sensitive areas of each odor sensor, the target odor may be accurately collected in multiple dimensions.

[0014] More specifically, key features are extracted from the odor data, such as main chemical components, concentrations, etc. This may involve feature selection and dimensionality reduction techniques, and data visualization techniques are used to generate odor display maps. The maps can be two-dimensional or three-dimensional graphics, showing the various characteristics of the odor and their distribution. If it is a two-dimensional map, the feature data can be mapped to a two-dimensional plane, such as using principal component analysis (PCA) to reduce high-dimensional data to two dimensions. If it is a three-dimensional map, a three-dimensional scatter plot or other three-dimensional visualization techniques are used to display the multi-dimensional characteristics of the odor.

[0015] It can be understood that through sensor data collection and processing, the chemical composition and concentration of the target odor can be described in detail. The generated odor display map can accurately display the distribution of odor characteristics and help understand the composition and intensity of odors. Through graphical expression, complex odor data can be presented in an intuitive form, which is easy to analyze and interpret. Users or researchers can quickly understand the characteristics of the target odor through the map, improve the usability and readability of the data, and the generated odor display map provides basic data support for subsequent user experience analysis, odor matching and simulation.

[0016] Specifically, in step S2 of the embodiment provided by the present invention, the odor sensor and the touch screen are integrated into one device. After the odor sensor collects odor data, a questionnaire or survey form designed to collect user experience feedback on the target odor is displayed on the touch screen to obtain the user's feelings and experience of the target odor. The questions include odor judgment, intensity perception, degree of liking, emotional response, etc. of the odor.

[0017] More specifically, in addition to the above solutions, a designated client or network service platform may be used for users to interact with each other and submit user experience data.

[0018] More specifically, questionnaires are distributed to users or surveys are conducted to collect users' subjective experience data on the target odor. Further analysis can be performed to obtain the correlation between the user experience and the target odor, thereby analyzing the user's olfactory function status and preference for odors, and then based on the user's olfactory function status and preference for odors, the user's perception of various odors can be simulated.

[0019] More specifically, the user's experience feedback is bound to the corresponding odor display map, usually by matching the user experience data with the odor features in the odor map, and the bound data is combined into an analysis data set. Each analysis data contains an odor display map and the corresponding user's experience feedback on the odor. All analysis data of the same user are merged into one user data set. Ensure the integrity and consistency of the data, organize all analysis data in the user data set, and facilitate subsequent analysis and processing. It can be classified and summarized by time, odor type, or other characteristics of the user.

[0020] It is understandable that by binding the user's subjective experience with a specific odor display map, the user's experience of different smells can be recorded and analyzed more accurately. The experience data of each user is combined with the data of the odor map to provide a detailed user feedback record for each smell, which is summarized in a user data set to facilitate personalized analysis, such as identifying user preferences and reaction patterns to different smells. By analyzing the user data set, insights into the user's behavioral and emotional reactions to smells can be gained, helping to improve product design and user experience strategies. Binding the user experience data with the odor display map and integrating them into a user data set enhances the availability of the data, facilitates subsequent analysis and application, and allows for multi-dimensional analysis, such as exploring the relationship between odor characteristics and user perception, or the different reactions of different user groups to the same smell.

[0021] Specifically, in step S3 of the embodiment provided by the present invention, the odor display map features of each group of analysis data and the corresponding user experience data are extracted from the user data set, and a model is established to parse the mapping relationship between the user experience data and the odor display map. Regression analysis, machine learning algorithms (such as support vector machines, random forests) or deep learning models (such as neural networks) can be used.

[0022] More specifically, experience mapping features are extracted from the established model, which describe the association between odor data and user experience. The experience mapping features are optimized and adjusted according to the analysis requirements to improve the accuracy and reliability of the mapping. The experience mapping features in each group of analysis data are integrated to form comprehensive analysis data. Statistical analysis, data mining techniques or machine learning algorithms are applied to analyze all experience mapping features, identify the distribution of users' mapping features, use statistical methods (such as principal component analysis PCA) to extract and summarize the distribution of users' mapping features, and use data visualization techniques (such as heat maps, scatter plots) to display the distribution of users' mapping features, so as to help understand users' odor experience capabilities.

[0023] More specifically, further, based on the results of the combined analysis, a user's mapping feature distribution map can be generated, and an analysis report on the user's odor experience ability can be written to describe the user's perception ability and preference for odors.

[0024] It is understandable that by analyzing the mapping relationship between user experience data and the odor display map, we can gain an in-depth understanding of how odor characteristics affect user experience. The generated experience mapping features can accurately describe the relationship between odor data and user experience, providing a basis for further analysis. The mapping feature distribution shows the user's experience ability and reaction pattern to different odors, helps understand the user's odor perception ability, and identifies the user's odor experience pattern, which can provide personalized suggestions and improvement plans for product design, marketing and other fields.

[0025] Specifically, in step S4 of the embodiment provided by the present invention, a suitable network service platform is selected, such as a data storage and visualization platform, which supports uploading and managing odor display maps, and all generated odor display maps are uploaded to the designated network service platform to ensure the integrity and accuracy of the map data, and the odor display maps are managed on the platform to ensure convenient retrieval and selection.

[0026] More specifically, by browsing and selecting odor display maps on the network service platform, users can select any map as a simulation object according to their needs, confirm the selected map as the simulation object, and prepare for simulation analysis and processing, and extract relevant data from the user's mapping feature distribution. The user's mapping feature distribution describes their ability to experience different smells.

[0027] More specifically, a simulation model is established to apply the user's mapping features to the selected simulation object. The simulation model can be adjusted and optimized based on the previous mapping relationship model, and the model parameters can be adjusted according to the characteristics of the simulation object to more accurately reflect the user experience. The simulation model is used to predict the user's experience effect on the simulation object, including the perception state of the smell, the perception intensity, the preference tendency, the emotional response, etc.

[0028] More specifically, the output results of the simulation model are analyzed to generate the user's simulated experience effect on the simulation object, and the simulated experience effect is displayed on the network service platform. The results can be presented using charts, reports or other visual methods. In addition, a feedback mechanism for the simulation experience results can be provided to users to collect their evaluation and opinions on the simulation effects.

[0029] It can be understood that by analyzing the simulation object according to the user's mapping feature distribution, it is possible to accurately predict the user's experience effects on different smells, provide personalized experience simulation, help adjust and optimize product design or services, and make improvements based on the user's simulated experience effects to meet the user's actual needs. Before the product is released or the service is implemented, the user's feedback can be predicted through the simulation experience effects to reduce uncertainty and risk. Through simulation analysis, the smell product or service can be optimized to improve user satisfaction and experience quality.

[0030] The present invention provides an odor feature interactive analysis method, which has the following beneficial effects: The present invention uses an odor sensor to collect odor data of a target odor and converts it into an odor display map, obtains user experience data of the target odor and binds it to the odor display map to form analysis data, analyzes the mapping relationship between the user experience data and the odor display map, generates experience mapping features, and obtains the user's mapping feature distribution in combination with the analysis, performs simulation analysis on the odor display map of any odor according to the user's mapping feature distribution, generates simulation results of user experience effects, accurately depicts the relationship between odor data and user subjective experience, provides personalized odor experience feedback, comprehensively understands the user's perception ability of different odors, uses a network platform to perform simulation analysis, and realizes efficient experience effect simulation, thereby solving the problem in the prior art that the individual's perception effect of odors cannot be accurately fed back only through odor data.

[0031] Preferably, the steps of collecting odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and performing a graph conversion process according to the odor data to generate an odor display graph for graph expression include: S11: continuously collecting odor signals of the target odor through the odor sensor, and summarizing the collected odor signals into a plurality of sequentially arranged unit data according to the time corresponding to the odor signal, and the unit data sequentially arranged in time order together constitute the odor data; wherein the unit data is the odor signal received by the odor sensor within a unit time; S12: extracting multi-dimensional data features of each of the unit data in the odor data to obtain multi-dimensional data features of the odor data, and converting representative vectors according to the multi-dimensional data features to obtain a plurality of graph representative vectors corresponding to the multi-dimensional data features; wherein the multi-dimensional data features are used to express the features of the odor data from multiple data analysis dimensions, and the data analysis dimensions include extreme value dimension, frequency dimension, variance dimension, tendency dimension and change pattern dimension; S13: collectively setting the respective graph representative vectors corresponding to the multi-dimensional data features in a pre-set basic graph, and using the basic graph in which the respective graph representative vectors are set as the odor display graph.

[0032] Specifically, the odor sensor is installed and calibrated to ensure that the signal of the target odor can be continuously and accurately collected, the data of the target odor is continuously collected through the odor sensor, and the collected odor signal is recorded in chronological order. According to the time corresponding to the odor signal, the collected odor signal is divided into odor data within a number of unit times, and arranged in sequence to form a time series.

[0033] More specifically, the maximum and minimum values ​​are extracted from the time series of the odor data to reflect the peak and trough changes of the odor signal. The frequency characteristics of the odor signal are extracted through spectral analysis to reflect the periodicity and fluctuation characteristics of the odor changes. The variance of the odor data is calculated to describe the discreteness or volatility of the odor signal over a period of time. The changing trend of the odor data is analyzed to identify the trend of the signal growing or weakening over time. The changing pattern of the odor signal is extracted through pattern recognition methods to identify the complex fluctuation or periodic structure of the odor signal.

[0034] More specifically, according to the extracted multi-dimensional data features, they are converted into multiple atlas representative vectors. These vectors are used to represent data of different feature dimensions in the odor display atlas. In order to ensure consistent data visualization effects, the generated representative vectors are standardized so that vectors of different dimensions can be compared and analyzed on the same scale. Space is reserved in the pre-designed basic atlas to fill in the corresponding atlas representative vectors. The basic atlas can use a two-dimensional or three-dimensional coordinate system to express the relative position and relationship of features of different dimensions. Each atlas representative vector is set at the corresponding position of the basic atlas according to its feature dimension. This process expresses multi-dimensional features on the same atlas to form a complete odor display atlas. The final odor display atlas is used to express the feature data of the target odor in different dimensions. The atlas can be exported as an image or used for subsequent data analysis.

[0035] It is understandable that by continuously collecting odor signals, the changes in the target odor can be monitored and analyzed in real time to ensure the timeliness of the data. Through multi-dimensional feature extraction, the complex changing characteristics of the odor signal can be fully and accurately expressed to improve the analysis accuracy. The multi-dimensional analysis of the odor signal, including extreme values, frequencies, variances and other dimensions, can fully capture the characteristics of the odor signal and provide richer data support. Through tendency and change pattern analysis, the trend changes and complex fluctuations of the odor signal can be identified to provide a basis for further odor research. The multi-dimensional characteristics of the odor signal can be displayed by graphs to intuitively express the odor changes, which is convenient for users to understand and analyze the characteristics of the odor signal. Through the standardization of the representative vectors, it is ensured that the features of different dimensions can be compared and analyzed in the same graph, which improves the data availability and intuitiveness. This technology can provide technical support and data reference for odor detection, odor research and product development in related fields. Through the odor display graph, the characteristics of different odors can be more intuitively displayed, allowing users to have a better experience in odor-related products or services.

[0036] Preferably, the steps of obtaining the user's user experience data of the target odor, binding the user experience data with the odor display map to obtain a set of analysis data, and summarizing the groups of analysis data belonging to the same user into a user data set include: S21: Acquire user experience data through a data interaction port in an account login state, and generate a corresponding user tag according to the account login state of the data interaction port; S22: Binding the user experience data with the odor display map to obtain a set of analysis data, and binding the analysis data with the user tag; S23: Grouping the analysis data with the same user tag into a user data set of the user corresponding to the user tag.

[0037] Specifically, ensure that the user logs in to the account through the data interaction port, verify the login status to obtain user experience data, collect the user's experience data on the target odor through the data interaction port, including perception, evaluation and feedback information, and generate a unique user tag for each user based on the account login status for data identification and management.

[0038] More specifically, the user experience data is bound to the corresponding odor display map to form a set of analysis data. The binding process includes matching the user's experience data with the feature data in the odor display map. The analysis data is generated through the binding process. These data reflect the user's actual experience effect and feedback on the odor display map.

[0039] More specifically, the generated analysis data is bound to the user tag to ensure that each set of analysis data is associated with a specific user tag, and the user tag is matched with the analysis data one by one to form a complete analysis data set, which is convenient for subsequent analysis and management. The analysis data sets with the same user tag are aggregated into a user data set. This data set contains all the experience data and analysis data related to the user. The user data set is stored in a database or data management system to ensure data security and accessibility.

[0040] It is understandable that by identifying and managing user experience data through user tags and ensuring that each set of data is associated with a specific user, it is easy to track and analyze. All analytical data of the same user can be summarized into one data set, which helps to fully understand the user's experience and feedback. By binding the user experience data with the odor display map, it is possible to specifically analyze the user's feelings about different smells and provide personalized experience feedback. By automatically binding the user experience data with the odor display map and generating analytical data, the efficiency and accuracy of data processing can be improved. By analyzing the user data set, it is possible to deeply understand the user's odor experience and preferences, and help formulate product strategies and service optimization plans that better meet user needs.

[0041] Preferably, the steps of respectively performing parsing processing on the mapping relationship between the user experience data and the odor display map according to each group of the analysis data in the user data set to obtain the experience mapping features corresponding to each group of the analysis data include: S31: Based on each of the spectrum representative vectors in the odor display spectrum in the analysis data, a simulation allocation process of mapping weights is performed on the user experience data to obtain several simulation mapping relationships between the odor display spectrum and the user experience data, and the several simulation mapping relationships between the odor display spectrum and the user experience data are collectively used as the experience mapping features; wherein the simulation mapping relationship includes the simulation mapping weights of each of the spectrum representative vectors, and the simulation mapping weights are used to describe a possible influence direction and influence degree of the spectrum representative vector in the user experience data of the target gas; S32: Repeat the above steps for each group of the analysis data in the user data set to obtain experience mapping features corresponding to each group of the analysis data.

[0042] Specifically, analysis data is extracted from the user data set, including the odor display map and the user experience data. Based on the representative vectors of each map in the odor display map, the user experience data is simulated and assigned mapping weights. A weight allocation model is defined to simulate the direction and degree of influence of each map representative vector in the user experience data. According to the weight allocation model, weights are allocated to the user experience data and the odor display map to generate several simulated mapping relationships. Several simulated mapping relationships between the odor display map and the user experience data are calculated. These relationships include the simulated mapping weights of each map representative vector, which describes its influence on the user experience data. All calculated simulated mapping relationships are summarized to form a complete experience mapping feature.

[0043] More specifically, the simulated mapping weight is used to describe the role of the graph representative vector in all the graph representative vectors in the odor display graph. That is to say, each graph representative vector in the odor display graph provides characteristic feedback of a certain angle to the target odor. The odor characteristics fed back by each graph representative vector in the odor display graph together create the user's experience effect on the target odor. The simulated mapping weight is a possible speculation on the specific direction and degree of creation of the odor characteristics fed back by the graph representative vector in creating the user's experience effect on the target odor. Based on this, the simulated mapping relationship is a possible speculation on how the odor characteristics fed back by each graph representative vector in the odor display graph create the user's experience effect on the target odor.

[0044] More specifically, multiple simulated mapping relationships can be inferred based on a single piece of analysis data, and it is impossible to determine which simulated mapping relationship is correct. Therefore, target data of multiple different forms of odor data can be obtained, thereby obtaining multiple simulated mapping relationships corresponding to the multiple odor data for subsequent analysis to determine the correct mapping relationship, that is, how the various characteristics of the odor create the user experience effect.

[0045] More specifically, the obtained several simulated mapping relationships are used as experience mapping features, which comprehensively reflect the matching relationship between user experience data and odor display maps. The experience mapping features are stored in the data management system to ensure the integrity and availability of the data.

[0046] More specifically, the above steps are repeated for each group of analysis data in the user data set to generate corresponding experience mapping features one by one, and all generated experience mapping features are integrated into a feature set to facilitate subsequent analysis and application.

[0047] It can be understood that by simulating the allocation of mapping weights for each representative vector in the odor display map, the mapping relationship in the user experience data can be described in detail. The simulated mapping relationship can be used to accurately reflect the direction and degree of influence of the odor display map in the user experience data. Several simulated mapping relationships can be used together as experience mapping features to comprehensively describe the relationship between the user experience data and the odor display map. Through the simulated allocation of mapping weights, rich experience mapping features can be generated to provide rich data support for subsequent analysis and model training. Through automated weight allocation and mapping relationship calculation, data processing efficiency can be improved and manual intervention can be reduced. Large-scale user data sets can be batch processed and experience mapping features can be quickly generated to improve data processing efficiency. Through experience mapping features, the matching relationship between user experience data and the odor display map can be deeply analyzed to help understand users' perception and feedback on odors. Based on the experience mapping features, key factors in user experience can be identified to provide data support for the optimization and improvement of odor products.

[0048] Preferably, the step of combining and analyzing the experience mapping features corresponding to each group of analysis data to obtain the mapping feature distribution of the user includes: S33: performing frequency query processing on several types of simulation mapping relationships in each group of the experience mapping features to obtain the occurrence frequencies of various types of simulation mapping relationships in each group of the experience mapping features, and generating first confidence weights of various types of simulation mapping relationships according to the occurrence frequencies of various types of simulation mapping relationships in each group of the experience mapping features; S34: Substituting the simulated mapping relationship in which the first mapping weight in the experience mapping feature meets the preset standard into each group of the analysis data, respectively, so that the simulated mapping feature is adjusted based on the odor display map and the user experience data in each group of the analysis data to obtain a fitting mapping feature of the simulated mapping feature; S35: performing analysis processing on the mapping suitability of each of the fitting mapping features according to each group of analysis data, so as to obtain a second confidence weight of each of the fitting mapping features; S36: Compare the second confidence weights of the respective fitting mapping features to obtain a comparison result, and use the fitting mapping feature with the best second confidence weight in the comparison result as the mapping feature distribution.

[0049] Specifically, all simulated mapping relationships are extracted from each group of experience mapping features, the frequency of occurrence of each simulated mapping relationship in each group of experience mapping features is calculated, the frequency data is recorded, and the first confidence weights of various simulated mapping relationships are generated according to the frequency of occurrence. The higher the frequency, the greater the weight, and vice versa.

[0050] More specifically, the experience mapping feature describes several mapping relationships between the user experience data and the odor display map in the analysis data. By analyzing the frequency of occurrence of the simulated mapping relationship in each experience mapping feature, it is determined whether the simulated mapping relationship is suitable for each experience mapping feature.

[0051] More specifically, the simulated mapping relationships whose first confidence weights meet the preset standards are screened out, and the screened simulated mapping relationships are respectively substituted into each group of analysis data, and the weights of the simulated mapping relationships are adjusted based on the odor display maps and user experience data in each group of analysis data to generate fitting mapping features.

[0052] More specifically, the simulated mapping relationship whose first confidence weight meets the preset standard does not necessarily fully match all experience mapping features. Therefore, it is necessary to make subtle adjustments to the allocated weights of the simulated mapping relationship whose first confidence weight meets the preset standard, so that the simulated mapping relationship can be closer to those analysis data that have not been satisfied without deviating from the original experience mapping features, thereby obtaining the fitting mapping features that meet all analysis data to the greatest extent. By comparing these fitting mapping features, the optimal mapping features of the mapping relationship that meet all analysis data can be obtained.

[0053] More specifically, for each set of analysis data, the mapping fitness of each fitting mapping feature is evaluated, and the fitness score of each fitting mapping feature is calculated. Based on the fitness score, a second confidence weight for each fitting mapping feature is generated. The higher the fitness, the greater the weight. The second confidence weights of all fitting mapping features are compared, the comparison results are recorded, and the fitting mapping feature with the best second confidence weight is selected as the final mapping feature distribution.

[0054] It can be understood that by querying and counting the frequency of simulated mapping relationships, the commonness of each mapping relationship in user experience data is quantified, and the frequency data is used to generate a first confidence weight to provide a basis for subsequent weight adjustment processing. Reliable simulated mapping relationships are screened out through preset standards to ensure the accuracy of the weight adjustment processing. Based on the actual analysis data, the weights of the simulated mapping relationships are dynamically adjusted to generate more accurate fitting mapping features. A fitness analysis is performed on each fitting mapping feature to ensure a high degree of match between the mapping features and the user experience data. According to the fitness score, a second confidence weight is generated to further optimize the mapping features. By comparing the second confidence weights of all fitting mapping features, the optimal mapping feature distribution is selected to ensure the reliability and accuracy of the results. Finally, the user's mapping feature distribution is obtained, providing solid data support for user experience analysis and scent optimization.

[0055] Preferably, the step of uploading each of the odor display graphs to a designated network service platform comprises: S41: transmitting each of the odor display graphs to a pre-built network service platform for interactive analysis of odor characteristics; S42: Instruct the network service platform to store and process each of the odor display graphs.

[0056] Specifically, confirm and access the pre-built network service platform for interactive analysis of odor characteristics, prepare the odor display map files to be uploaded, ensure their format and data integrity, upload each odor display map file to the designated platform through the network interface, and ensure that the upload process is stable and the files are complete.

[0057] More specifically, the platform receives and verifies the uploaded odor display map file, confirms that the file is correct, and stores the odor display map file in a designated data storage system to ensure the security and persistence of the data. The platform records the storage location and metadata of the odor display map for subsequent retrieval and management.

[0058] It is understandable that ensuring that the odor display map file is uploaded to the platform in a complete and stable manner enables the platform to achieve secure storage, protect the data from loss or damage, facilitate subsequent retrieval and analysis operations, and improve data utilization efficiency.

[0059] Preferably, the step of selecting any of the odor display graphs from the network service platform as a simulation object, and performing simulation analysis on the simulation object according to the mapping feature distribution of the user to obtain the simulation experience effect of the simulation object includes: S43: selecting any of the odor display graphs from the network service platform as a simulation object; S44: constructing a simulation target based on the simulation object; wherein the simulation target is used to output user experience data to provide feedback on the user's simulation experience effect on the simulation object; S45: enabling the mapping feature distribution to perform data mapping processing on the simulation object and the simulation target, so that the simulation target outputs the user experience data of the simulation object corresponding to the mapping feature distribution, and uses the output user experience data as the simulation experience effect of the simulation object.

[0060] Specifically, any odor display map is selected from the network service platform as a simulation object. The simulation object will serve as the basic data for subsequent simulation analysis, and the selected odor display map will be loaded into the simulation analysis system to ensure its integrity and availability.

[0061] More specifically, a simulation target is constructed based on the selected simulation object. The simulation target is used to generate and output feedback data related to the user experience. In the simulation target, an output function is designed to enable it to output user experience data related to the odor display map. This output is used to reflect the user's perception effect of the odor.

[0062] More specifically, the user's mapping feature distribution data is loaded, which is used to describe the user's reaction and perception to various odor display maps, and the user's mapping feature distribution is data mapped with the simulation object and the simulation target. Through this process, the simulation target can output user experience data that matches the user's mapping feature distribution.

[0063] More specifically, the simulation target is made to output user experience data corresponding to the simulation object. The output data represents the user's perception effect under the odor display map simulation, and the output user experience data is used as the final simulation experience effect of the simulation object to analyze and provide feedback on the user's response to the odor.

[0064] It can be understood that by matching the user's mapping feature distribution with the odor display map, the simulation process can more accurately reflect the user's perception of a specific odor. According to the user's mapping feature distribution, the simulation object outputs a personalized experience effect, ensuring the fit and authenticity of the simulation results. The simulation target can output user experience data in real time, provide users with instant feedback, and help evaluate the simulation effect. Through feedback data, the mapping features in the simulation process can be dynamically adjusted to improve the accuracy and effect of the simulation analysis.

[0065] Preferably, it also includes: obtaining user experience data of the user for the several experience smells by applying several pre-specified experience smells to the user, and binding the user experience data with the odor display map of the experience smell to obtain several groups of analysis data for performing analysis processing of mapping feature distribution.

[0066] Specifically, in order to obtain the user's olfactory perception state, it is necessary to analyze the mapping relationship between the user's user experience data and the odor display map, and this analysis requires the support of multiple groups of data to ensure the reliability of the analysis effect. Therefore, before calculating the mapping feature distribution, the user can specify several experience smells in turn, and let these experience smells respectively serve as the target smell to calculate the mapping relationship between the user's user experience data for this experience smell and the odor display map of the experience smell, so as to obtain the user's mapping feature distribution and know the user's olfactory perception state.

[0067] More specifically, the experience smell can be packaged in a designated container. When the experience smell needs to be released, the experience smell is connected to the smell release device, and the experience smell is released into the air through the smell release device, so that the user can experience it to obtain the user's user experience data.

[0068] Reference Figure 2 As shown, in a second aspect, the present invention provides an application of an odor feature interactive analysis method, which is used to implement an odor feature interactive analysis method described in any one of the first aspects, including: An odor collection module is used to collect odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and perform a graph conversion process based on the odor data to generate an odor display graph for graphical expression; An odor processing module is used to obtain the user's user experience data of the target odor, and bind the user experience data with the odor display map to obtain a set of analysis data, and summarize the groups of analysis data belonging to the same user into a user data set; A mapping analysis module, used to analyze the mapping relationship between the user experience data and the odor display map according to each group of analysis data in the user data set, so as to obtain experience mapping features corresponding to each group of analysis data, and combine and analyze the experience mapping features corresponding to each group of analysis data to obtain the user's mapping feature distribution; wherein the experience mapping feature is used to describe the correlation between the odor data and the user experience, and the mapping feature distribution is used to describe the user's own odor experience ability; The experience simulation module is used to upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulated experience effect of the simulation object; wherein the simulated experience effect is a simulation of the user's experience of the simulation object.

[0069] In this embodiment, for the specific implementation of each module in the application embodiment of the above method, please refer to the description in the above method embodiment, which will not be repeated here.

[0070] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for interactive analysis of odor characteristics, characterized in that: include: The odor data of the target odor is collected by an odor sensor to obtain odor data corresponding to the target odor, and a graph conversion process is performed according to the odor data to generate an odor display graph for graphical expression; Acquire user experience data of the target odor from the user, and bind the user experience data with the odor display map to obtain a set of analysis data, and summarize the groups of analysis data belonging to the same user into a user data set; According to each group of the analysis data in the user data set, the mapping relationship between the user experience data and the odor display map is analyzed and processed respectively to obtain the experience mapping features corresponding to each group of the analysis data, and the experience mapping features corresponding to each group of the analysis data are combined and analyzed to obtain the mapping feature distribution of the user; wherein the experience mapping feature is used to describe the correlation relationship between the odor data and the user experience, and the mapping feature distribution is used to describe the odor experience ability of the user itself; Upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulation experience effect of the simulation object; wherein the simulation experience effect is a simulation of the user's experience of the simulation object.

2. The odor feature interactive analysis method according to claim 1, characterized in that: The steps of collecting odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and performing a graph conversion process according to the odor data to generate an odor display graph for graph expression include: The odor sensor continuously collects odor signals of the target odor, and summarizes each odor signal collected into a plurality of sequentially arranged unit data according to the time corresponding to the odor signal, and each of the unit data sequentially arranged in time order together constitutes the odor data; wherein the unit data is the odor signal received by the odor sensor within a unit time; Performing a multi-dimensional data feature extraction process on each of the unit data in the odor data to obtain the multi-dimensional data features of the odor data, and performing a representative vector conversion process according to the multi-dimensional data features to obtain a plurality of graph representative vectors corresponding to the multi-dimensional data features; wherein the multi-dimensional data features are used to express the features of the odor data from multiple data analysis dimensions, and the data analysis dimensions include extreme value dimension, frequency dimension, variance dimension, tendency dimension and change pattern dimension; The respective graph representative vectors corresponding to the multi-dimensional data features are collectively set in a pre-set basic graph, and the basic graph provided with the respective graph representative vectors is used as the odor display graph.

3. The odor feature interactive analysis method according to claim 1, characterized in that: The steps of obtaining user experience data of the target odor from the user, and binding the user experience data with the odor display map to obtain a set of analysis data, and summarizing the analysis data of each group belonging to the same user into a user data set include: Acquire user experience data through a data interaction port in an account login state, and generate a corresponding user tag according to the account login state of the data interaction port; Binding the user experience data with the odor display map to obtain a set of analysis data, and binding the analysis data with the user tag; The various analysis data having the same user tag are summarized as a user data set of the user corresponding to the user tag.

4. The odor feature interactive analysis method according to claim 2, characterized in that: The steps of respectively performing parsing processing on the mapping relationship between the user experience data and the odor display map according to each group of the analysis data in the user data set to obtain the experience mapping features corresponding to each group of the analysis data include: Based on each of the spectrum representative vectors in the odor display spectrum in the analysis data, a simulation allocation process of mapping weights is performed on the user experience data to obtain several simulation mapping relationships between the odor display spectrum and the user experience data, and the several simulation mapping relationships between the odor display spectrum and the user experience data are collectively used as the experience mapping features; wherein the simulation mapping relationship includes a simulation mapping weight of each of the spectrum representative vectors, and the simulation mapping weight is used to describe a possible influence direction and influence degree of the spectrum representative vector in the user experience data of the target gas; Repeat the above steps for each group of the analysis data in the user data set to obtain experience mapping features corresponding to each group of the analysis data.

5. The odor feature interactive analysis method according to claim 4, characterized in that: The step of combining and analyzing the experience mapping features corresponding to each group of analysis data to obtain the mapping feature distribution of the user includes: Performing frequency query processing on several types of simulation mapping relationships in each group of the experience mapping features to obtain the occurrence frequencies of various types of simulation mapping relationships in each group of the experience mapping features, and generating first confidence weights of various types of simulation mapping relationships according to the occurrence frequencies of various types of simulation mapping relationships in each group of the experience mapping features; Substituting the simulated mapping relationship in which the first mapping weight in the experience mapping feature meets the preset standard into each group of the analysis data, respectively, so that the simulated mapping feature is adjusted based on the odor display spectrum and the user experience data in each group of the analysis data to obtain a fitting mapping feature of the simulated mapping feature; According to each group of analysis data, each of the fitting mapping features is analyzed for mapping suitability to obtain a second confidence weight of each of the fitting mapping features; The second confidence weights of the respective fitting mapping features are compared to obtain a comparison result, and the fitting mapping feature having the best second confidence weight in the comparison result is used as the mapping feature distribution.

6. The odor feature interactive analysis method according to claim 1, characterized in that: The step of uploading each of the odor display graphs to a designated network service platform includes: Transmitting each of the odor display graphs to a pre-built network service platform for interactive analysis of odor characteristics; The network service platform is instructed to store and process each of the odor display graphs.

7. The odor feature interactive analysis method according to claim 1, characterized in that: The steps of selecting any of the odor display graphs from the network service platform as a simulation object, and performing simulation analysis on the simulation object according to the mapping feature distribution of the user to obtain the simulation experience effect of the simulation object include: Select any of the odor display graphs from the network service platform as a simulation object; Constructing a simulation target based on the simulation object; wherein the simulation target is used to output user experience data to provide feedback on the user's simulation experience effect on the simulation object; The mapping feature distribution is used to perform data mapping processing on the simulation object and the simulation target, so that the simulation target outputs the user experience data of the simulation object corresponding to the mapping feature distribution, and the output user experience data is used as the simulation experience effect of the simulation object.

8. The odor feature interactive analysis method according to claim 1, characterized in that: Also includes: By applying several pre-specified experience smells to the user, the user experience data of the user for the several experience smells is obtained, and the user experience data is bound to the odor display map of the experience smell to obtain several groups of analysis data for analysis processing of mapping feature distribution.

9. An application of an odor feature interactive analysis method, characterized in that: A method for interactively analyzing odor characteristics according to any one of claims 1 to 7, comprising: An odor collection module is used to collect odor data of a target odor through an odor sensor to obtain odor data corresponding to the target odor, and perform a graph conversion process based on the odor data to generate an odor display graph for graphical expression; An odor processing module is used to obtain the user's user experience data of the target odor, and bind the user experience data with the odor display map to obtain a set of analysis data, and summarize the groups of analysis data belonging to the same user into a user data set; A mapping analysis module, used to analyze the mapping relationship between the user experience data and the odor display map according to each group of analysis data in the user data set, so as to obtain experience mapping features corresponding to each group of analysis data, and combine and analyze the experience mapping features corresponding to each group of analysis data to obtain the user's mapping feature distribution; wherein the experience mapping feature is used to describe the correlation between the odor data and the user experience, and the mapping feature distribution is used to describe the user's own odor experience ability; The experience simulation module is used to upload each of the odor display maps to a designated network service platform, select any of the odor display maps from the network service platform as a simulation object, and perform simulation analysis on the simulation object according to the user's mapping feature distribution to obtain a simulated experience effect of the simulation object; wherein the simulated experience effect is a simulation of the user's experience of the simulation object.

Images