Car purchase user experience evaluation method and system

Abstract

The invention discloses a car purchase user experience evaluation method and system. The method comprises the steps of collecting social media data about car purchase user experience; constructing a user experience evaluation questionnaire according to the social media data; and taking the survey result of the user experience evaluation questionnaire as market feedback data. According to the invention, the user experience evaluation questionnaire is constructed based on social media data, objective and real vehicle use experience of the vehicle purchase user is mined based on media big data, and comprehensive and objective evaluation items are obtained through data analysis. The defects that in a traditional questionnaire, the sample size is small, representativeness is insufficient, evaluation item setting is difficult to cover all dimensions, evaluation items are lack of objectivity and the like are overcome, and therefore more accurate market feedback and improvement directions are provided for vehicle enterprises and dealers.

Description

Technical Field

[0001] This application relates to the technical field of vehicles, and more specifically, to a method and system for evaluating the purchase user experience of vehicles. Background Art

[0002] In the context of the increasingly fierce competition in today's automotive market, the purchase user experience has become the focus of attention for automotive manufacturers and dealers. The quality of the user experience directly affects consumers' purchase decisions, and in turn affects the reputation and market share of automotive brands. Therefore, accurately evaluating the user experience is crucial for vehicle manufacturers to enhance user reputation.

[0003] Traditional user experience evaluations mainly rely on questionnaires to collect users' opinions and experience feedback. Although this method is relatively convenient to implement, it exposes significant defects, such as a small sample size, insufficient representativeness, and since the questionnaire design lacks data support, it is difficult for the evaluation items to cover all dimensions, and the evaluation items on the questionnaire are greatly affected by the subjective feelings of the designers, lacking objectivity. These limitations make it difficult for this method to accurately depict the real experience of users during actual vehicle use. Summary of the Invention

[0004] This application provides a method and system for evaluating the purchase user experience of vehicles, constructs a user experience evaluation questionnaire based on social media data, mines the objective and real vehicle use experience of purchase users based on media big data, obtains comprehensive and objective evaluation items through data analysis, and makes up for the defects such as small sample size, insufficient representativeness, difficulty in covering all dimensions in the evaluation item setting, and lack of objectivity in the evaluation items in traditional questionnaires, so as to provide more accurate market feedback and improvement directions for vehicle manufacturers and dealers.

[0005] This application provides a method for evaluating the purchase user experience of vehicles, including:

[0006] Collect social media data regarding the purchase user experience;

[0007] Construct a user experience evaluation questionnaire based on the social media data;

[0008] Use the survey results of the user experience evaluation questionnaire as market feedback data.

[0009] Preferably, constructing a user experience evaluation questionnaire based on social media data specifically includes:

[0010] Analyze the social media data to obtain multiple subjective psychological latent variables;

[0011] Determine multiple observed variables corresponding to the subjective psychological latent variables;

[0012] Design a user experience evaluation questionnaire based on all the observed variables.

[0013] Preferably, in the user experience evaluation questionnaire, the observed variables are evaluated by a scale method.

[0014] Preferably, analyze the social media data to obtain multiple subjective psychological latent variables, specifically including:

[0015] Identify the attitude of car-buying users towards vehicle use based on social media data;

[0016] Mine multiple core themes that car-buying users are concerned about based on social media data;

[0017] Take the usage intention, attitude of car-buying users, and the latent variables corresponding to all core themes as subjective psychological latent variables.

[0018] Preferably, constructing a user experience evaluation questionnaire based on social media data further includes:

[0019] Conduct a survey using the user experience evaluation questionnaire;

[0020] Based on the survey results of subjective psychological latent variables and the views of car-buying users on vehicles, determine whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data;

[0021] If not, adjust the observed variables corresponding to the subjective psychological latent variables and redesign the user experience evaluation questionnaire.

[0022] Preferably, the car-buying user experience evaluation method further includes:

[0023] Analyze the significance of the influence between pairwise subjective psychological latent variables based on market feedback data;

[0024] Adjust the marketing strategy according to the influence significance.

[0025] This application also provides a car-buying user experience evaluation system, including a collection module, a questionnaire construction module, and a feedback module;

[0026] The collection module is used to collect social media data on car-buying user experience;

[0027] The questionnaire construction module is used to construct a user experience evaluation questionnaire based on social media data;

[0028] The feedback module is used to take the survey results of the user experience evaluation questionnaire as market feedback data.

[0029] Preferably, the questionnaire construction module includes a first analysis module, an observed variable determination module, and a design module;

[0030] The first analysis module is used to analyze social media data to obtain multiple subjective psychological latent variables;

[0031] The observation variable determination module is used to determine multiple observation variables corresponding to the subjective psychological latent variables;

[0032] The design module is used to design a user experience evaluation questionnaire based on all the observation variables.

[0033] Preferably, in the user experience evaluation questionnaire, the observation variables are evaluated by a scale method.

[0034] Preferably, the first analysis module includes an attitude recognition module, a topic recognition module, and a latent variable determination module;

[0035] The attitude recognition module is used to identify the attitude of car-buying users towards using vehicles based on social media data;

[0036] The topic recognition module is used to mine multiple core topics that car-buying users are concerned about based on social media data;

[0037] The latent variable determination module is used to use the latent variables corresponding to the usage intention, attitude, and all core topics of car-buying users as subjective psychological latent variables.

[0038] Preferably, the questionnaire construction module further includes a research module, a judgment module, and a first adjustment module;

[0039] The research module is used to conduct research using the user experience evaluation questionnaire;

[0040] The judgment module is used to judge whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data based on the research results of the subjective psychological latent variables and the views of car-buying users on the vehicle;

[0041] The first adjustment module is used to adjust the observation variables corresponding to the subjective psychological latent variables when the user concerns reflected in the user experience evaluation questionnaire are inconsistent with the user concerns obtained from social media data; and,

[0042] The design module is used to redesign the user experience evaluation questionnaire based on the adjusted observation variables.

[0043] Preferably, the car user experience evaluation system further includes a second analysis module and a second adjustment module;

[0044] The second analysis module is used to analyze the significance of the influence between pairwise subjective psychological latent variables based on market feedback data;

[0045] The second adjustment module is used to adjust the marketing strategy based on the influence significance.

[0046] Other features and advantages of the present application will become apparent from the following detailed description of exemplary embodiments of the present application with reference to the accompanying drawings. Description of the Drawings

[0047] The drawings incorporated in and constituting a part of this specification illustrate embodiments of the present application and, together with the description thereof, serve to explain the principles of the present application.

[0048] Figure 1 Flowchart of the method for evaluating the purchase user experience provided by the present application;

[0049] Figure 2 Probability graph of the theme and related keywords provided by the present application;

[0050] Figure 3 Path relationship graph of the subjective psychological latent variables of an embodiment provided by the present application;

[0051] Figure 4 Structure diagram of the purchase user experience evaluation system provided by the present application. Detailed Embodiments

[0052] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application.

[0053] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way a limitation on the present application or its application or use.

[0054] Techniques, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods, and devices should be considered as part of the specification.

[0055] In all examples shown and discussed herein, any specific values should be construed as merely exemplary and not as a limitation. Thus, other examples of exemplary embodiments may have different values.

[0056] The present application provides a method and system for evaluating the purchase user experience. A user experience evaluation questionnaire is constructed based on social media data. The objective and real vehicle use experience of purchase users is mined from media big data. Through data analysis, comprehensive and objective evaluation items are obtained, making up for the defects in traditional questionnaires such as small sample size, insufficient representativeness, difficulty in covering all dimensions in the setting of evaluation items, and lack of objectivity in evaluation items. Thus, more accurate market feedback and improvement directions are provided for automobile manufacturers and dealers. Further, the present application mines the core themes concerned by purchase users, as well as the vehicle use attitudes and use intentions of users from social media data, thereby forming subjective psychological latent variables that conform to the actual psychology of users. Based on this, a user experience evaluation questionnaire is constructed, which can comprehensively and accurately obtain the feelings of users and improve the quality of the questionnaire survey. In addition, the present application analyzes the significance of the influence between pairwise subjective psychological latent variables based on the questionnaire survey, thereby determining the influence of these latent variables on the user experience, and thus formulating targeted marketing strategies.

[0057] As Figure 1 shown, the method for evaluating the purchase user experience provided by the present application includes:

[0058] S110: Collect social media data on the purchase user experience.

[0059] As an embodiment, based on a Python crawler program, collect the experience feedback of purchase users on a certain vehicle model in a social media platform (such as an automobile platform). The data includes user ID, comment content, comment time, and IP address.

[0060] It can be understood that before executing S120, it is necessary to perform data preprocessing on the collected social media data, including: removing noise and irrelevant information in the comment content text, such as emojis, links, and punctuation marks; removing abbreviations, misspellings, and informal terms, and uniformly formatting the text to ensure data consistency; calling the word segmentation function of the Jieba library, taking the text to be segmented as input, and segmenting the text according to the words and phrases in the dictionary; referring to multiple authoritative sources, setting conjunctions and stop word lists according to the text to avoid interference of common but meaningless words in the text analysis task.

[0061] S120: Construct a user experience evaluation questionnaire based on the social media data.

[0062] As an embodiment, constructing a user experience evaluation questionnaire based on the social media data specifically includes:

[0063] S1201: Analyze the social media data to obtain multiple subjective psychological latent variables.

[0064] As an embodiment, analyzing the social media data to obtain multiple subjective psychological latent variables specifically includes:

[0065] P1: Identify the attitudes of vehicle users based on social media data.

[0066] As an example, the sentiment analysis method is used to classify the sentiment of the comment content text to identify the attitudes of users towards vehicle use.

[0067] As an example, by calling the SnowNLP library, the sentiment of the comment content text is classified using a pre-trained Bayesian model with an accuracy of over 80%. The output result of the model is a sentiment score in the interval [0, 1]. A score of 0.7 or higher is considered a positive sentiment, a score in the range of 0.3 - 0.7 is a neutral sentiment, and otherwise it is a negative sentiment.

[0068] P2: Mine multiple core topics that vehicle purchase users are concerned about based on social media data.

[0069] As an example, the topic model is used to mine the core topics that users are concerned about. The topic model is the Latent Dirichlet Allocation (LDA) model, which is an unsupervised probabilistic generative model for discovering hidden topics in text data. The model assumes that a document is generated by a mixture of multiple latent topics, and each topic is a probability distribution of a set of words. By iteratively reassigning topics to each word in the comment content text and updating the word distribution under each topic until the model converges, the topic distribution of each comment content text and the word distribution of each topic are finally output.

[0070] Among them, the model needs to pre-specify the number of topics. Too many topics may make the model too complex, while too few may not be able to capture the subtle differences in the text. In this application, the topic coherence score is used to determine the number of topics. When the coherence score shows an inflection point, the number of topics is the most appropriate.

[0071] Since LDA is an unsupervised learning method, each topic needs to be manually labeled. The labeling process refers to the keywords with higher frequencies under the topic, thereby ensuring that the LDA model accurately extracts the core topics. As an example, Figure 2 the top 10 keywords with the highest probabilities related to each topic are listed, and at the same time, a topic and related keyword probability graph is drawn according to the topic-keyword probability.

[0072] As an example, the core topics that vehicle purchase users are concerned about mined based on social media data include price, usefulness, ease of use, government subsidies, and publicity.

[0073] P3: Take the usage intention, attitude of vehicle purchase users, and the latent variables corresponding to all core topics as subjective psychological latent variables.

[0074] When determining the subjective psychological latent variables corresponding to the core theme, first collect relevant subjective psychological latent variables and their original definitions from the classical behavioral theory frameworks involved in the existing literature; during the analysis process, if the core theme conforms to the definition of a certain subjective psychological latent variable, then according to this behavioral theory, designate this subjective psychological latent variable as the label of this core theme.

[0075] Preferably, after determining the above corresponding relationship, several experienced researchers were also invited to independently carry out the matching work between the core theme and the subjective psychological latent variables; then they jointly participated in the discussion to ensure an accurate match between the subjective psychological latent variables and the core theme.

[0076] In the above example, the matching results between the core theme and the subjective psychological latent variables are shown in Table 1:

[0077] Table 1: Matching Results between Core Theme and Subjective Psychological Latent Variables

[0078]

[0079] Thus, the subjective psychological latent variables include attitude, price factor, perceived usefulness, perceived ease of use, conditional value, media exposure, and usage intention.

[0080] Among them, "attitude" refers to the positive or negative evaluation of users during the vehicle usage process, that is, the degree of preference or recognition of users for the vehicle; "price factor" refers to the cost of the vehicle during the vehicle usage process, which can be divided into four aspects: vehicle price, fuel cost, parking cost, and maintenance cost; "perceived usefulness" refers to users' perception of the usefulness of the vehicle in meeting traffic needs, environmental protection performance, economy, convenience, etc.; "perceived ease of use" refers to users' perception of aspects such as vehicle operation convenience, learning cost, and interface friendliness; "conditional value" refers to users' attention to the subsidies from central and local governments that can reduce the vehicle purchase cost; "media exposure" refers to the degree of relevant information that users come into contact with on automotive platforms and other social media; "usage intention" refers to users' tendency to purchase or use the vehicle in the future.

[0081] S1202: Determine multiple observed variables corresponding to the subjective psychological latent variables.

[0082] Latent variables are variables that cannot be directly observed and need to be indirectly measured through multiple observed variables. Observed variables are variables that can be directly observed and measured, that is, the external expressions that can reflect subjective psychological latent variables.

[0083] As an example, the observed variables of the subjective psychological latent variable attitude include: I think using new energy vehicles is a good idea; I think using new energy vehicles is a wise decision; I think using new energy vehicles is a pleasant experience.

[0084] The observed variables of the subjective psychological latent variable price factor include: I think the price of fuel is much higher than that of electricity; I think the maintenance cost of new energy vehicles is very high; I think the maintenance cost of new energy vehicles is lower.

[0085] The observed variables of the subjective psychological latent variable perceived usefulness include: I think the functions of new energy vehicles are very practical; I think new energy vehicles can reduce environmental pollution; I think it is very convenient to charge new energy vehicles.

[0086] The observed variables of the subjective psychological latent variable perceived ease of use include: I think the structure of new energy vehicles is simple and the degree of intelligence is high; I think there is no need to queue up during vehicle annual inspection; I think it is very convenient to park new energy vehicles.

[0087] The observed variables of the subjective psychological latent variable conditional value include: When the government subsidies for purchasing vehicles reach my expectation, I will buy new energy vehicles; When the discounts given by enterprises reach my expectation, I will buy new energy vehicles; When the promotional activities of manufacturers reach my expectation, I will buy new energy vehicles.

[0088] The observed variables of the subjective psychological latent variable media exposure include: I have seen my friends share information about new energy vehicles in the circle of friends; I have seen evaluation information about new energy vehicles on social media; I have seen news reports about new energy vehicles on popular social platforms.

[0089] The observed variables of the subjective psychological latent variable usage intention include: I plan to use new energy vehicles; I plan to use new energy vehicles; I want to use new energy vehicles.

[0090] S1203: Design a user experience evaluation questionnaire based on all observed variables.

[0091] The user experience evaluation questionnaire includes the basic information of the respondents and all observed variables, and evaluates the observed variables through a scale method. Among them, the basic information includes gender, age, income, and the ownership of private vehicles, etc.

[0092] As an embodiment, the Likert seven-point scale is used to measure the respondents' cognition of the observed variables, with 7 being the best and 1 being the worst. Among them, 1 means very disagree, 7 means very agree, and the transition is gradual in between.

[0093] Preferably, the user experience evaluation questionnaire also includes the socioeconomic attributes of the respondents, subjective variables, and their views on the target vehicle.

[0094] Preferably, a user experience evaluation questionnaire is constructed based on social media data, and the verification of the questionnaire is also included, specifically including:

[0095] S1204: Conduct a survey using the user experience evaluation questionnaire.

[0096] S1205: Based on the survey results of subjective psychological latent variables and the views of car-buying users on the target vehicle, determine whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data. If so, execute S130. Otherwise, execute S1206.

[0097] As an embodiment, the survey results of subjective psychological latent variables are the total scale value or the average scale value of each subjective psychological latent variable. By sorting the survey results of all subjective psychological latent variables and combining the views of car-buying users on the target vehicle, the user concerns reflected in the user experience evaluation questionnaire can be determined. By comparing it with the user concerns obtained from social media data (such as the above-mentioned core themes), it can be seen whether there is a deviation in the concerns.

[0098] S1206: If the user concerns reflected in the user experience evaluation questionnaire are inconsistent with the user concerns obtained from social media data, it means that the observed variables are not accurate enough. Then adjust the observed variables corresponding to the subjective psychological latent variables, and then return to S1203 to redesign the user experience evaluation questionnaire. Until the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data.

[0099] S130: Take the survey results of the user experience evaluation questionnaire as market feedback data.

[0100] As an embodiment, each item in the user experience evaluation questionnaire is statistically analyzed, and the statistical results are used as the survey results to form market feedback data to assist in making marketing decisions.

[0101] However, simple data statistics cannot reflect the impact of subjective psychological latent variables on user experience. Based on such considerations, preferably, the car-buying user experience evaluation method provided by this application further includes:

[0102] S140: Analyze the significance of the influence between subjective psychological latent variables pairwise based on the market feedback data.

[0103] As an embodiment, a structural equation model is used to mine the significance of the influence between subjective psychological latent variables and the corresponding observed variables, as well as between subjective psychological latent variables pairwise, so as to mine the influence on user experience.

[0104] The structural equation model covers two key elements, namely the measurement model and the structural model. In the measurement model, the corresponding subjective psychological latent variables are measured through observed variables, while allowing for the measurement characteristics of the observed variables. The focus of the measurement model is to construct a reliable framework to accurately measure the latent variables. The basic form of the measurement model is as follows:

[0105] A = Λ a α + ε

[0106] B = Λ b β + δ

[0107] Where A is the exogenous observed variable; B is the endogenous observed variable; α is the exogenous latent variable (including price factor, perceived usefulness, perceived ease of use, conditional value, media exposure); β is the endogenous latent variable (including attitude and usage intention); ε is the measurement error of the exogenous observed variable A; δ is the measurement error of the endogenous observed variable B; Λ a is the factor loading matrix, representing the influence of the exogenous latent variable α on the exogenous observed variable A; Λ b represents the influence of the endogenous latent variable β on the endogenous observed variable B.

[0108] The structural model is used to describe the causal relationship between subjective psychological latent variables based on a system of linear equations, revealing the interaction and influence of subjective psychological latent variables. The basic form of the structural model is as follows:

[0109] β1 = μ2β2 + Γα + γ

[0110] Where, β1 represents the explained endogenous latent variable, i.e., the dependent variable; β2 represents another endogenous latent variable other than β1, which is the independent variable; μ2 is the coefficient matrix between endogenous latent variables, reflecting the mutual influence of endogenous latent variables; Г is the coefficient matrix between exogenous latent variables and endogenous latent variables, reflecting the influence of exogenous latent variables on endogenous latent variables; Υ is the residual term that cannot be explained by the endogenous latent variable in the equation.

[0111] After constructing the structural equation model, factor analysis needs to be used to explore the relationship between subjective psychological latent variables and observed variables, and confirmatory factor analysis is used to verify the reliability and validity of the measurement model to ensure the reliability and validity of the data.

[0112] As an example, the KMO (Kasiser-Meter-Olkin) test is used to measure the correlation between observed variables, and the Bartlett sphericity test is used to test the independence between observed variables. When the KMO index is higher than 0.7 and the significance of the Bartlett sphericity test is less than 0.001, the data is considered suitable for factor analysis.

[0113] As an example, based on the principal component analysis method, the observed variables included in the subjective psychological latent variable are orthogonally rotated using the varimax method to obtain the rotated component matrix. Extract the factors with factor load values greater than 0.5 to ensure a one-to-one correspondence between the subjective psychological latent variable and the number of factors, which proves that the factor analysis effect is good.

[0114] As an example, for the reliability test, the Cronbach's alpha coefficient and the composite reliability (CR) are used as evaluation indicators to verify whether the respondents have good consistency when answering different observed variables of the same subjective psychological latent variable. If the Cronbach's alpha coefficient is greater than 0.7 and the CR value is greater than 0.7, it indicates that the questionnaire scale has high stability and reliability. The calculation formula of CR is as follows:

[0115]

[0116] where ω is the standardized factor loading score and μ1 is the error measurement value of the observed variable.

[0117] The validity test evaluates the accuracy of the measurement method for the measured object according to the convergent validity and discriminant validity. As an example, the validity is tested based on the standardized factor loading coefficient and the average variance extracted (AVE) value of the subjective psychological latent variable. If all standardized factor loading coefficients are greater than 0.6 and the AVEs are greater than 0.5, it indicates that the scale has good convergent validity; if the square root of the AVE of each variable is higher than the correlation coefficient with other variables, it proves that the scale has good discriminant validity. The calculation formula of AVE is as follows:

[0118]

[0119] where ω 2 is the index reliability of the observed variable.

[0120] The test results of the discriminant validity are shown in Table 2:

[0121] Table 2: Test Results of Discriminant Validity

[0122]

[0123] Subsequently, it is necessary to test the fitness of the structural equation model to verify whether the theoretical framework matches the actual data. Common indicators include the chi-square to degrees of freedom ratio (χ 2 / df), the root mean square error of approximation (RMSEA), the standardized root mean square residual (SRMR) value, the comparative fit index (CFI), and the Tucker-Lewis index (TLI) value. The test results of the model fitness are shown in Table 3. It can be seen from Table 3 that the model goodness of fit is excellent and has good effectiveness.

[0124] Table 3: Results of the Fitness Test for the Structural Equation Model

[0125]

[0126] On this basis, path analysis is carried out on the structural equation model (for example, using Mplus software), and the influence degree of each subjective psychological latent variable on the user experience evaluation is evaluated by calculating the standardized path coefficient and the significance of the P-value. The positive or negative nature of the path coefficient reveals the direction of the relationship between variables, while the magnitude of the P-value indicates whether there is a significant relationship between variables. * is the significance level, where * indicates P < 0.05, ** indicates P < 0.01, and *** indicates P < 0.001. The path relationship of the subjective psychological latent variables is shown in Table 4, Figure 3 For the visualization result.

[0127] Table 4: Path Relationship of Latent Variables

[0128]

[0129]

[0130] S150: Adjust the marketing strategy according to the influence significance.

[0131] From Table 4 and Figure 3 it can be seen that perceived usefulness (β1 = 0.35, P < 0.01), attitude (β1 = 0.196, P < 0.05), and conditional value (β1 = 0.276, P < 0.001) have a significant positive impact on usage intention. When formulating the marketing strategy, it is necessary to adopt strategies to improve the user's perceived usefulness, attitude, and conditional value.

[0132] Media exposure (β1 = -0.151, P < 0.01), perceived usefulness (β1 = 0.73, P < 0.001), and perceived ease of use (β1 = 0.413, P < 0.001) have a significant impact on attitude, which indirectly reflects that media exposure, perceived usefulness, and perceived ease of use can have an indirect effect on usage intention through the mediating role of attitude. When formulating the marketing strategy, it is necessary to pay attention to the impact of the strategy on media exposure, perceived usefulness, and perceived ease of use.

[0133] Based on the above, the present application also provides a purchase user experience evaluation system. As Figure 4 shown, the purchase user experience evaluation system includes a collection module 410, a questionnaire construction module 420, and a feedback module 430.

[0134] The collection module 410 is used to collect social media data on the purchase user experience.

[0135] The questionnaire construction module 420 is used to construct a user experience evaluation questionnaire based on social media data.

[0136] The feedback module 430 is used to take the survey results of the user experience evaluation questionnaire as market feedback data.

[0137] Preferably, the questionnaire construction module 420 includes a first analysis module 4201, an observed variable determination module 4202, and a design module 4203.

[0138] The first analysis module 4201 is used to analyze social media data to obtain multiple subjective psychological latent variables.

[0139] The observed variable determination module 4202 is used to determine multiple observed variables corresponding to the subjective psychological latent variables.

[0140] The design module 4203 is used to design a user experience evaluation questionnaire based on all the observed variables.

[0141] Preferably, in the user experience evaluation questionnaire, the observed variables are evaluated by a scale method.

[0142] Preferably, the first analysis module 4201 includes an attitude recognition module, a topic recognition module, and a latent variable determination module.

[0143] The attitude recognition module is used to recognize the attitude of car-buying users towards vehicle use based on social media data.

[0144] The topic recognition module is used to mine multiple core topics that car-buying users are concerned about based on social media data.

[0145] The latent variable determination module is used to take the latent variables corresponding to the usage intention, attitude of car-buying users, and all core topics as subjective psychological latent variables.

[0146] Preferably, the questionnaire construction module further includes a research module 4204, a judgment module 4205, and a first adjustment module 4206.

[0147] The research module 4204 is used to conduct research using the user experience evaluation questionnaire.

[0148] The judgment module 4205 is used to judge whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data based on the research results of the subjective psychological latent variables and the views of car-buying users on the vehicle.

[0149] The first adjustment module 4206 is used to adjust the observed variables corresponding to the subjective psychological latent variables when the user concerns reflected in the user experience evaluation questionnaire are inconsistent with the user concerns obtained from social media data. And,

[0150] The design module 4203 is used to redesign the user experience evaluation questionnaire according to the adjusted observed variables.

[0151] Preferably, the vehicle purchase user experience evaluation system further includes a second analysis module 440 and a second adjustment module 450.

[0152] The second analysis module 440 is used to analyze the significance of the influence between pairwise subjective psychological latent variables based on market feedback data.

[0153] The second adjustment module 450 is used to adjust the marketing strategy according to the significance of the influence.

[0154] Although some specific embodiments of the present application have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration and not for limiting the scope of the present application. Those skilled in the art should understand that the above embodiments can be modified without departing from the scope and spirit of the present application. The scope of the present application is defined by the appended claims.

Claims

1. A method for evaluating the car - purchasing user experience, characterized in that, Including: Collecting social media data on the user experience of car purchases; Constructing a user experience evaluation questionnaire based on the social media data; Taking the survey results of the user experience evaluation questionnaire as market feedback data.

2. The method for evaluating the purchase user experience according to claim 1, wherein Constructing a user experience evaluation questionnaire based on the social media data specifically includes: Analyzing the social media data to obtain multiple subjective psychological latent variables; Determining multiple observed variables corresponding to the subjective psychological latent variables; Designing a user experience evaluation questionnaire based on all the observed variables.

3. The method for evaluating the purchase user experience according to claim 2, wherein In the user experience evaluation questionnaire, the observed variables are evaluated by a scale method.

4. The method for evaluating the purchase user experience according to claim 2, wherein Analyzing the social media data to obtain multiple subjective psychological latent variables specifically includes: Identifying the attitudes of car-purchasing users towards vehicle use based on the social media data; Mining multiple core topics that car-purchasing users are concerned about based on the social media data; Taking the usage intention of car-purchasing users, the attitudes, and the latent variables corresponding to all core topics as the subjective psychological latent variables.

5. The method for evaluating the purchase user experience according to claim 2, wherein Constructing a user experience evaluation questionnaire based on the social media data further includes: Conducting a survey using the user experience evaluation questionnaire; Judging whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from the social media data based on the survey results of the subjective psychological latent variables and the views of car-purchasing users on the vehicle; If not, adjusting the observed variables corresponding to the subjective psychological latent variables and redesigning the user experience evaluation questionnaire.

6. A vehicle purchase user experience evaluation system, characterized in that, Including a collection module, a questionnaire construction module, and a feedback module; The collection module is used to collect social media data on the user experience of car purchases; The questionnaire construction module is used to construct a user experience evaluation questionnaire based on the social media data; The feedback module is used to take the survey results of the user experience evaluation questionnaire as market feedback data.

7. The vehicle purchase user experience evaluation system according to claim 6, wherein The questionnaire construction module includes a first analysis module, an observed variable determination module, and a design module; The first analysis module is used to analyze the social media data to obtain multiple subjective psychological latent variables; The observed variable determination module is used to determine multiple observed variables corresponding to the subjective psychological latent variables; The design module is used to design a user experience evaluation questionnaire based on all the observed variables.

8. The vehicle purchase user experience evaluation system according to claim 7, characterized in that In the user experience evaluation questionnaire, the observed variables are evaluated by a scale method.

9. The vehicle purchase user experience evaluation system according to claim 7, wherein The first analysis module includes an attitude recognition module, a topic recognition module, and a latent variable determination module; The attitude recognition module is used to identify the attitudes of car-purchasing users towards vehicle use based on the social media data; The topic recognition module is used to mine multiple core topics that car-purchasing users are concerned about based on the social media data; The latent variable determination module is used to take the usage intention of car-purchasing users, the attitudes, and the latent variables corresponding to all core topics as the subjective psychological latent variables.

10. The vehicle purchase user experience evaluation system according to claim 7, wherein, The questionnaire construction module further includes a survey module, a judgment module, and a first adjustment module; The survey module is used to conduct a survey using the user experience evaluation questionnaire; The judgment module is used to judge whether the user concerns reflected in the user experience evaluation questionnaire are consistent with the user concerns obtained from social media data based on the research results of subjective psychological latent variables and the views of car buyers on vehicles; The first adjustment module is used to adjust the observed variables corresponding to the subjective psychological latent variables when the user concerns reflected in the user experience evaluation questionnaire are inconsistent with the user concerns obtained from social media data; and, The design module is used to redesign the user experience evaluation questionnaire based on the adjusted observed variables.

Images