A method and system for measuring consumer purchase intent based on EEG technology

Abstract

The application discloses a kind of consumer purchase willingness measurement method, system based on EEG technology, comprising: EEG signal acquisition equipment gathers the EEG signal generated after receiving stimulus, the original signal collected is introduced into EEGLAB module in Matlab and is preprocessed, then is transmitted to data analysis system;Data analysis system carries out the extraction of electroencephalogram, and the electroencephalogram is converted into the data that can be statistically analyzed, and then the data is transmitted to discriminant identification system;Discriminant identification system carries out consumer purchase willingness analysis by KNN algorithm.The application first extracts the electroencephalogram of consumer frontal lobe part by EEG equipment, then extracts characteristic value, and uses KNN machine learning model to detect the change of consumer purchase willingness, the application can intelligently detect the change of purchase willingness when consumer receives different external stimulus, improve the efficiency and accuracy of analysis.

Description

Technical Field

[0001] This invention relates to the field of EEG signal measurement technology, and in particular to a method and system for measuring consumer purchase intention based on EEG technology. Background Technology

[0002] According to data from the 51st Statistical Report on Internet Development in China published by CNNIC, as of December 2022, the number of online shopping users in my country reached 845 million, accounting for 79.2% of all internet users; online retail sales reached 13.79 trillion yuan, a year-on-year increase of 4.0%. Online retail has become an important force driving consumption expansion. The emergence of online shopping has greatly simplified the transaction process, allowing consumers to complete transactions in a short period of time. Numerous studies have also shown that online shopping influences consumers' purchasing intentions; however, most of these studies are based on traditional questionnaires and interviews. These methods rely on post-hoc recall, which introduces subjectivity and can lead to biases and inaccuracies in the results. Currently, there are no automated devices or methods on the market that can intelligently analyze brainwave signals to determine consumers' purchasing intentions.

[0003] Therefore, improving the accuracy of measurements has become an urgent problem to be solved. Summary of the Invention

[0004] To overcome the aforementioned shortcomings of existing technologies, this invention provides a method and system for measuring consumer purchase intention based on EEG technology, thereby enabling objective analysis of electroencephalogram (EEG) signals to identify changes in consumer purchase intention and improve measurement accuracy.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for measuring consumer purchase intention based on EEG technology, comprising:

[0006] The EEG signal acquisition device acquires the EEG signals generated after receiving stimulation, imports the acquired raw signals into the EEGLAB module in Matlab for data preprocessing, and then transmits them to the data analysis system.

[0007] The data analysis system extracts EEG signals, converts them into data that can be statistically analyzed, and then transmits the data to the discrimination and recognition system.

[0008] The discrimination and identification system uses the K Nearest Neighbor (KNN) algorithm to analyze consumer purchase intentions.

[0009] Preferably, the preprocessing includes removing useless electrodes, rereferencing, filtering, resampling, segmentation and baseline correction, independent component analysis and / or removing bad segments.

[0010] Preferably, the removal of useless electrodes refers to the removal of active electrodes except for reference and special purposes.

[0011] Preferably, the rereference refers to selecting bilateral mastoid references before filtering.

[0012] Preferably, filtering refers to setting a 0.5Hz high-pass filter to filter out noise below this lower limit; setting a 30Hz low-pass filter to filter out noise above this upper limit; setting a 49Hz to 51Hz dip filter to filter out noise within this range; and setting a 0.5-30Hz band-pass filter to filter out signals that pass through this upper and lower limit.

[0013] Preferably, resampling is not mandatory; simply input the desired sampling rate in the software.

[0014] Preferably, segmentation is necessary because the experiment requires recording consumers' EEG data throughout the entire process, but what we really need are the EEG signals generated when consumers receive external stimuli. Therefore, it is necessary to mark the entire data segment according to the requirements and segment the data.

[0015] Preferably, baseline correction is used to eliminate EEG noise caused by spontaneous EEG activity, with data before time 0 serving as the baseline. Assuming that the EEG signal before time 0 represents spontaneous EEG activity upon receiving stimulation, subtracting the average of the data before time 0 from the data after time 0 can eliminate some of the spontaneous EEG noise.

[0016] Preferably, the bad segments are removed using spherical interpolation, and the number of bad segments removed cannot exceed 10% of the total.

[0017] Preferably, EEG signals are a mixture of source signals and noise. Independent component analysis (ICA) decomposes each data point into individual components through inverse operations. Artifacts, such as blinking and muscle tension, are then removed to obtain relatively clean data. Processing steps:

[0018] (1) Preprocessing of EEG signal data by mean reduction and normalization;

[0019] (2) Determine the number of independent components;

[0020] (3) Process the data using maximum likelihood estimation;

[0021] (4) The mixed signal is converted into a set of independent signals by the transformation matrix;

[0022] (5) Estimate the mixture matrix A;

[0023] (6) Multiply the independent component signals by the inverse mixing matrix to achieve signal recovery;

[0024] (7) Interpret the amplitude, frequency, etc. of the signal through statistical analysis.

[0025] Preferably, the characteristics of the EEG signal include time-domain characteristics, frequency-domain characteristics, and time-frequency characteristics.

[0026] Preferably, the time-domain feature refers to the distribution of EEG signals in a certain region over time, and its parameters include mean, variance, standard deviation and peak value.

[0027] F1 = max(X) i (n));

[0028] F2=min(X i (n));

[0029]

[0030]

[0031]

[0032] F6 = F1 - F2;

[0033] Where max is the function for finding the maximum value; min is the function for finding the minimum value; F3 is the mean, F4 is the variance, F5 is the standard deviation, and F6 is the peak value; N is the total number of sampling points; X i (n) represents the amplitude of the nth sampling point.

[0034] Preferably, the frequency domain features refer to the distribution of EEG signals in a certain region as frequency changes, and the power spectral density estimate is obtained through short-time Fourier transform, windowing, and other methods.

[0035] Preferably, the frequency domain feature extraction uses the Welch algorithm for power spectral density estimation, and the steps are as follows:

[0036] (1) Given a time-domain signal x(n) of length N, divide the signal into L segments, each segment having a length of M, i.e.: N = LM;

[0037] (2) Apply a window function (hamming window) and FFT to the signal of each segment to obtain the frequency domain representation of each segment:

[0038] (3) Perform amplitude squaring on the frequency domain representation to obtain the power spectral density estimate for each segment: I i (w)=X i (w) 2 ;

[0039] (4) Average the power spectral density estimates of all segments to obtain the average power spectral density estimate of the signal:

[0040] Preferably, the time-frequency features are time and frequency information obtained by wavelet transform, combining time-domain and frequency-domain features.

[0041] Preferably, the time-frequency feature extraction steps are as follows:

[0042] (1) Convert the time-domain signal using discrete wavelet transform:

[0043]

[0044] Where a is the proportional parameter and b is the displacement parameter. For mother and child waves.

[0045] (2) After filtering, the approximation coefficients and detail coefficients are obtained:

[0046]

[0047]

[0048] (3) After wavelet energy decomposition, the following calculations are performed:

[0049] E(l)=∑ n |C l [n]| 2 ;

[0050] Where L represents the number of decomposition layers, and C[n] represents the approximate coefficients or detail coefficients of the wavelet coefficients in the l-th layer;

[0051] (4) Wavelet mean value processing:

[0052]

[0053] Where N represents the total number of approximation coefficients or detail coefficients;

[0054] (5) Wavelet normalization processing:

[0055]

[0056] Where M represents the wavelet decomposition order, and the sum of P(l) is 1;

[0057] (6) Calculate wavelet entropy:

[0058]

[0059] Preferably, the KNN algorithm is used to train a classification model and output consumer purchase intention analysis results.

[0060] A consumer purchase intention measurement system based on EEG technology, used to implement the method, includes:

[0061] EEG signal acquisition equipment is used to collect consumers' electroencephalogram (EEG) signals, and transmit the preprocessed EEG signals to the data analysis system.

[0062] The data analysis system extracts features from EEG signals and then transmits the data to the discrimination and recognition system.

[0063] A discriminant identification system is used to detect changes in consumer purchase intentions using the KNN algorithm.

[0064] Compared with the prior art, the beneficial effects of the present invention are:

[0065] This invention provides a method and system for measuring consumer purchase intention based on EEG technology, so as to objectively analyze brain signals to determine changes in consumer purchase intention and improve the accuracy of measurement results.

[0066] This invention provides a method and system for measuring consumer purchase intention based on EEG technology. It collects electroencephalogram (EEG) signals from the frontal lobe of the consumer's cerebral cortex using an EEG device and extracts feature values. A KNN machine learning model is then used to detect changes in EEG signals in the frontal lobe of the cerebral cortex, which is responsible for decision-making, when the consumer receives external stimuli.

[0067] This invention uses artificial intelligence algorithms to automatically monitor changes in consumer purchasing intentions in real time, and its portable EEG device makes it convenient for researchers or marketing personnel to use, thus improving detection efficiency.

[0068] This invention first collects EEG signals from the frontal lobe of consumers using an EEG device, then extracts feature values, and uses a KNN machine learning model to detect changes in consumers' purchasing intentions. This invention can intelligently detect changes in purchasing intentions when consumers receive different external stimuli, improving the efficiency and accuracy of the analysis. Attached Figure Description

[0069] Figure 1 A flowchart illustrating the specific implementation of a consumer purchase intention measurement method based on EEG technology provided by the present invention;

[0070] Figure 2 This is a schematic diagram of a consumer purchase intention measurement system based on EEG technology provided by the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0072] The core of this invention is to provide a method and system for measuring consumer purchase intention based on EEG technology, so as to objectively analyze brain signals to determine changes in consumer purchase intention and improve the accuracy of measurement results.

[0073] A method for measuring consumer purchase intention based on EEG technology includes:

[0074] The EEG signal acquisition device acquires the EEG signals generated after receiving stimulation, imports the acquired raw signals into the EEGLAB module in Matlab for data preprocessing, and then transmits them to the data analysis system.

[0075] The data analysis system extracts EEG signals, converts them into data that can be statistically analyzed, and then transmits the data to the discrimination and recognition system.

[0076] The discrimination and identification system uses the K Nearest Neighbor (KNN) algorithm to analyze consumer purchase intentions.

[0077] To acquire high-quality EEG signals, the electrodes were placed with the nasal root and occipital protuberance as reference points front-to-back, the external auditory canals as reference points left-to-right, and the center of the top of the head as the starting point. After the test begins, a evoked signal is given, and the EEG device automatically records the consumer's evoked EEG. Offline analysis is used during the acquisition process; the test is stopped if large areas of artifacts are observed.

[0078] The preprocessing includes removing useless electrodes, rereferencing, filtering, resampling, segmentation and baseline correction, independent component analysis and / or removing bad segments.

[0079] The removal of useless electrodes refers to the removal of active electrodes except for reference and special purposes.

[0080] The term "rereference" refers to selecting bilateral mastoid references before filtering.

[0081] Filtering refers to setting a 0.5Hz high-pass filter to filter out noise below this lower limit; setting a 30Hz low-pass filter to filter out noise above this upper limit; setting a 49Hz to 51Hz dip filter to filter out noise within this range; and setting a 0.5-30Hz band-pass filter to filter out signals that pass through this upper and lower limit.

[0082] Resampling is not mandatory; simply input the desired sampling rate into the software.

[0083] Segmentation is necessary because the experiment requires recording consumers' EEG data throughout the process, but what we really need are the EEG signals generated when consumers receive external stimuli. Therefore, it is necessary to mark the entire data segment and segment the data according to the requirements.

[0084] Baseline correction aims to eliminate EEG noise caused by spontaneous EEG activity, using data before time 0 as the baseline. Assuming that the EEG signals before time 0 represent spontaneous EEG activity upon receiving stimulation, subtracting the average of the data before time 0 from the data after time 0 can eliminate some of the spontaneous EEG noise.

[0085] Bad segments are removed using spherical interpolation, and the number of bad segments removed cannot exceed 10% of the total.

[0086] EEG signals are composed of a mixture of source signal and noise. Independent component analysis (ICA) decomposes each data point into its constituent components through inverse operations. Artifacts, such as blinking and muscle tension, are then removed to obtain relatively clean data. Processing steps:

[0087] (1) Preprocessing of EEG signal data by mean reduction and normalization;

[0088] (2) Determine the number of independent components;

[0089] (3) Process the data using maximum likelihood estimation;

[0090] (4) The mixed signal is converted into a set of independent signals by the transformation matrix;

[0091] (5) Estimate the mixture matrix A;

[0092] (6) Multiply the independent component signals by the inverse mixing matrix to achieve signal recovery;

[0093] (7) Interpret the amplitude, frequency, etc. of the signal through statistical analysis.

[0094] The characteristics of the EEG signal include time-domain characteristics, frequency-domain characteristics, and time-frequency characteristics.

[0095] The time-domain features refer to the distribution of EEG signals in a certain region over time, and their parameters include mean, variance, standard deviation, and peak value.

[0096] F1 = max(X) i (n));

[0097] F2=min(X i (n));

[0098]

[0099]

[0100]

[0101] F6 = F1 - F2;

[0102] Where max is the function for finding the maximum value; min is the function for finding the minimum value; F3 is the mean, F4 is the variance, F5 is the standard deviation, and F6 is the peak value; N is the total number of sampling points; X i (n) represents the amplitude of the nth sampling point.

[0103] The frequency domain features refer to the distribution of EEG signals in a certain region as frequency changes. The power spectral density estimate is obtained through short-time Fourier transform, windowing, and other methods.

[0104] The frequency domain feature extraction employs the Welch algorithm for power spectral density estimation, and the steps are as follows:

[0105] (1) Given a time-domain signal x(n) of length N, divide the signal into L segments, each segment having a length of M, i.e.: N = LM;

[0106] (2) Apply a window function (hamming window) and FFT to the signal of each segment to obtain the frequency domain representation of each segment:

[0107] (3) Perform amplitude squaring on the frequency domain representation to obtain the power spectral density estimate for each segment: I i (w)=X i (w) 2 ;

[0108] (4) Average the power spectral density estimates of all segments to obtain the average power spectral density estimate of the signal:

[0109] The time-frequency features are information in time and frequency obtained by wavelet transform, combining time-domain and frequency-domain features.

[0110] The time-frequency feature extraction steps are as follows:

[0111] (1) Convert the time-domain signal using discrete wavelet transform:

[0112]

[0113] Where a is the proportional parameter and b is the displacement parameter. For mother and child waves.

[0114] (2) After filtering, the approximation coefficients and detail coefficients are obtained:

[0115]

[0116]

[0117] (3) After wavelet energy decomposition, the following calculations are performed:

[0118] E(l)=∑ n |C l [n]| 2 ;

[0119] Where L represents the number of decomposition layers, and C[n] represents the approximate coefficients or detail coefficients of the wavelet coefficients in the l-th layer;

[0120] (4) Wavelet mean value processing:

[0121]

[0122] Where N represents the total number of approximation coefficients or detail coefficients;

[0123] (5) Wavelet normalization processing:

[0124]

[0125] Where M represents the wavelet decomposition order, and the sum of P(l) is 1;

[0126] (6) Calculate wavelet entropy:

[0127]

[0128] The KNN algorithm is used to train a classification model and output consumer purchase intention analysis results.

[0129] A consumer purchase intention measurement system based on EEG technology, used to implement the method, includes:

[0130] EEG signal acquisition equipment is used to collect consumers' electroencephalogram (EEG) signals, and transmit the preprocessed EEG signals to the data analysis system.

[0131] The data analysis system extracts features from EEG signals and then transmits the data to the discrimination and recognition system.

[0132] A discriminant identification system is used to detect changes in consumer purchase intentions using the KNN algorithm.

[0133] Example 1:

[0134] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a specific implementation of a consumer purchase intention measurement method based on EEG technology provided by this invention. The frontal lobe is the region where most conscious thoughts and decisions occur; therefore, in this embodiment, the electrodes of the EEG signal acquisition device are mainly distributed in the frontal lobe, with the electrode positions symmetrically distributed left and right. The EEG signal acquisition device collects the electroencephalogram (EEG) signals of the consumer when receiving image or video stimuli.

[0135] The measurement method used in this embodiment specifically includes the following process:

[0136] 1. When subjects receive external stimuli, electrodes are placed in the frontal lobe of the cerebral cortex to record electroencephalograms (EEGs) of neuronal firing phenomena over a period of time. The EEG signals are then preprocessed and transmitted to a data analysis system.

[0137] In the preprocessing steps: re-reference is performed by selecting an average reference, which eliminates errors caused by changes in the reference electrode potential in the original recording; filtering is used to eliminate 50Hz ambient noise and some high-frequency or low-frequency noise. Therefore, this embodiment designs a 0.5Hz high-pass filter, a 30Hz low-pass filter, and a 49Hz to 51Hz concave filter; to avoid the original data sampling frequency being unsuitable for the experiment, this embodiment requires resampling, with the resampling frequency being twice that of the analysis band; then, target data segments are selected according to the requirements of data analysis; to prevent baseline differences due to data instability, this embodiment also requires baseline correction to keep them on the same level; and independent component analysis is used to remove electrooculography artifacts. After the above preprocessing, a clean signal related to the experimental stimulus can be obtained.

[0138] 2. The data analysis system extracts relevant features of the EEG signal according to the needs of the experiment, and then transmits them to the judgment and recognition system.

[0139] The features extracted from the EEG signal include six features in three categories: time-domain features, frequency-domain features, and time-frequency features. Time-domain features utilize mean, variance, standard deviation, and peak value; frequency-domain features utilize power spectral density estimates. The time-frequency features are obtained through wavelet transform to obtain wavelet entropy measurements.

[0140] In this embodiment, an eight-channel device was used to collect EEG signals, that is, the EEG signals of eight channels were extracted, and feature extraction was performed on the EEG signals of each channel, resulting in a total of 48 features.

[0141] 3. The system uses the KNN algorithm to train a classification model to identify changes in consumer purchasing intentions.

[0142] The KNN algorithm is a classic and relatively easy-to-understand algorithm in machine learning. This algorithm describes a sample's class as follows: if the majority of the K most similar samples in the feature space belong to a certain class, then that sample also belongs to that class. However, when choosing the value of K, care must be taken: a K value that is too small means the overall model becomes too complex, making it prone to overfitting; while a K value that is too large means the model becomes too simple, leading to prediction errors.

[0143] This invention provides a method and system for measuring consumer purchase intention based on EEG technology. It collects electroencephalogram (EEG) signals from the frontal lobe of the consumer's cerebral cortex using an EEG device and extracts feature values. A KNN machine learning model is then used to detect changes in EEG signals in the frontal lobe of the cerebral cortex, which is responsible for decision-making, when the consumer receives external stimuli.

[0144] In detail, the cerebral cortex can be divided into the frontal lobe, temporal lobe, parietal lobe, and occipital lobe. The electrical signals generated by different areas represent different functions. The frontal lobe is where most conscious thoughts and decisions are made; the temporal lobe is responsible for processing auditory information and is also related to memory and emotion; the parietal lobe is responsible for integrating information from the outside world as well as information from internal sensations such as those from skeletal muscles, limbs, head, eyes, and ears; and the occipital lobe is responsible for visual processing.

[0145] This invention can automatically monitor changes in consumer purchasing intentions in real time through artificial intelligence algorithms, and its portable EEG device makes it convenient for researchers or marketing personnel to use, thus improving detection efficiency.

[0146] Example 2:

[0147] Please refer to Figure 2 , Figure 2 This is a schematic diagram of a system for measuring changes in consumer purchase intention based on EEG (electroencephalography) provided by the present invention. This system is used to implement the above-mentioned method and includes:

[0148] EEG signal acquisition device 101 is used to acquire consumers' electroencephalogram (EEG) signals, and transmit the preprocessed EEG signals to the data analysis system.

[0149] The data analysis system 102 extracts the features of the EEG signal and then transmits the data to the discrimination and recognition system.

[0150] The discrimination and recognition system 103 is used to detect changes in consumer purchase intentions using the KNN algorithm.

[0151] As can be seen, the system first collects EEG signals from the frontal lobe of consumers using an EEG device, then extracts feature values, and uses a KNN machine learning model to detect changes in consumers' purchasing intentions. This invention can intelligently detect changes in purchasing intentions when consumers receive different external stimuli, improving the efficiency and accuracy of the analysis.

[0152] For a description of the identification system provided by this invention, please refer to the foregoing embodiment of measuring changes in consumer purchase intention based on EEG devices. The embodiments of this invention will not be repeated here. The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. As the system disclosed in the embodiments corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0153] The foregoing has provided a detailed description of the consumer purchase intention measurement method and system based on EEG technology provided by this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of this invention. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this invention.

[0154] The functions described herein can be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions can be stored as one or more instructions or codes on or transmitted via a computer-readable medium. Other examples and embodiments are within the scope and spirit of this invention and the appended claims. For example, due to the nature of software, the functions described above can be implemented using software executed by a processor, hardware, firmware, hardwired, or any combination thereof. Furthermore, the functional units can be integrated into a single processing unit, or each unit can exist physically separately, or two or more units can be integrated into a single unit.

[0155] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0156] The units described as separate components may or may not be physically separate. Similarly, the components of the control device may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0157] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

Claims

1. A method for measuring consumer purchase intent based on EEG technology, characterized in that: The application relates to a system for analyzing consumer purchase willingness, which comprises the following steps: An EEG signal acquisition device is used to collect EEG signals generated after stimulation, the collected original signals are introduced into an EEGLAB module in Matlab for data preprocessing, and then the data are transmitted to a data analysis system; The data analysis system extracts EEG signals, converts the EEG signals into data that can be statistically analyzed, and then transmits the data to a discriminant recognition system; The discriminant recognition system analyzes consumer purchase willingness through a K Nearest Neighbor (KNN) algorithm; The preprocessing comprises the following steps: rejecting useless electrodes, re-referencing, filtering, resampling, segmenting and baseline correction, independent component analysis and / or rejecting bad segments; The independent component analysis decomposes data of each data point through inverse operation, removes artifact components, and obtains clean data, and the processing steps comprise the following steps: (1) preprocessing of mean removal and normalization of EEG signal data; (2) determining the number of independent components; (3) processing the data through maximum likelihood estimation; (4) converting mixed signals into a group of independent signals through a transformation matrix; (5) estimating a mixing matrix A; (6) multiplying the independent component signals by the inverse mixing matrix to realize signal recovery; (7) explaining the amplitude and frequency of the signals through statistical analysis; The characteristics of the EEG signals comprise time domain characteristics, frequency domain characteristics and time-frequency characteristics. The time-frequency characteristics are information on time and frequency obtained through wavelet transformation by combining time domain and frequency domain characteristics.

2. The method of measuring consumer purchase intention based on EEG technology according to claim 1, characterized in that: The time domain characteristics refer to the distribution of the EEG signals in a certain region with time, and the parameters comprise mean, variance, standard deviation and peak value.

3. The method of measuring consumer purchase intent based on EEG technology according to claim 2, characterized in that: The frequency domain characteristics refer to the distribution of the EEG signals in a certain region with frequency, and the power spectral density estimation value is obtained through short-time Fourier transformation and windowing.

4. The method of measuring consumer purchase intent based on EEG technology according to claim 3, characterized in that: The frequency domain characteristic extraction adopts a Welch algorithm to perform power spectral density estimation.

5. The method of measuring consumer purchase intent based on EEG technology according to claim 1, wherein: The KNN algorithm is used to train a classification model and output the consumer purchase willingness analysis result.

6. A system for measuring consumer purchase intention based on EEG technology, for implementing the method for measuring consumer purchase intention based on EEG technology according to any one of claims 1-5, characterized in that: The application also relates to a system for analyzing consumer purchase willingness, which comprises the following steps: An EEG signal acquisition device is used to collect EEG signals of consumers, and the EEG signals are transmitted to a data analysis system after preprocessing; The data analysis system is used to extract characteristics of the EEG signals, and then transmits the data to a discriminant recognition system; The discriminant recognition system is used to detect changes in consumer purchase willingness through a KNN algorithm.

Images