Universal method of neuromarketing

Technical field

The invention is intended to realize a method and a whole system for creating a special effect and analyzing the human effect during the sales process. Such systems make it possible to study the reaction of the person concerned to the material used for marketing purposes, usually an interactive video, and to modify it according to short- and medium-term reactions. Such a system is based on long-term research and uses the achievements of artificial intelligence (AI) systems. The system is universal and independent of the type of trade, allows to study buyers' reactions proactively and can map the consumer's reactions to emotional, affective and physiological states according to the intensity of reactions both directly (region, age group, natural cycles) and post factum - post-crisis, high emotional meaning of the event.

The aim of the invention is to create a new type of neuromarketing method and equipment for its implementation, which would allow to determine the impact of the presented marketing material on the subject, change the video content, display time and other parameters so that the expected marketing effect is as good as possible.

Neuro-marketing is a new area of ​​marketing that uses biometric methodologies to examine customer priorities and reactions to marketing initiatives and to determine why customers make different decisions. Current neural marketing techniques are not a full-fledged tool that can help real-time impactful product and service marketing campaigns that emphasize the emotional reactions of customers.

In recent years, a variety of emotional and affective metrics have been studied, including Net Emotional Value, Service Encounter Emotional Value, Hedonic Value, and Perceived Value. ) in order to develop a criterion for evaluating the customer's response and to express it formally through a numerical value.

One such criterion is Net Emotional Value (GEV), calculated from positive emotion estimates minus negative emotion estimates. When calculating GEV, the average of positive emotions (happy, satisfied, confident, valued, attentive, safe, focused, forgiving, stimulated, exploring, interested, energetic) is subtracted from the average of negative emotions (dissatisfied, overwhelmed, frustrated, frustrated, upset, , ignored, rushed) average of ratings. A higher value of the indicator corresponds to a better impact result.

Over time, neurodarketing methodologies have been supplemented by new biometric methods based on instrumental observations of pulse, skin conductivity, and pupil dilation. A new specialized field of research has emerged - biometrics, which uses various biometric tools (functional magneto-resonance imaging (fMRI), electroencephalograms, facial expressions, tone of voice, pulse, etc.). They help to identify the customer's reactions to certain stimuli classified according to the SIS International Research methodology. Our proposed method and the equipment required for its implementation allow us to perform neurodarketing using research-based methods.

Instrumental measurement data need to be processed appropriately, so this technical solution uses a special data processing technique that uses neurastrices to perform multi-criteria analysis and find the optimal assessment of the impact of marketing material. When assessing the interconnection of data and the level of intensity of their interfaces, neurocorrelation matrices are applied, which provide the highest reliability of the data connection, corresponding to the real impact of the marketing material on the research client.

State of the art

A number of technical solutions in the field of neuromarketing are currently known. They include technical solutions for the determination of physiological parameters, describe the equipment, determine the parametric equipment, but do not directly cover the proposed method and the equipment intended for its implementation.

Methods for identifying brain responses to present stimuli are provided by inventors Greičius and Menon US13153465. Their method is based on a functional brain map obtained by medical methods used in direct clinical practice. Assessing the effects on the brain according to their cognitive status allows the created map to be used for neurodarking. The specific state of cognition is recognized using a matrix of brain activity and brain activity connectivity, the method of obtaining which is not fully detailed. Such an approach is powerful and beneficial for non-clinical neurodarketing, but can be very difficult to apply in practice, brain activity detection equipment is expensive, requires qualified and expensive staff, and cannot be performed remotely.

WIPO document WO2015093926A1, invented by Ana Cristina DE LA PENA

OSORIO (2013) describes the equipment used in neurodarking and its operating processes, which include similar equipment to our proposed one, but presents data storage in the reaction matrix; the test signals are obtained using an electroencephalogram (EEG) device, an eye pupil motion sensor, an electrical conductivity measuring device, and then the obtained data are combined in that matrix, which is then used for the application of the neuromarketing method. The difference from our proposed method is data processing using local resources (PC), using more data variants, and the fact that our proposed method is based on dynamic behavioral estimation according to the Russell model, not on memorized behavioral matrices. The accumulated experience matrices used in this method require high computational power and are difficult to adapt to cyclical and targeted changes in the environment.

U.S. Pat. No. 12,835,714, invented by Hans C. Lee (2009), discloses a system for the purpose of neuromarketing based on a detailed analysis of pupil movement in the human eye and a region of interest (ROI) method. The very serious and detailed method described in the patent differs from the method and device we provide in terms of its data processing concept and the amount of data used.

U.S. Pat. No. 12,172,914 to Hans C. Breiter (2009) describes a method and apparatus for researching marketing information. This patent provides a method for evaluating marketing material based on a customer's conscious reaction at the click of a button. Such an approach is simple, effective, and inexpensive, but it does not study the unconscious human reaction, which is often more significant than the expressed conscious reaction, and it is not suitable for all personality types. The proposed method is appropriate for prioritizing one or more participants, but is fundamentally different from our proposed method in terms of both data types and information processing.

In US20140365240A1, inventor James Canton provides a system for telemedicine purposes that analyzes and compares the emotional state of a client with a database collected. The system includes data analysis, predictive analysis, near-field communication data, interactive voice response analysis, short code data analysis, and neurotransmission information used to identify feelings. Semantic analysis can be used for emotion analysis to identify priorities and personal needs. Neuromarketing research examines consumers ’neurological responses to marketing stimuli and products. Functional magnetic resonance imaging, electroencephalography, topography, and other sensors can be used to evaluate biometric information to find out how users make decisions and which part of the brain determines them. Such neuromarketing information obtained from a user can be used to compare a particular user with already available neurological profiles. The system is statically trained by updating the database, it does not work in real-time training mode, and uses an emotion database in a format not described in the patent.

In U.S. Patent Application US20160093078A1, inventor Collin Charles Davis and co-authors provide a system for creating a virtual world using images collected from a variety of sources, including those obtained using clustered resources, to create images of a virtual world. The system allows the use of images related to a specific real-world location (such as digital photos, digital frames from videos, etc.) Using image processing techniques, an artificial image of the location is created and virtual models are created. From the latter, video content is created that connects the real and virtual worlds and is streamed to the client device (s). The collection and processing of feedback data is not completely deterministic, the system is quite different from the one we offer.

U.S. Patent Application US20180063602A1 discloses a system and method for real-time context-sensitive video generation and streaming. You may receive a request from a user elsewhere to submit a personal video, select and follow the video template creation rules, including the order of the scenes, create video and audio, create frames for the first part of the video, encode them, and stream the first part user while creating frames for the next part of the video. What differs from our proposal is that the feedback is done deliberately and the real consumer reactions remain underestimated.

U.S. Pat. No. 9,913,878B2 discloses a device capable of generating video material comprising a reproducible image sequence comprising the following components: a defined function rendering module, object adaptation functions, a function timeout time module for time-limited functions, time-limited functions a sequencing module and a conversion module that applies the appropriate effects to the objects. What differs from our proposal is that the feedback is done deliberately and the real consumer reactions remain underestimated.

Some patents disclose one or more parts of a video production system that can generate dynamic, data-driven video clips using video configuration information based on user, user device, and / or specific video clip information. In additional cases, the video clip generation method also allows to obtain a dynamic, data-driven video on demand, to obtain video configuration information, to request video resources, and to generate and submit a video clip after collecting video resources (US20130125181A1).

Video data can be very useful when selling real estate (RE) information online. However, using available video data processing techniques, data users cannot efficiently and conveniently transfer and combine frames that present different perspectives and other properties of a real estate object. Among other things, such shortcomings are addressed by the invention, which proposes methods and systems that will allow better processing of image data for real estate marketing, reporting and analysis (US8307286B2).

This communication system converts the advanced services multimedia data into video, which is streamed to the user's mobile device. The system consists of the following components: a database that stores user profile data; no. mail systems in which users are stored in audiovisual format. letters; an application server for storing and retrieving data in a database and repository; a multimedia converter that receives multimedia data and converts it to a streaming video according to a selected template; and an access server that provides the multimedia data to the multimedia converter, receives the generated streaming video, and transmits it to the communication network (US7813724B2).

A video content editing support system capable of effectively modifying the content of a video clip is described in U.S. Patent Application US20050152665A1. It consists of an imaging device that writes image content data to a medium, an editing terminal, and a network that connects the imaging device and the editing terminal. The video content editing help system allows you to select the required video content data from the title information and edit it directly. This highly intelligent system differs from ours in that the content displayed is pre-selected rather than based on an instrumental customer response setting.

The virtual environment user interface described in US20100169797A1 has been supplemented with several new landmarks for novices to help them navigate in a virtual, computer-generated 3D environment and avoid common mistakes. Newcomer landmarks clearly distinguish the real environment from the virtual environment by restricting the vertical field of view of the novice avatar with application controls so that the user does not just look straight up or down, providing the user with visual information. The system differs from ours in that the feedback to the video content is active here because the user controls the system, but the unconscious level is not assessed here.

U.S. Patent No. 9,565,460B1 describes a method for modifying interactive video content based on a pre-described pattern of behavior based on information provided by social networks. This method identifies the profile of a single customer or audience and offers contextual information. It differs from our device in the form of interactivity and does not use an instrumental data collection system, but collects data from existing audience aggregated reactions.

Patent CN101946500B describes a method and system that allows an end user to embed images in video content by organically embedding them. The user then feels like he or she is in the video himself or herself, and he or she can interact with the famous people shown in that video. When creating video content and the first computer program, certain temporary templates are inserted into the video content. The end user prepares the video according to the requirements of the embedded template. The second computer program then combines the two video streams by swapping templates and inserting the end-user video into the content of the original video. The system also allows you to insert promotional information in the background, creating custom, personalized ads and product demonstration ads in real time. The technical solution presented in this patent differs from the solution we offer in terms of its data collection, processing methodology, and instrumental data collection.

Similar technical solutions are provided in JP2009131964, JP2009274085, JP2009131964, US12172914, but they differ from our proposed solution by one of the following criteria:

there is no integrated analysis of viewer emotional states, valence and excitement, viewer emotional attitudes, physiological viewer states, videos, and contextual metrics;

personalized video advertising alternatives do not use biometric and emotive computing technologies and do not create all possible options and select the most effective ones;

neuro and neurocorrelation matrices were not used in the analysis of alternatives;

utilitarian / instrumental, hedonistic, user-perceived and emotional values ​​were not identified through the integrated application of biometric and intellectual technologies.

The essence of the invention

The essence of the present invention is a neural marketing method, which obtains an integrated emotion, affective and physiological state of a research client from data determined by instrumental emotion, affective and physiological state parameters, processing information on a local computer and presenting it in Russell's emotional circle, neurocorrelation , which allows for a formal assessment of the impact of the displayed marketing video, the dynamics of the reaction in real time and the response to it.

Instrumental methods use an EEG device, an pupil analysis device, a skin electrical conductivity device. As an alternative to the verification data flow, there is a conscious response of the client to the questions formed after each video review, providing a balance between conscious and unconscious evaluation, expressed on a circular scale of emotional, affective, and physiological states. The calculated values ​​can be used immediately or stored for further processing using artificial intelligence systems.

The result of each demonstration is evaluated by a numerical quantity, which is the value of the summary criterion in the system of evaluation of emotional, affective and physiological conditions. This indicator is a quality criterion that the aim is to increase when creating an advertisement, performing its display and choosing an alternative after repeated broadcasting.

Data revealing the essence of the invention

The research developed the VINERS method, which consists of two parts: VINERS1 and VINERS2. The method of analysis and evaluation of the impact of electronic advertising (advertising of created content) (VINERS1) allows to learn more about the effectiveness of advertising at each stage of its development, helps to improve it to the most attractive option for the viewer. With the help of this method, it is possible to answer exactly which video frame had a greater effect on short-term and long-term memory, which frames evoke the strongest impressions and feelings. The intuitive method of electronic advertising broadcast (advertising of already created content) (VINERS2) allows to perform an integrated assessment of the neurobiological response of the viewers and to select the most effective version of the advertising broadcast in real time .

Such a system is based on research, the essential methodology of which is presented below:

Advertising objects are investigated using multicriteria analysis to form neuromatrices. They provide linked details about the object of the advertisement, the affective attitudes, emotional and physiological states of the viewers of the advertisement. This method makes it possible to economically accommodate an array of twelve layers of data obtained during the study, consisting of more than 400 million. anonymized data units. Using the neuromatrix method, over 60,000 moderate to strong correlations between displayed content and emotions, affective, and physiological states were found from these data, followed by neurocorrelation matrices.

On the basis of these neuro and neurocorrelation matrices and the Russell emotion model (1980), the VINERS neuro-advertising emotion model was developed, which is used to map the advertising content of viewers ’affective attitudes, emotional and physiological states. They allow for a much more accurate and real-time assessment of the effectiveness of the advertising cycle and do not detach the whole process of creation, broadcasting and delivery.

In order to perform a detailed analysis of this research object, the VINERS method was developed, which integrated the Damascus Somatic Marker Hypothesis (1994), neuro and neurocorrelation matrices, Russell emotion model (1980), biometric, physiological, affective computing, statistical analysis (LOGIT, KNN, MBP, FPRA), decision making, referral, big data and text analytics, categorical, spatial and individualized emotional content and map form analysis, neurosavian analysis, five multi-criteria analysis methods developed by the applicants.

The result of the VINERS method is a relative criterion for evaluating advertising, expressed in numbers and allowing to compare the improvement of the same advertising in the development process or to compare the effectiveness of two different advertisements.

Nevertheless, the VINERS method is used for instrumental neuro-survey.

The method is suitable for real-time negotiations. The answers obtained are examined by the proposed method. The big data method requires a time perspective, but it can also be used for subsequent analysis of the data set. The VINERS method can be used to automatically create many alternative advertising options by integrating the viewer's emotional (happy, sad, angry, scared, disgusted) and affective (boredom, interest, confusion) states, valence, arousal and physiological (facial temperature, pulse, pulse, etc.), to select the most efficient version of advertising broadcasting in real time.

Description of drawings

The hardware part of the neuromarketing research system VINERS shown in FIG. 1, consists of an investigative purchaser marked with position 1, who is subject to the study of emotional, affective, and physiological parameters during a neuromarketing process. We scan his emotional, affective, and physiological responses for the marketing information he provides with a set of sensors 2 that encompasses many emotional, affective, and physiological responses. Physiological responses include brain electromagnetic wave (EEG) sensor, pupil response, body temperature and its dynamics, sweating rate, and many other known sensors of physiological responses. These sensors, with their data processing equipment, provide the data to a data bus 3, which transmits the data directly to a data acquisition and processing device (data acquisition) 4. It normalizes the data and prepares it for a computer or computer cloud 5. The bus connection between the data processing device and the computer can be standard (USB, Eth, WiFi) or specialized via an additional data channel for large amounts of data. The data is processed, stored and presented to the operator 6 in real time on a computer or in the cloud 5 for on-screen display or further data analysis.

The data processing of the neuromarketing system VINERS, the hardware of which is shown in Fig. 1, is shown in Figs. 2. Data processing is performed from the data stream 7, which is presented as data in time 8, and transmitted through the computer interface to the processing computer 5, and a time matrix of emotion, affective and physiological state and impact correlations 9 is formed according to the given time discrete. emotional, affective, and physiological state reactions. Such neuromatrices can be formed once per session or with a given time discrete to show a dynamic change in human responses during a sales session. We then transform the significance of the data into a neurocorrelation matrix 10, which provides the significance of the human response required for the final estimate. The data are further transmitted for processing and, after calculations, are expressed in a user-friendly form, ie the Russell diagram 11. This compact and receptive form shows the customer’s behavior, his motivational stimuli, and reveals a map of the emotional, affective, and physiological states, as well as the neuromatrix needed for a successful sales process.

Description of the implementation of the invention

The proposed structural and principal scheme of VINERS emotion, affective and physiological state processing system for neuromarketing (Fig. 1, 2) is realized by connecting sensors and neuroreactive sensors (emotional and affective state and physiological state data) to a personal computer connected to the general computer network. transmitted via data acquisition boards or directly via universal USB interfaces for further processing by unique software. The computer interface to large data systems, if required, is connected to a universal computer network.