NeurAssio - An EEG based system for advertisement impact assessment

Project Title

NeurAssio - An EEG based system for advertisement impact assessment

Project Area of Specialization NeuroTechProject Summary

Advertising is very important for marketing purposes to maintain their advertising quality. This can be done by knowing the persons genuine feedback about the advertisement. In this project we will use a low cost EEG headset for collecting the brain signals of the subjects, viewing the advertisements, that aims in providing the statistics of the advertisement’s impact and the likelihood of a person purchasing the advertised product. By the analysis of brain signals using the latest methods we will be able to distinguish between di?erent brain waves like alpha, beta, gamma and delta each of them being speci?c to a certain kind of activity.This analysis will help us provide the statistical trend for a certain kind of advertisement on the basis of various factors. This will greatly help the marketing sector of multinational companies to improve their advertising campaign which will bring a revolutionary change in how they perceive their brand.

Project Objectives

INDUSTRY OBJECTIVES

• The system must allow the analysts to ?ll in the gaps left by traditional marketing methods.
• It must help the marketing analysts understand the consumer preferences in order to improve brand quality, advertisements, packaging etc.
• Our project should focus on the study of brain responses to marketing stimuli which will revolutionize marketing sector of industries.
• It must be so economical causing the industries to save on their marketing expenses by recording consumers response in an e?ective and e?cient way.
• System must cut o? losses of industries in marketing sector, by making them predict consumers response using advanced techniques accurately.
• Main aim of our project is to develop a system that is cost e?ective and error free than the previously designed systems which will eventually lead the industrial ?eld to progress
• It must help the industrial sector to make necessary changes wherever required in the product or advertisement beforehand by analyzing consumers behavior.

RESEARCH OBJECTIVES

• The main objective is to acquire the brain signals in a CSV ?le from the headset to have a closer look at the signals attained and to understand them better.
• Projects research is based on correct distinction between di?erent brain signals, representing di?erent emotions and behavior. So initially the research is based on the understanding di?erent signals connected with di?erent emotions. In order to draw conclusions better.
• The toughest part is the detection and classification of brain signals into di?erent emotions using di?erent classi?ers, whichever is more accurate in prediction.
• There should be a thorough research, so that the developers must be able to discover new techniques which are more e?cient and faster than previous ones.
• Techniques that are not yet used must be preferred. Such techniques must be understood properly and used in such a way that they help in the development of an e?cient and intelligent system that is able to analyze human emotions more accurately than older systems.

ACADEMIC OBJECTIVES

• The main goal of this project is to allow the developers learn major areas of computer ?eld more specifically machine learning and signal processing.
• Developers must learn problem solving techniques in order to solve real world problems that they encounter in their career.
• The developers should be exposed to new technologies which help them connect with the technological world in a better way.
• The developers must implement the test and trial method. They must test di?erent techniques and use those that are more suitable for the project. This will also help them in their future works.
• Practical performance must be worked upon more and more.
• The developers must practice risk and change management in the course of their project development. Also, they would be able to polish their decisive power.

Project Implementation Method

It is basically a desktop application in unity which is implemented in Csharp and Python. The methodology of the app is as follows:

DATA ACQUISITION

The EEG signals are acquired from each subject using the Neurosky Mindwave Mobile 2 headset. The subject is shown a video while wearing the EEG headset and the brain signals are acquired during this process. These acquired signals are then saved in a CSV file. 

The data collection session schedule is specified as follows:

• Number of Ads: 10 videos of 4 categories.
• Categories: Smartphones, Automobiles, Beverages, Food.
• No. of subjects: 9 subjects (4 males and 5 females), age (18-25 years).
• Session Duration: 30 minutes per subject.

The CSV file of raw EEG data obtained, is as follows:

 

Each subject is shown 10 videos of all 4 categories. They are also asked to fill in a questionnaire that is then used to label each file that is obtained.

DATA PREPROCESSING

The signals in the CSV file are then preprocessed. The redundancy is removed and it is normalized.

FEATURE EXTRACTION

The CSV file is then visualized by plotting each signal in each file. On the basis of this visualization the most important features are selected for classification and prediction.

CLASSIFICATION

The CSV file is then passed through 3 classifiers. The classifiers then classify the signals into two classes. It determines whether the user likes the advertisement or not.

Three classifiers are as follows:

• • SVM (Support Vector Machine) 
  • LR (Linear Regression)
  • DT (Decision Tree)

The classifier predicts the user's likeness or dislike ness. The result  is then shown graphically in unity, also showing the percentage to which the person liked or disliked the video(advertisement.)

Benefits of the Project

MARKETING SECTOR
The proposed system will help the companies to improve their promotional advertisement video and to make it more attractive and productive by giving them response of di?erent customers. Therefore, it will eventually upgrade their marketing.

ADVERTISING INDUSTRY
As it is the global industry of public relations and marketing companies, when the proposed system will enhance the advertisements to provide e?ective information about products and services in an e?cient manner, it will be bene?cial for the advertising industry as well.

INCREASE IN ECONOMY
In a country where consumer spending determines the long run of the economy advertising motivates them to pay more. So, by playing a positive role in both marketing sector and advertising industry , which provides goods, services and jobs, proposed system will deep down increase the economy of the country.

ENTERTAINMENT INDUSTRY
As the system gives the feedback on a video, it can be used in entertainment industry as well to enhance the music videos, short movies, di?erent parts of movies or dramas etc. according to the viewers taste before releasing them.

EDUCATION SECTOR
By getting the students feedback on online lectures, proposed system can be used to upgrade the teaching methods accordingly.

RESEARCH IN NEUROMARKETING
In the proposed system, neuromarketing (the use of modern brain science to measure the impact of marketing and advertising on consumers) will be used. So, a person doing research in this ?eld can use the system.

OTHER
With some advancements if the proposed system is trained to work without the video then it can be used in the classrooms to get the students feedback about teachers or the classroom environment to improve education system and also to enhance a companys environment by getting employees feedback etc.

Technical Details of Final Deliverable

• Efficient system for providing an authentic review of a product
• For improving the quality of product according to customer intrests
• For determining the user engagement and excitement level while watching an advertisement
• For finding resulting short advertisements with high imapct which will help the companies for reducing their marketing cost
• For predicting the sales of a product before its launch in market

Final Deliverable of the Project Software SystemCore Industry MediaOther Industries Education Core Technology NeuroTechOther Technologies Artificial Intelligence(AI)Sustainable Development Goals Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Partnerships to achieve the GoalRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	29000
Nuerosky Mindwave Mobile 2	Equipment	1	24000	24000
Recharable cell	Equipment	10	500	5000