EMINER: An Android Application for Emotion Detection using AI Techniques

Project Title

EMINER: An Android Application for Emotion Detection using AI Techniques

Project Area of Specialization Computer ScienceProject Summary

Emotions are the essence of what makes us human. In educational institutes students have different emotions like no interest in studies, family problems, and loose relations with peers, and inferiority complexion which drastically affect their progress. In order to know their behavior and solve their problems, there is a need for an application that can be used throughout the semester to automatically detect students emotions. Emotion Miner is one of such applications that recognize students' emotions by detecting their facial expressions which helps in understanding their problems. The general workflow of the proposed project is that it takes an image as input, applying the pre-processing technique on this input image, then executing an emotion detection algorithm, and finally matching it with a data set that generates a result. The proposed model is an Android application which will be developed using the Scrum framework of agile methodology.

Project Objectives

There is rapid progress in IT related to the medical field, everyone is focused to address medical issues. There is a need to address the sentiment analysis of students' behavior, which detects what makes them failing to achieve their goal. That’s the reason of this project to address the students' emotional issues and give suggestions on time. The main objectives of this project are:

• Detect student’s emotions and generate reports about their emotional state for a specific period.
• Based on the report, an analysis of the previous history could be useful to know about their health status.
• The teacher can give feedback to students and can motivate them if required.
• The proposed application revolves around the idea of providing benefit to teachers and students by giving them the facility of an in-hand device that can be used anytime and anywhere.

Project Implementation Method

Android Studio will be used to build the application.  With Android Studio, Java classes are used as models. For image processing, Google Vision API and Scikit-image library will be deployed. The application will match the input image with any image dataset preferably the Extended Cohn Kanade dataset. The well-known Convolutional Neural Network algorithm will be used for emotion detection and further image processing. The report will be generated in SQL server 2018.

We are using Agile’s scrum methodology which consists of the following phases:

1. Product Backlog

This one is the most crucial part of Scrum, here we gather information about the system from different sources such as from the internet, research reports, stakeholders, etc. Here the information kept in the backlog and the new changes could be incorporated in this section. For example, what are the issues in existing apps such as are they generating emotions accurately? How much processing time these applications take?

2. Sprints

In this section, the development has been divided into different modules and each module will be assigned a time. In that time span that module should be ready. That is called sprints.

3. Sprints Planning

 In this section, requirements gathered in the product backlog will be utilized and according to that, the tasks will be prioritized. The team lead decides and plans which functionality will develop when.

4. Sprints Execution

Now planning of each sprint has been done. In this section, according to the planning, each sprint execution incorporated.

5. Daily Scrum

Daily the team schedules a meeting where they explain how much work has been done and how much left behind etc.

6. Sprint Review

In this section, the team reviews the sprint which is being developed, to check either it’s meeting the requirement or not. This inspection conduct in between the development of the application. Such as either the application is showing a valid result or not etc.

7. Sprint Retrospective

This is the inspection conducted at the end of the sprint. To check whether the desired feature or functionality is achieved or not? Like whether the application is accurately providing search results or showing errors.

Benefits of the Project

• The application will be used by teachers and students, they can use it anytime anywhere.
• Emotion detection allows teachers to know about the student’s emotional state in advance so that immediate assistance can be provided.
• The proposed application would generate a report which helps teachers in the assessment of student’s behavior and attitude.
• Our application is envisioned to be used in academic institutions, so they can assess student’s mental health on time and can guide them so that in the future it wouldn’t lead them to major diseases such as depression, etc.

Technical Details of Final Deliverable

The final deliverable of the EMINER will be an android based system along with a documentation manual having all technical details of the application. It will be developed in Java along with the SQLite database to store and retrieve data. We will develop a python's API that we will integrate with the android application. The application will be compatible with android API version 21 and above. Laptop and Android phone will be used to install Android Studio and run the application respectively. OTP gateway will be used to generate a one-time password.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries Education Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Quality EducationRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Document Printing	Miscellaneous	1	5000	5000
OTP Service (10,000 sms)	Miscellaneous	1	5000	5000
Android Phone for application testing	Equipment	1	30000	30000
IP Camera	Equipment	1	20000	20000
ADATA 2TB Portable Hard Drive	Equipment	1	20000	20000
-	Equipment	0	0	0