Have you ever felt the need of quote that inspires you keep going forward with your life or simply to make you feel better?
Here's InspireMe for you!
InspireMe is a python-based application made using Kivy that provides motivational quotes to the user depending upon their mood by analyzing their written text using Machine Learning Algorithm.
- Tech Stack
- Features
- Application Overview
- ML Algorithm
- Challenges Faced
- Future Scope
- Learnings
- Contributers
Language-: Python
Framework-: Kivy
Kivy is a free and open source Python framework for developing mobile apps and other multitouch application software with a natural user interface.
- The app offers the firstly, the basic feature for the user to sign-in and register.
- After successfully logging in, the user is taken to a screen where they can type their feelings(or sentences)
- The ML algorithm analyses the sentiment/emotions of the written text and return one of the emotions - anger, joy, fear, sadness, surprise, love.
- The quotes for each of the emotions is present in the quotes folder. The program randomly chooses a quote and displays to the user.
- First/Home Screen of the App
- Sign-in/Login Page
- Analysing Text and Displaying Quote
- Algorithm Used: Naive Bayes Algorithm
- It is a classification technique based on Bayes' Theorem with an assumption of independence among predictors. In simple terms, a Naive Bayes classifier assumes that the presence of a particular feature in a class is unrelated to the presence of any other feature.
- Dataset Used: Emotions Dataset from Kaggle with 16000 sentences
- Accuracy Obtained: 80% Accuracy was obtained.
The following are the challenges I came across whilst making this project
- Lack of resources to study the framework of Kivy which made the app development difficult.
- The ML model obtained only 80% accuracy because of lesser data.
- This app can be modified into a full Mental Health app by including more features. For example, incorporating posting blogs, chatbot and other mental health resources.
- We can improve the accuracy of the model by adding cleaner and larger data.
- The app can be deployed for both android and ios uers.
I learnt a lot as a developer while making this project. I learned an entirely new framework of Kivy, strengthened my skills in Python, dived a little deeper into data science, mainly NLTK and Classification Algorithms.
The repository has been made and maintained by Kashika Akhouri