Data-Analysis-With-Automotive-Industry-Engage

Caralysis

Caralysis Caralysis is a web app which is aimed at solving these problems which provides dashboards, visualisations, trends, solutions for solving various challenges faced by automobiles as well as the improvisation they can bring to their business sales as well as serving customer’s satisfaction

Tech Stack

Data Analysis: Google Colab

Libraries: Pandas, Numpy, Scikitlearn

Data Visualisation Matpotlib, Seaborn, Dabl, Altair, charts.html

Framework: Streamlit

Installation and getting started

Run Caralysis in your local enviornment

Clone this repository

  git clone https://github.com/KhushbooGupta2111/Data-Analysis-With-Automotive-Industry-Engage

Download latest python version 3.10.2

https://www.python.org/downloads/
  

Head to the caralysis directory and install the libraries using pip

  cd caralysis
  pip install pandas
  pip install numpy
  pip install matpotlib
  pip install seaborn
  pip install streamlit
  pip install streamlit_multipage
  pip install streamlit_option_menu
  pip install Pillow
  pip install altair
  pip install opencv_python
  pip install SQLAlchemy
  pip install plotly

Running streamlit app on terminal

streamlit run app.py

Features

1. It provides dynamic features selections based on vehicle make and model preferences
2. It provides a dashboard wherein specifications including interior, exterior, safety performance, and latest features of variants of various models of Automobile Manufacturers are provided. Along with this visualization, plots have been shown for __ understanding the market trends.
3. A similar dashboard for Electric vehicles launched in India with their features and visualization
4. Data about CO2 emissions of various automobiles in India along with solutions to bring improvisation in the same have been provided. Data visualization of various parameters in automobiles and
5. Since Two Wheelers share a good market and demand for them is quite high in India. Therefore an existing or a newly established automobile industry can launch two-wheeler vehicles, catalysis also provides a dashboard along with visualization of various two-wheeler models.

Demo

Youtube link: https://www.youtube.com/watch?v=6OwziFN51M8

Screenshots

Learnings and Development Cycle

1. Analysing the challenges faced by the automobile industry and and scope of improvisation in their vehicles. It took me time to understand data science and how it helps to solve and provide business solutions based on data.
2. Examining the dataset given and searching for relevant datasets to solve other problems.
3. After this, various functions and ways were implemented and excuted for data cleaning which involved handling missing data,removing outliers, units, typecasting data for accurate data visualisation.
4. Learnt various libraries like pandas, numpy, scikitlearn for data analysis
5. Learnt various libraries like matplotlib, seaborn, plotly,altair and dabl for data visualisation.
6. Learnt about Feature Selection using heatmaps and cleaning the data based upon this.
7. Ploting visuals, correlations, heatmaps using the libraries mentioned above.
8. Learnt about Machine Learning models tried implenting the same for price prediction using the dataset given in the repository and which resulted in an accuracy of 0.76.
9. Choosing the viable framework/ library for building the frontend of the application using streamlit.
10. Testing the app and checking for bugs.

Future Scope

1. Giving an option to the automobile venture to upload their dataset and compare it with the features of well-established automobile manufacturers.
2. Deploying Machine Learning models to predict the price of various vehicles based on performance and features provided.
3. Learning about Neural Networks and implementing them to improve their accuracy.

🔗 Profile