This project focuses on multi-label classification of fast food restaurants using both traditional machine learning and deep learning models. By analyzing detailed addresses and various features of each restaurant, the goal is to accurately categorize them into multiple relevant categories.
- Data Preprocessing: Cleaning data, handling missing values, and engineering relevant features.
- Exploratory Data Analysis (EDA): Visualizations including word clouds and geographical distributions.
- Machine Learning Models: Implementation of Logistic Regression, Naive Bayes, SVM, Random Forest, and K-Nearest Neighbors.
- Deep Learning Models: Development of Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Artificial Neural Networks (ANN).
- Evaluation Metrics: Assessment using AUC ROC, Precision, Recall, and F1 Score.
- Visualization: Comparative plots of model performances and confusion matrices for detailed analysis.
- Programming Language: Python 3.x
- Libraries:
- Data Manipulation:
pandas,numpy - Visualization:
matplotlib,seaborn,wordcloud,folium,plotly - Natural Language Processing:
nltk,re - Machine Learning:
scikit-learn - Deep Learning:
tensorflow(with Keras API) - Geocoding:
geopy - Miscellaneous:
nbformat
- Data Manipulation: