- About the Project
- Part 1: Exploratory Data Analysis (EDA) and Data Wrangling
- Part 2: Machine Learning Models
The project is divided into two parts: exploratory data analysis and machine learning models. The goal of part 1 of this project is to gain a comprehensive understanding of the hotel reviews. This involves examining the structure, content, and distribution of the reviews to identify key patterns and insights. We will preprocess the reviews and prepare the dataset for machine learning. This includes text cleaning, tokenization, lemmatization, and feature extraction to ensure the data is in a suitable format for modeling. The goal of part two is to perform sentiment analysis by developing and evaluating several machine learning models to accurately label the sentiment expressed in hotel reviews as positive or negative.
I will perform EDA and data wrangling techniques to gain comprehensive insights into the dataset.
The following open source packages are used in this part of the project:
- Natural Language Toolkit (nltk)
- Pandas
- Numpy
- Plotly
- Matplotlib
- SciKit Learn (sklearn)
The dataset can be downloaded here.
I will develop several machine learning models to correctly label the sentiment behind hotel reviews. The machine learning models I will use are Logistic Regression, K-Nearest Neighbors, and Decision Trees.
The following open source packages are used in this part of the project:
- NumPy
- Pandas
- Matplotlib
- SciKit Learn (sklearn)
A cleaned and preprocessed dataset will be used to do the analyses stated above. The dataset can be downloaded here.