===========================================================================

EDA Amazon Books Reviews

An exploratory analysis of the Amazon Book Reviews. An expansion to this project is carried out using Collaborative Filtering models from the research: A Comparative Analysis of Amazon Book Ratings Using Collaborative Filtering.

◘ Introduction

The general idea behind this research is to identify and establish a pattern or a set of patterns and relationships among a plethora of features available within the acquired data. A thorough analysis is performed by using a multitude of tools and packages using Python so that a set of statistical and/or machine learning models can be applied to accomplish better generalization.

◘ Study Flowchart

◘ Project Organization

---

├── LICENSE
├── Makefile          				<- Makefile with commands.
├── README.md             	<- The top-level README for developers using this project.
├── data
|
|
├── features                		<- Set of files to construct a more readable and useable data.
|   |── data_filtering.py		<- Script that filters through the data and keeps important features.
|   |── data_processing.py   <- File where data is cleaned and visually analyzed for its distribution.
|   |── data_inspection.py    <- Script to observe and search for patterns and relationship among various features.
|   └── nlp.py                     <- Text from the review features are processed and analyzed for better intuition.
│	 
|
├── figures            				<- Generated graphics and figures to be used in reporting (includes IDE and Notebooks generated graphs).
│    				     
│
├── notebooks          			<- Additional script for Jupyter Notebooks for better visualization.
│
│
├── requirements.txt    		<- The requirements file for reproducing the analysis environment, e.g.
│                         				    generated with `pip freeze > requirements.txt`
│
├── setup.py           			<- makes project pip installable (pip install -e .) so src can be imported
|
│
|	
├── visualization           		<- Create exploratory and results oriented visualizations.
|   |── analysis_charts.py		<- Script to better facilitate abstractions for generating simple graphs.
|   └── mining_charts.py		<- Can be used to generate a more specific type of graph to be utilized during data inspection.
│
└── tox.ini            				<- tox file with settings for running tox; see tox.readthedocs.io

---

◘ Modules Required:

• pandas 2.0.0
• plotly 5.15.0
• missingno 0.5.2
• vaderSentiment 3.3.2
• spacy 3.5.3
• matplotlib 3.7.1
• seaborn 0.12.2
• wordcloud 1.9.2

===========================================================================