Facebook-UCI-Prediction

This project includes some basic data analysis on the UCI Facebook dataset.

It contains a basic exploratory data analysis with answers to some questions, and a model for predicting Lifetime post consumers for a post, based on the provided dataset variables.

Files

• README.md: Short description and instructions
• requirements.txt: Python library dependencies
• preparation.py: Python module with data download and preparation functions
• model.py: Python module with functions for modeling the Lifetime post consumers variable.
• EDA.ipynb: Part 1 of the analysis: answering some basic questions on the data set.
• Model.ipynb: Part 2 of the analysis: Modelling Lifetime Post Consumers

Instructions

Requirespython 3.7 or higher. You need to create a virtual environment:

python3 -m venv /path/to/virtual/environment

Activate the virtual environment, and install the libraries required using:

source /path/to/virtual/environment/bin/activate
pip install -r requirements.txt

You need to add the virtual environment as a jupyter kernel, in order to be able to select it in Jupyter:

ipython kernel install --name "venv" --user

Then launch Jupyter notebook:

jupyter notebook

You can then open and run the notebooks EDA.ipynb and model.ipynb.

Source

The data for this analysis are available here:

https://archive.ics.uci.edu/ml/datasets/Facebook+metrics

Moro, S., Rita, P., & Vala, B. (2016). Predicting social media performance metrics and evaluation of the impact on brand building: A data mining approach. Journal of Business Research, 69(9), 3341-3351.