https://github.com/joe-stifler/tik-tok-video-count-analysis
https://github.com/joe-stifler/tik-tok-video-count-analysis
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/joe-stifler/tik-tok-video-count-analysis
- Owner: joe-stifler
- Created: 2024-12-19T22:46:49.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2024-12-23T15:33:05.000Z (10 months ago)
- Last Synced: 2025-02-04T13:58:05.584Z (8 months ago)
- Language: Jupyter Notebook
- Size: 3.77 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# TikTok Exploratory Data Analysis (EDA) Project
[Link to Tableau Dashboard](https://public.tableau.com/app/profile/jose.ribeiro6623/viz/Book1_17346387436440/Story1?publish=yes)
## Overview
This repository contains a Python notebook and supporting materials for an Exploratory Data Analysis (EDA) project focused on a TikTok dataset. The project aims to understand user behavior, content characteristics, and engagement patterns on the platform, with a specific emphasis on distinguishing between videos categorized as "claim" versus "opinion." This project is part of a larger effort to gather actionable insights that can drive business decisions. Dataset source: [here](https://www.coursera.org/learn/go-beyond-the-numbers-translate-data-into-insight).
## Key Objectives
* **Data Exploration:** Inspect and clean a TikTok dataset, addressing missing values and outliers.
* **Data Visualization:** Create a range of visualizations (box plots, histograms, bar charts, scatter plots, and pie charts) using Python's `matplotlib`, `seaborn`, and `pandas` libraries to identify key trends and patterns within the data.
* **Outlier Analysis:** Investigate the presence of outliers in video engagement metrics and analyze their impact on data.
* **Tableau Integration:** Create a Tableau dashboard to present key findings and visualizations in an accessible and shareable format.
* **Actionable Recommendations:** Generate evidence-based recommendations to enhance the platform and guide future actions.
## Content
* `tiktok_dataset.csv`: The raw dataset used for this analysis.
* `notebook.ipynb`: A Python Jupyter Notebook containing all code, analysis, and visualizations.
## Analysis Areas
The analysis explores key variables such as:
* `claim_status` (claim vs. opinion)
* `verified_status` (verified vs. not verified)
* `author_ban_status` (active, under review, banned)
* Video engagement metrics (`video_view_count`, `video_like_count`, etc.)