This project focuses on understanding, cleaning, and analyzing the Netflix Titles dataset using Python. The objective is to explore how real-world datasets are inspected, cleaned, and analyzed before applying machine learning techniques.
The project demonstrates the early stages of a typical data science workflow, including dataset inspection, data cleaning, exploratory data analysis (EDA), and visualization of important patterns within the data.
The goal is to build a strong foundation in working with structured datasets and understanding how proper data preparation supports future analytical and machine learning tasks.
Dataset Used: Netflix Titles Dataset
The dataset contains information about movies and TV shows available on Netflix.
Each row represents a single Netflix title (movie or TV show).
Dataset structure:
- Rows: 8807
- Columns: 12
| show_id | type | title | director | country | release_year | rating | duration |
|---|---|---|---|---|---|---|---|
| s1 | Movie | Dick Johnson Is Dead | Kirsten Johnson | United States | 2020 | PG-13 | 90 min |
| s2 | TV Show | Blood & Water | Unknown | South Africa | 2021 | TV-MA | 2 Seasons |
| Column | Description |
|---|---|
| show_id | Unique identifier for each title |
| type | Indicates whether the title is a Movie or TV Show |
| title | Name of the movie or show |
| director | Director of the title |
| cast | Main actors appearing in the title |
| country | Country of production |
| date_added | Date when the title was added to Netflix |
| release_year | Year the title was released |
| rating | Audience rating classification |
| duration | Length of movie or number of seasons |
| listed_in | Genre or category of the title |
| description | Short description of the content |
The dataset contains several rating categories including TV-MA, TV-14, TV-PG, PG-13, and R, indicating the intended audience for each title.
The project follows a structured data analysis workflow:
Load Dataset
↓
Dataset Understanding
↓
Basic Data Checks
↓
Data Cleaning (handling missing values and duplicates)
↓
Save Cleaned Dataset
↓
Statistical Summary
↓
Exploratory Data Analysis
↓
Visualization and Insights
This pipeline converts raw data into a clean dataset suitable for analysis.
Before performing analysis, the dataset was inspected and cleaned to ensure data quality.
Cleaning steps included:
- Inspecting dataset structure using
df.info() - Reviewing sample records using
df.head() - Identifying missing values
- Filling missing categorical values with
"Unknown" - Checking for duplicate records
- Verifying dataset consistency after cleaning
Example cleaning operation:
df[col] = df[col].fillna("Unknown")After cleaning, the dataset was verified to confirm that no missing values or duplicate rows remained.
Exploratory Data Analysis was performed to understand the distribution and characteristics of the dataset.
Key columns explored during analysis:
- type
- release_year
- country
- rating
- duration
Example command used during analysis:
df['type'].value_counts()EDA helps identify patterns, trends, and relationships within the dataset.
Movies dominate the Netflix catalog, significantly outnumbering TV Shows in the dataset.
The number of titles increases significantly after approximately 2015, reflecting the expansion of streaming platforms and original content production.
The United States produces the largest share of Netflix titles, followed by India and the United Kingdom, indicating strong contributions from both Western and international markets.
Most titles fall under TV-MA and TV-14, suggesting that the majority of Netflix content targets adult and teen audiences.
Many TV shows in the dataset have only one season, while movies typically range between 90 and 110 minutes in length.
Although this project does not implement machine learning models, the dataset could support several potential ML tasks such as:
- Predicting whether a title is a Movie or TV Show
- Predicting the content rating
- Genre classification using the description field
Before applying machine learning algorithms, additional preprocessing and feature engineering would be required to convert categorical and text data into numerical features.
The following tools and technologies were used in this project:
- Python
- pandas
- matplotlib
- Git
- GitHub
Through this project, the following skills were developed:
- Understanding and inspecting real-world datasets
- Data cleaning and preprocessing
- Exploratory data analysis
- Data visualization
- Organizing a structured data project
- Version control using Git
This dataset provides opportunities for further exploration, including:
- Advanced data analysis and trend exploration
- Feature engineering for machine learning
- Text analysis using the description column
- Building predictive models using the cleaned dataset
Future projects will expand on these possibilities by applying machine learning and more advanced analytical techniques.