INDEX

Table of Contents
1. Abstract
2. Introduction
3. Problem Statement
4. Objective
5. Tools & Technologies
6. Dataset Description
7. Data Collection
8. Data Preparation
9. Exploratory Data Analysis (EDA)
10.Visualization Techniques
11.Genre-Wise Analysis
12.Rating Distribution Analysis
13.Votes vs Rating Correlation
14.Release Year Trends
15.Key Insights
16.Challenges Faced
17.Limitations
18.Future Scope
19.Conclusion
20.References
21.Appendix A (Full Code)
22.Appendix B (Graph Outputs)
Chapter-1
Abstract
The rise of digital platforms and global access to entertainment has led to an unprecedented
explosion in the production and consumption of movies. In this context, understanding
viewer preferences through data analytics becomes increasingly vital for filmmakers, critics,
streaming services, and content creators. This project, titled “Movie Ratings Analysis,”
focuses on analysing a curated dataset of movie attributes—such as genre, rating, number of
votes, and release year—using Python and visualization libraries to extract valuable insights
and trends.
The primary goal of the project is to discover how factors like movie genre, public votes, and
year of release influence a film's average rating. By employing libraries such as Pandas,
Matplotlib, and Seaborn, we perform structured data cleaning, exploration, and
visualization. The dataset consists of 10 sample movies from various genres like Action,
Romance, Drama, Thriller, Sci-Fi, and Crime. Though the dataset is small and synthetic, it
mirrors the patterns commonly seen in real-world data and serves as a prototype for larger
analyses.
The analysis reveals several interesting patterns: Drama and Sci-Fi movies tend to receive
higher ratings on average, while Action movies, though more frequent, show moderate
rating values. A positive correlation is observed between the number of votes and the
overall rating, indicating that movies with higher viewer engagement often perform better in
terms of rating. Additionally, release year trends show a steady increase in the number of
films, reflecting industry growth.
Through bar charts, scatter plots, histograms, and line graphs, we present these insights in a
visual, easy-to-understand format. These visualizations help stakeholders make informed
decisions about audience preferences, genre popularity, and potential content strategies.
Although the project is based on a small dataset, it successfully demonstrates the
capabilities of data analytics in transforming raw data into strategic intelligence. The study
concludes with suggestions for future improvement, including the integration of real-time
data from sources like IMDb APIs, prediction models using machine learning, and interactive
dashboards for non-technical users.
This project serves as a foundational template for academic, professional, and business use-
cases where media data analytics is crucial. It highlights the importance of combining
statistical understanding with visual communication to unlock the full potential of data-
driven storytelling.
Chapter-2
Introduction
The global entertainment industry, particularly the film sector, has seen exponential growth
over the past few decades. With the emergence of online streaming platforms such as
Netflix, Amazon Prime, Disney+, and others, the accessibility and consumption of movies
have expanded across diverse audiences and regions. As the number of films released every
year increases, so does the volume of data associate with these films—ratings, reviews,
genres, cast, runtime, budgets, box office collections, and more.
Among these data points, movie ratings—usually out of 10—are one of the most critical
indicators of a film's success and public perception. Ratings provide a quick snapshot of how
a movie was received by its audience. When combined with additional features like genres,
number of votes, and release year, these ratings can offer powerful insights into trends in
viewer behaviour, genre popularity, and audience engagement.
This project, titled "Movie Ratings Analysis," is an exploratory data analytics project
designed to uncover meaningful patterns in a dataset of fictional but realistic movie entries.
The dataset includes key variables:
• Movie Title
• Genre
• Release Year
• Rating (on a scale of 0 to 10)
• Number of Votes
Using Python and data visualization libraries such as Pandas, Matplotlib, and Seaborn, the
project walks through a complete data analysis pipeline: from data loading and preparation,
to cleaning, transformation, exploration, and visualization.
The purpose of this study is to answer several core questions:
• What genres tend to receive higher ratings?
• Do movies with more votes tend to have better ratings?
• Which years saw more movie releases?
• What is the overall distribution of ratings in this sample?
Though the dataset is relatively small (10 entries), the techniques and concepts applied are
scalable to large real-world datasets such as those from IMDb or TMDb. This project not only
provides hands-on experience in data visualization and interpretation but also offers a
template for how to perform movie analytics using a structured, replicable approach.
Furthermore, this report emphasizes the importance of visual communication in data
science. By representing complex numeric relationships through bar charts, histograms, and
scatter plots, it becomes easier for both technical and non-technical stakeholders to
understand and act upon the insights derived from the data.
In the modern era of content recommendation engines, targeted advertising, and machine-
learning-driven entertainment platforms, such analytics play a crucial role in personalizing
the user experience and predicting trends. This project lays the foundation for such
advanced analyses by starting with exploratory techniques and ending with actionable
insights based on viewer ratings and preferences.
Chapter-3
Problem Statement
The movie industry produces thousands of films across multiple genres every year. While
some movies achieve massive critical and commercial success, others fail to meet audience
expectations. For producers, directors, streaming platforms, and marketing teams,
understanding why certain movies perform better than others is crucial for making informed
business decisions.
With the rise of online platforms such as IMDb, Rotten Tomatoes, and Metacritic, millions of
viewers now contribute their ratings and reviews for movies they've watched. These ratings,
combined with information about genres, release years, and the number of votes, represent
valuable data that can help stakeholders answer questions like:
• Which genres are consistently well-received by audiences?
• Do movies with a higher number of votes generally have higher ratings?
• Is there a relationship between the year of release and movie success?
• What is the typical rating distribution across the movie industry?
• How can this information guide future movie production and marketing strategies?
However, despite the abundance of available data, many organizations struggle to extract
actionable insights from it due to:
• The large volume of unstructured data.
• The complexity involved in analysing multiple variables.
• The lack of expertise in data visualization and interpretation.
This project aims to address these challenges by providing a structured approach to movie
ratings analysis using data science techniques.
Chapter-4
Objective
The primary objective of this project is to perform a comprehensive analysis of movie
ratings data using Python-based data analytics and visualization techniques. The study aims
to extract meaningful patterns, trends, and insights from the dataset, which can help
stakeholders in the entertainment industry better understand audience behaviour, genre
popularity, and movie success factors.
Key Objectives of the Study:
1. Data Preparation and Cleaning
• Load the provided dataset into a suitable analysis environment.
• Inspect the data for missing values, inconsistencies, or errors.
• Clean and preprocess the data to ensure it is analysis-ready.
• Convert relevant fields (e.g., Release Year) into correct data types.
2. Exploratory Data Analysis (EDA)
• Use descriptive statistics to summarize the data.
• Examine the central tendencies, dispersion, and shape of the data distribution.
• Understand the distribution of ratings across different movies and genres.
3. Genre-wise Analysis
• Group movies by their genre to compute average ratings.
• Identify which genres consistently achieve higher ratings from audiences.
• Compare performance across genres using bar plots and statistical summaries.
4. Rating Distribution Analysis
• Visualize how movie ratings are distributed across the dataset.
• Identify common rating ranges and outliers.
• Analyse whether ratings follow a normal, skewed, or bimodal distribution.
5. Votes vs. Ratings Correlation
• Investigate whether the number of votes influences the movie’s rating.
 • Use scatter plots to visualize any positive or negative correlation.
• Interpret whether highly rated movies tend to attract more votes.

6. Release Year Trend Analysis

• Analyse the number of movies released per year.
• Detect any growth trends or patterns in production volume over time.
• Provide insights into whether certain years witnessed more highly-rated movies.
7. Visualization and Communication
• Use Python visualization libraries such as Matplotlib and Seaborn to create:
o Bar Charts
o Histograms
o Scatter Plots
o Line Graphs
• Represent insights visually to make the findings accessible to both technical and non-
technical stakeholders.
8. Summarization of Key Insights
• Summarize the major takeaways derived from the analysis.
• Provide actionable recommendations based on data-driven observations.
9. Identify Limitations and Future Scope
• Recognize the scope and boundaries of the current analysis.
• Suggest areas for further improvement, such as using larger real-world datasets, live
data scraping, or machine learning for predictive analytics.

➢ Ultimate Goal:
To demonstrate the power of data analytics in converting raw movie ratings data into
meaningful insights that can inform better decisions in:
• Movie production planning
• Content recommendation algorithms
• Audience targeting strategies
 • Marketing and promotional efforts.

Chapter-5
Tools & Technologies Used
In this project, several tools and technologies were employed to handle data processing,
analysis, and visualization. Each tool plays a specific role in the overall data analytics
pipeline, from data loading to generating insights.

1. Python (Programming Language)

• Purpose:
Python serves as the primary programming language for this entire project.
• Why Python?
Python is widely regarded as one of the most popular languages for data science and
analytics due to its simplicity, readability, and an extensive ecosystem of open-source
libraries.
• Key Features for This Project:
o High readability and ease of use.
o Rich support for data analysis libraries.
o Strong community support.
o Wide usage in both academia and industry.

2. Pandas (Data Manipulation and Analysis Library)

• Purpose:
Used to load, manipulate, clean, and analyze structured data.
• Key Functions in This Project:
o Reading data from CSV files (pd.read_csv()).
o Data cleaning (handling missing values, type conversions).
o Grouping and aggregating data (groupby()).
o Basic statistical analysis (describe(), info()).
o Filtering and slicing data for focused analysis.
 • Why Pandas?
Pandas provides high-performance data structures (like DataFrames) that make it
very easy to perform complex data operations with minimal code.

3. Matplotlib (Data Visualization Library)

• Purpose:
To create static, interactive, and publication-quality plots for data visualization.
• Key Functions in This Project:
o Bar charts for genre and year-wise analysis.
o Line plots for temporal data.
o Customization of plots (labels, titles, axes).
• Why Matplotlib?
Matplotlib offers full control over every aspect of a figure, making it highly
customizable for visualizing complex datasets.
4. Seaborn (Advanced Visualization Library built on Matplotlib)
• Purpose:
To create attractive, informative statistical graphics that are aesthetically pleasing and
easier to interpret.
• Key Functions in This Project:
o Histograms to analyze rating distribution.
o Scatter plots for correlation analysis (Votes vs Ratings).
o Bar plots for genre-wise average rating.
o Automatic handling of complex statistical plots.
• Why Seaborn?
Seaborn simplifies many tasks that would require extensive coding in Matplotlib. Its
themes and color palettes greatly enhance the appearance of the plots.
5. VS Code (Visual Studio Code – Code Editor)
• Purpose:
Used as the Integrated Development Environment (IDE) to write, debug, and execute
Python code.
• Key Features Used:
o Syntax highlighting for Python.
 o Integrated terminal to run scripts.
o Extensions for Python debugging and linting.
o Easy integration with external libraries.

• Why VS Code?
It's lightweight, highly customizable, free, and widely used by both beginners and
professionals in the data science community.

6. CSV File Format (Comma-Separated Values)

• Purpose:
CSV file was used as the source of input data for this project.
• Why CSV?
o Simple text-based format.
o Easy to read and write by both humans and programs.
o Supported natively by Python’s Pandas library.
Chapter-6
Dataset Description
In any data analytics project, understanding the dataset thoroughly is a crucial step before
performing any meaningful analysis. This section provides a comprehensive description of
the dataset used in the Movie Ratings Analysis Project.

Dataset Overview
The dataset is a fictional, sample dataset created to simulate real-world movie ratings data.
While small in size, it contains all the essential attributes commonly found in actual movie
databases such as IMDb, Rotten Tomatoes, or TMDb.

Dataset Structure
The dataset consists of 10 movie entries with the following structure:
Attributes in Detail
1. Movie Title
• Type: Text/String
• Description: The unique title of each movie.
• Example: The Great Adventure
2. Genre
• Type: Categorical/Text
• Description: The category or genre the movie falls under.
• Categories Present:
o Action
o Romance
o Thriller
o Sci-Fi
o Drama
o Crime
 • Importance: Helps identify viewer preferences across different genres.
3. Release Year
• Type: Integer
• Range: 2015 – 2021
• Description: The year in which the movie was released.
• Importance: Used to analyze trends in movie production over time.
4. Rating
• Type: Float
• Range: 6.9 – 9.0
• Description: Average rating based on viewer feedback, on a scale of 0 to 10.
• Importance: Main performance metric for movie quality and reception.
5. Votes
• Type: Integer
• Range: 1200 – 6100
• Description: Total number of audience votes submitted for each movie.
• Importance: Used to assess viewer engagement and correlation with ratings.
Chapter-7
Data Collection
Introduction to Data Collection

Data collection is one of the most critical steps in any data analytics
project. The accuracy, completeness, and reliability of the dataset directly
affect the validity of the analysis and the conclusions that can be drawn
from it. In real-world scenarios, data about movies can be collected from
multiple sources such as online databases, APIs, web scraping, or through
licensed datasets.

For the purposes of this project, the data collection process was simplified
by creating a synthetic sample dataset that closely resembles actual data
structures used by popular movie databases like IMDb, TMDb, and
Rotten Tomatoes.

Sources of Data (In Real-World Context)

In practical, real-world projects, movie data can be collected from:
Chapter-8
Data Preparation
Introduction to Data Preparation
Data preparation, also known as data preprocessing, is a crucial step that transforms raw
data into a clean, organized, and analysis-ready form. Without proper preparation, even the
most sophisticated analytical models can produce misleading or incorrect results.
In this project, data preparation involved importing the data, cleaning any inconsistencies,
ensuring correct data types, and organizing the dataset for seamless analysis.

Steps Involved in Data Preparation

1. Importing Required Libraries
The first step is to load the necessary Python libraries used throughout the project for data
handling and visualization:

pandas for data manipulation.

matplotlib and seaborn for data visualization.

2. Loading the Dataset

The dataset stored as a CSV file (movie_data.csv) was loaded into a Pandas DataFrame:

This loads the data into a tabular format that allows easy manipulation and
exploration.
3. Initial Data Inspection
To get an overview of the dataset:
head() shows the first few records.
info() shows data types and non-null counts.
describe() provides summary statistics for numerical fields.

Why is Data Preparation Important?

• Avoids analysis errors.
• Ensures reliable and valid results.
• Makes downstream processing efficient.
• Helps identify any irregularities early on.

Prepared Dataset Summary

After preparation, our dataset consisted of:
• 10 movies.
• 5 well-defined columns.
• No missing values or duplicates.
• Correct data types.
Chapter-9
Exploratory Data Analysis (EDA)
Purpose of EDA

• Exploratory Data Analysis (EDA) is the phase in which we explore the dataset to
understand its structure, detect patterns, spot anomalies, and form hypotheses. In this
project, EDA is performed using both numerical summaries and graphical
representations to gain valuable insights about the movies dataset.

Genre Distribution

Objective:

• To identify which movie genres are most common in the dataset.

Insight:
• Action and Sci-Fi were the most frequent genres.
• Drama and Crime were less represented in this sample.

Average Rating by Genre

Objective:

• To determine if some genres generally receive better ratings than others.

Insight:
• Drama, Crime, and Sci-Fi genres tend to receive higher ratings.
• Action movies, although frequent, had slightly lower average ratings.

Votes vs Rating Correlation

Objective:

• To determine if popular movies (more votes) tend to have higher or lower ratings.

Insight:
• There is no strong linear correlation between votes and rating.
• Some highly voted movies had moderate ratings and vice versa.

Average Rating Over Years

Objective:

• To observe how average movie ratings have changed across years.

Insight:
• Ratings fluctuated across years, peaking in 2016 and 2020.
• 2017 and 2021 had relatively lower average ratings.
Chapter-10
Data Visualization & Insights
Purpose of this Section
While data visualization is part of Exploratory Data Analysis (EDA), this section focuses on
interpreting the meaning behind the graphs. Good visualizations are not just about making
charts — they help you communicate complex ideas clearly and effectively. Here, we
analyze the key visual outputs generated and extract actionable insights.
1. Genre-wise Movie Count
Visualization Recap:
A bar plot showing the count of movies per genre.

Interpretation:
• Action genre appears most frequently, suggesting its popularity in production.
• Genres like Drama and Crime are underrepresented but can be high performers in
terms of audience satisfaction.
• Strategic Insight: A high number of movies in a genre doesn’t guarantee better
audience ratings (as we’ll see later). Quality over quantity is critical.
2. Average Rating by Genre
Visualization Recap:
A bar plot comparing average IMDb ratings across genres.

Interpretation:
• Drama had the highest average rating in the dataset, followed by Crime and Sci-Fi.
• Action movies had the lowest average rating, despite being the most frequent.
• Strategic Insight: For studios aiming at critical acclaim or higher audience approval,
investing in Drama and Sci-Fi content may offer better returns in reputation and
awards.

3. Votes vs Rating (Scatter Plot)

Visualization Recap:
A scatter plot mapping the number of votes to the corresponding movie rating.
 Interpretation:
• No strong linear trend is visible — some highly rated movies received fewer votes.
• One movie had fewer votes but a very high rating, suggesting it may be a hidden gem
or critically appreciated but under-watched.
• Strategic Insight:
o Popular ≠ Best — marketing, reach, or distribution often drive vote count.
o For recommendation systems, both votes and rating should be considered
together.

4. Year-wise Average Ratings

Visualization Recap:
A line plot showing the average rating of movies by release year.
Interpretation:
• Ratings peaked in 2016 and 2020, with average scores nearing 9.0 and 8.4,
respectively.
• Dips observed in 2017 and 2021, which may indicate years where production focus
shifted or genre preferences changed.
• Strategic Insight: Movie quality and reception fluctuate over time. Industry trends
(e.g., streaming boom during pandemic years) might explain rating spikes.

Visualization Summary Table

Chapter-11
Genre-Wise Analysis

Drama genre has highest average rating.

Action genre is the most frequent.
Chapter-12
Rating Distribution Analysis

Most ratings are between 7 and 8.5

Normal distribution pattern observed.
Chapter-13
Votes vs Rating Correlation

Positive correlation between votes and ratings.

Some genres (like Drama) have higher rating regardless of vote count.
Chapter-14
Release Year Trends

Maximum movies released in 2019 & 2020.

Industry output growing steadily.
Chapter-15
Key Insights

1.Action Movies Are Common but Not the Best

• There were more Action movies than any other genre.
• But Action movies had lower ratings compared to others.
• So, just making more Action movies doesn’t mean they’ll be liked more.
2.Drama and Romance Are Loved by People
• Drama and Romance movies had the highest ratings and also got a lot of
votes.
• These kinds of movies connect with people’s emotions, which makes
them more enjoyable.
3.More Votes Doesn’t Always Mean Better Movie
• Some movies had a lot of votes but didn’t get high ratings.
• Example: One movie had over 3,000 votes but only a 6.9 rating.
• Another had fewer votes but a 9.0 rating.
• So, popular movies are not always the best ones.
4.Best Year Was 2016
• Movies released in 2016 got the highest average rating.
• 2017 was the worst year in terms of ratings.
• This shows that movie quality can change over the years.
5.Sci-Fi Movies Are Strong Performers
• Sci-Fi movies had high ratings and good audience response.
• Even with fewer Sci-Fi movies, they did really well.
• A good choice for producers who want both quality and popularity.
6.Action and Thriller Movies Have Mixed Quality
• Some Action and Thriller movies were good, others not so much.
• Their ratings were not consistent.
• These genres need better stories and ideas to keep the quality high.
7.One Good Movie Can Make a Big Difference
• In some years, one excellent movie made that year stand out.
• For example, The Pianist’s Tale in 2016 got the best rating and most
votes.
• One strong movie can lift the whole year’s performance.
8.No Strong Relationship Between Votes, Ratings, or Year
• We checked the data and found that:
o Votes and Ratings are only weakly related.
o Newer movies don’t always get more votes or better ratings.
• This means we can’t predict success just by looking at votes or release
year.
9.Each Genre Attracts Different Audiences
• Drama and Crime genres had more consistent ratings.
• Action and Thriller were less predictable.
• This helps in choosing which type of movie to recommend or promote to
different people.

10.Charts Help Us See Patterns Clearly

• Charts like bar graphs and scatter plots made it easy to:
o Compare genres
o See how ratings and votes change over time
 • Visuals help explain data better than plain numbers.

Summary Table
Chapter-16
Challenges Faced
Every data analysis project comes with its own set of challenges — technical,
analytical, and sometimes even creative. In this project, we faced several
obstacles while analyzing the movie dataset using Python and data visualization
tools.
Here are the major challenges explained in simple terms:
1.Small Dataset
What Happened:
• The dataset had only 10 movies, which is very small.
• With fewer data points, some insights (like correlations and trends) may
not be fully accurate or strong.
How We Handled It:
• We focused on qualitative insights (like genre performance) instead of
heavy statistics.
• Used visualization and interpretation to make meaningful observations
from limited data.
2. Balancing Genres
What Happened:
• Some genres had only one movie (e.g., Drama, Romance, Crime), while
others had more.
• This made it hard to compare genres fairly.
How We Handled It:
• Compared average ratings and votes instead of total counts.
• Made sure insights were clearly explained, keeping the data imbalance
in mind.
3. Weak Correlation Results
What Happened:
• Correlation analysis didn’t show any strong relationships between
variables.
• This could make it seem like the data has no story to tell.
How We Handled It:
• We looked beyond correlation — using scatter plots and genre-wise
analysis to find hidden patterns.
• Explained that popularity and quality are not always linked — which is
an insight in itself.
4. Visual Clarity
What Happened:
• With limited data, some charts looked flat or less meaningful.
• For example, a bar chart with only one value (like Drama or Crime) can
be hard to interpret.
How We Handled It:
• Added text-based insights below each graph.
• Used color schemes and labels to improve visual appeal and
understanding.
5. Choosing the Right Questions
What Happened:
• It was challenging to choose good questions to explore, especially with a
small dataset.
• We had to be careful to avoid over-analyzing limited data.
How We Handled It:
• Focused on basic but useful questions:
o What genre gets the highest ratings?
o Which year had the best movies?
o Is there a link between votes and ratings?
6. Code and Plot Adjustments
What Happened:
• At times, plot labels overlapped or charts didn’t show well in VS Code.
• Minor bugs in plotting code (e.g., incorrect axis labels or colors) slowed
progress.
How We Handled It:
• Used plt.tight_layout() and label rotation for clean visuals.
• Regularly debugged and improved the code for better presentation.
Chapter-17
Limitations
1.Small Number of Movies
• The dataset had only 10 movies.
• That’s too small to find very strong or accurate results.
This means the results might change if we use a bigger dataset.

2. No Audience Information
• We don’t know who voted (age, gender, or country).
So, we can’t say which group of people liked which movies the most.

3. Covers Only a Few Years

• Movies are only from 2015 to 2021.
Older movies or very new ones (like from 2022 to 2025) are not included.

4. Ratings Don’t Change Over Time

• In this project, each movie has one fixed rating and vote count.
But in real life, ratings can change as more people watch and review the
movie.

5. Only One Genre Per Movie

• Each movie was listed with just one genre (like Action or Drama).
But many real movies are a mix — like Action-Comedy or Sci-Fi-Thriller.
6. Weak Relationships Between Data
• We didn’t find strong links between:
o Ratings and Votes
o Year and Votes
So, we can’t say “more votes means a better movie” or “newer movies always
do better.”

7. No Machine Learning Used

• The project only used basic analysis and charts.
It didn’t use tools like prediction, AI, or smart recommendations.

8. Charts May Be Interpreted Differently

• Some charts were simple, but people might understand them in different
ways.
Especially with a small dataset, it’s easy to make mistakes when guessing
patterns.

9. Basic Tools Only

• We used only Python and VS Code for analysis.
No dashboards or interactive tools like Excel, Power BI, or Tableau were used.
Chapter-18
Future Scope
This project can be improved and expanded in many ways in the future:
1. Use a Bigger Dataset
• Include hundreds or thousands of movies for better results.
• More data = more reliable insights.
2. Add More Details
• Include data like budget, box office, cast, language, and awards.
• Helps understand what makes a movie successful.
3. Add Audience Info
• Include age, gender, and location of viewers.
• Useful for personalized movie suggestions.
4. Analyze Reviews
• Use text analysis (NLP) to study user reviews.
• Learn what people really think about the movie.
5. Use Machine Learning
• Predict movie ratings or popularity using AI.
• Makes the project more advanced and useful.
6. Create Recommendations
• Suggest movies to users based on their likes.
• Similar to how Netflix or YouTube works.
7. Make Interactive Dashboards
• Use tools like Power BI or Tableau.
• Easier for users to explore data visually.
Chapter-19
Conclusion
In this project, we studied a small dataset of 10 movies using Python and data
visualization tools like Matplotlib and Seaborn. Our goal was to find useful
insights about movie ratings, genres, votes, and release years.

What We Did:
• Cleaned and organized the movie data.
• Created different charts to understand trends.
• Compared genres like Action, Drama, Sci-Fi, etc.
• Looked at which years had the best-rated movies.
• Checked if more votes meant better ratings.
What We Found:
• Drama and Sci-Fi movies had the highest ratings.
• Action movies were the most common but had lower ratings.
• More votes didn’t always mean a better movie.
• 2016 was the best year for movie ratings.

Why It Matters:
• This analysis helps understand what kind of movies people like.
• It can guide movie makers or streaming platforms to improve content.
• Even a small dataset can give valuable insights when analyzed properly.
Chapter-20
References
Below are the websites, tools, and libraries that helped in completing this
project:

Tools and Libraries Used

• Python – Programming language used for data analysis
• Pandas – Used for handling and analyzing data
• Matplotlib – For creating charts and graphs
• Seaborn – For beautiful and informative visualizations
• Jupyter Notebook / VS Code – Used to write and run the code
Online Resources
• www.w3schools.com – For learning Python basics
• pandas.pydata.org – Official Pandas documentation
• matplotlib.org – Matplotlib library docs
• seaborn.pydata.org – Seaborn documentation
• kaggle.com – For sample datasets and ideas
• towardsdatascience.com – For articles on data analysis
Other Sources
• Classroom notes and tutorials
• YouTube tutorials on Python data visualization
• Sample CSV datasets from online platforms
Chapter-21
Appendix : A (Full Code)
Chapter-22
Appendix : B (Graph Outputs)
1.
2.

3.
4.