Original size 900x1289

Television Trends in Numbers: An Analytical Look at the Drama Industry

PROTECT STATUS: not protected
Longread translated automatically

Data Selection and Its Source

For this study, I selected a dataset concerning popular Korean dramas, which includes 3,550 rows of information covering the series titles, release years, ratings, genres, age classifications, synopses, and cast members. The data was sourced from open repositories on Kaggle.

Reasons for Selection

The study of Korean dramas was chosen as a topic because they represent not only a significant cultural phenomenon and an art form, but also a powerful global industry with its own laws of development and trends. Data analysis allows us to identify key patterns in viewer preferences and general trends within this media market:

— the dependence of ratings and popularity on genre, cast, and production studio — the distribution of drama popularity across different countries and viewer demographics — an analysis of the themes and plot archetypes most in demand during different periods — identifying the relationship between script elements, visual style, and audience success

This analysis can form the basis for strategic decisions in media production, international content distribution, and the marketing positioning of new projects.

Choosing visualization types

To effectively present the data, I selected several different types of visualizations:

  1. «Number of Dramas by Year (since 2000)» — a bar chart, used to show the dynamics of drama releases over the years.
  2. «Drama Rating Distribution» — a histogram with KDE, which illustrates the distribution of user ratings.
  3. «Vote Count Distribution» — a histogram that shows how the popularity of dramas is distributed by the number of votes.
  4. «Top 10 Genres by Drama Count» — a horizontal bar chart that reflects the most common genres.
  5. «Genres with the Highest Average Rating» — a bar chart that compares genres based on average rating.
  6. «Rating and Vote Count» — a scatter plot, demonstrating the relationship between popularity and score.
  7. «Drama Distribution by Age Rating» — a bar chart that shows which audience demographic most dramas target.
  8. «Rating Distribution by Age Rating» — a box plot, used to compare rating distributions across different age categories.

Each visualization type was chosen based on the nature of the data and the specific analytical objectives.

Stages of Work

The analysis process involved systematically working with the K-drama dataset and comprised several stages.

  1. Data Loading and Preparation: The dataset was loaded using the pandas library; initial data cleaning was performed, column types were standardized, and relevant features were selected; the table structure, presence of missing values, and anomalous values were checked.

  2. Preliminary Analysis: Key statistical characteristics of the numerical variables were calculated; the distributions of ratings, votes, release years, and categorical features (genres, age ratings) were analyzed; key metrics of interest for further analysis and visualization were identified.

  3. Aggregation and Transformation: The data was grouped by year, genre, and age rating; average ratings, drama counts, and other aggregated metrics were calculated; summary tables and filtered data subsets were generated for chart construction.

  4. Visualization and Interpretation: Various chart types were constructed (histograms, bar charts, scatter plots, and box plots); axis labels, titles, and value range constraints were added; conclusions were drawn from the visualizations regarding the release trends of dramas, the popularity of genres, and the relationship between ratings and the number of votes.

The primary tools for the analysis were the pandas library for data processing and aggregation, along with matplotlib and seaborn for creating and styling the visualizations.

Using Neural Networks

During the process, I utilized Google’s Gemini language model. The neural network helped me to:

  1. Structure the code for efficient data processing.
  2. Select optimal visualization types for various analytical aspects.
  3. Create dictionaries for translating categories and colors.
  4. Interpret the analysis results.

The primary prompts used when working with Gemini focused on selecting visualization types, styling them, and effectively presenting the outcomes.

Chart Styling

To create a professional and harmonious data presentation, I developed a unified style for all visualizations:

  1. Applied consistent formatting to titles, axes, and labels.
  2. Added numerical values to key chart elements.
  3. Used a grid to improve data perception.
  4. Maintained a consistent structure and design across all visualizations.

These principles ensured the integrity and professional appearance of all charts, making the presented information easier to perceive and analyze.

Final visuals and their interpretation

Original size 3552x1751

The chart titled «Number of Dramas by Year (since 2000)» illustrates the dynamic release schedule of dramas over time. The X-axis displays the years, and the Y-axis displays the number of projects. A steady increase in the number of dramas is visible in recent years. Conclusion: the drama industry is growing and developing rapidly.

Original size 2952x1450

The «Drama Rating Distribution» chart illustrates how user scores are spread out. Most values are concentrated in the above-average range. Extremely low and high ratings are rare. Conclusion: Overall, the dramas receive consistently positive reviews.

Original size 2952x1750

The «Top 10 Drama Genres by Episode Count» chart displays the most frequent genres. Drama and romance hold the leading spots. The remaining genres appear significantly less often. Conclusion: the industry is focused on emotionally rich narratives.

Original size 2952x1750

The «Genres with Highest Average Rating» chart compares genres based on audience ratings to gauge quality. Some genres with fewer projects exhibit higher average ratings, suggesting high satisfaction among the target audience. Conclusion: a genre’s popularity does not always equate to its quality.

Original size 2952x1450

The «Distribution of Votes» chart illustrates the popularity of K-dramas among viewers. The majority of projects receive a small number of votes. Only a few dramas garner a very large audience. Conclusion: high popularity is characteristic of a limited number of hits.

Original size 2953x1752

The «Rating and Number of Votes» chart illustrates the relationship between popularity and evaluation. There is no evident linear correlation. A high number of votes does not guarantee a high rating. Conclusion: widespread appeal does not always signify high quality.

Original size 2952x1750

The «Distribution of Dramas by Age Ratings» chart shows which audiences are most frequently targeted by these projects. Most dramas are aimed at a broad audience. Content with strict limitations appears less often. Conclusion: creators aim to reach the widest possible viewership.

Original size 2953x1750

The «Rating Distribution by Age Rating» chart compares scores across age groups. The rating distributions in all categories are similar. No significant differences are observed between them. Conclusion: the age rating has a weak influence on the final drama score.

Key Takeaways

The analysis conducted revealed several key patterns within the world of the K-drama media industry:

— There is a steady annual increase in the number of dramas released, indicating active industry growth and rising audience interest. — Although most dramas receive above-average ratings, high popularity, as measured by the number of votes, does not always correlate with high viewer scores. — The drama and romance genres dominate the number of projects, yet some less common genres demonstrate higher average ratings. — The age rating of dramas has a weak influence on user scores, as the rating distributions across categories are largely similar. — An analysis of the relationship between rating and vote count showed no strong correlation, underscoring the importance of content quality over sheer volume.

These findings can be utilized by analysts and producers to forecast viewer interest, select priority genres, and optimize content release strategies.

Description of Generative Model Application

Television Trends in Numbers: An Analytical Look at the Drama Industry
Project created at 16.01.2026
creativecity.academy
A joint project of the HSE ART AND DESIGN SCHOOL and the Institute for the Development of Creative Industries, focused on working with Russian regions. The Creative City digital platform is a media outlet that aggregates projects related to territorial branding, a community of professionals, and a service that allows you to create local websites, work with the local identity and popularity of cities and places. Study the map of the Creative City. Get acquainted with the projects that have already been implemented. Add your... Show more
events
travelling
history
art and design
We use cookies to improve the operation of the HSE website and to enhance its usability. More detailed information on the use of cookies can be fou...
Show more