Bancie/Reading-Dataset

📘 Reading Performance Dataset

📌 Overview

The Reading Performance Dataset is a small-scale tabular dataset designed to analyze reading behavior and performance characteristics across diverse individuals. It captures demographic attributes, reading habits, attention levels, and content difficulty to support educational data analysis and machine learning experiments.

This dataset is suitable for educational analytics, behavioral analysis, and predictive modeling related to reading performance.

📊 Dataset features

Each row represents a single reading session. The dataset contains the following columns:

Feature Name	Type	Description
AGE_CATEGORY	Categorical	Age group of the reader (e.g., young_adult).
GENDER	Categorical	Gender of the participant (male, female, other).
MAJOR	Categorical	Academic major or role of the reader (e.g., student).
MINUTES_READING	Numeric	Total duration (in minutes) spent actively reading during the session.
MINUTES_BREAK	Numeric	Total break time (in minutes) taken during reading.
FOCUS_LEVEL	Ordinal	Level of focus during the reading session. Higher values indicate stronger concentration.
PAGES	Numeric	Number of pages read during the session.
CONTENT_LEVEL_ENUM	Ordinal	Difficulty level of the reading material.
READING_GENRE	Categorical	Genre of the reading material (e.g., academic, language, self-development).
SOUND_VOLUME	Ordinal	Environmental sound intensity during reading, ranging from silent to extremely loud.
DEVICE	Categorical	Device or medium used for reading (e.g., laptop, smartphone, book, tablet).
LOCATION	Categorical	Physical environment where the reading session took place (e.g., home, library, café).
WEATHER	Categorical	Weather condition during the reading session (e.g., sunny, rainy, cloudy).
MOOD	Ordinal	Emotional state of the reader during the session, from very bad to very good.
HUNGER	Ordinal	Hunger or satiety level during reading.
AROUSAL	Ordinal	Cognitive arousal level, from very low to very high.
MENTAL_IN_BREAK	Ordinal	Mental state during breaks (e.g., chaotic, neutral, calm).
MENTAL_FATIGUE	Ordinal	Level of mental fatigue experienced during the session.
INTERESTING_LEVEL	Ordinal	Degree of interest in the reading material.
COMPREHENSION_DEPTH	Ordinal	Depth of understanding achieved during reading.
LINKING_TO_PREVIOUS_KNOWLEDGE	Ordinal	Extent to which new content was connected to prior knowledge.

🎯 Tasks

This dataset can be used for:

• Text / behavior classification
• Educational performance analysis
• Student reading habit modeling
• Correlation analysis between focus, time, and output
• Regression or classification experiments

Although categorized under text classification, the dataset is also applicable to tabular machine learning tasks.

🛠 Data Source & Generation

The dataset was synthetically curated / collected for educational and analytical purposes. No personally identifiable information (PII) is included.

⚖️ License

This dataset is released under the MIT License, allowing free use, modification, and distribution with attribution.

🚀 Usage Example (Python)

Basic

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("Bancie/Reading-Dataset")

# View dataset structure
print(dataset)

# Access the default split (usually 'train')
data = dataset['train']
print(f"Number of examples: {len(data)}")
print(f"Features: {data.features}")

# Access a specific example
print(data[0])

# Convert to pandas DataFrame for analysis
df = data.to_pandas()
print(df.head())

Streaming

For large datasets or when you want to process data without downloading it entirely:

from datasets import load_dataset

# Load dataset in streaming mode
dataset = load_dataset("Bancie/Reading-Dataset", streaming=True)

# Iterate through examples
for example in dataset['train']:
    print(example)
    # Process your data here
    break  # Remove break to process all examples

📚 Citation

@misc{c.b_nguyen_2026,
    author       = { C.B Nguyen },
    title        = { Reading-Dataset (Revision 19cfbb0) },
    year         = 2026,
    url          = { https://huggingface.co/datasets/Bancie/Reading-Dataset },
    doi          = { 10.57967/hf/7416 },
    publisher    = { Hugging Face }
}