README.md

metadata

license: apache-2.0
task_categories:
  - text-classification
language:
  - en
tags:
  - indian-names
  - gender-prediction
size_categories:
  - 10K<n<100K
configs:
  - config_name: default
    data_files:
      - split: train
        path: Names_with_Gender_Labels_Dataset.csv

🇮🇳 Indian Names & Gender Dataset (Balanced)

Dataset Summary

This dataset contains 42,000 samples designed for training models to identify Indian names and classify their gender. It is perfectly balanced across three categories, making it ideal for training robust classifiers that can distinguish between real names and random text.

Dataset Structure

Column	Type	Description
Name	String	The text string (Name or Random Word).
Label	Integer	Class ID representing the category (0, 1, or 2).

Label Mapping

Label ID	Class	Count	Examples
0	Neutral / Non-Name	14,000	Ability, About, Agra, Abohar
1	Male Name	14,000	Aabharan, Aabhat, Aadalalagan
2	Female Name	14,000	Aabarana, Aabitha, Aadanya

Usage

This dataset is suitable for:

1. Gender Classification: Distinguishing between Male and Female Indian names.
2. Name Entity Recognition (NER): Filtering valid Indian names from common English words or place names (Label 1/2 vs Label 0).

Quick Start (Python)

from datasets import load_dataset

dataset = load_dataset("shisha-07/Indian_Names_with_Gender_Dataset")

# Print first example
print(dataset['train'][0])
# Output: {'Name': 'A', 'Label': 0}

🤖 Models Trained on this Dataset

• Llama-3-Indian-Gender-Classifier: A fine-tuned Llama-3 model that achieved 86.4% accuracy using this data.

• 🔗 Source Code

The code used to create and train on this dataset is available here: