---
license: mit
language:
- en
datasets:
- marketeam/Nemotron-Personas-Marketing
size_categories:
  - 1K<n<1M
task_categories:
  - text-generation
tags:
- personas
- marketing
- nvidia
- business
- filtered
- synthesize
- synthetic
configs:
  - config_name: synthesize
    data_files:
      - split: synthesize_nvidia
        path: marketing_personas.csv
    description: A complete marketing personas dataset
pretty_name: Marketing Personas (Filtered from Nemotron-Personas)
---

# Marketing Personas (Filtered from Nemotron-Personas)

![Dataset Preview](./marketing_personas.jpeg)

---

## Dataset Card for Marketing Personas (Filtered from Nemotron-Personas)

This dataset contains only **Marketing-related personas** filtered from the original  
[nvidia/Nemotron-Personas](https://huggingface.co/datasets/nvidia/Nemotron-Personas) dataset.  

The filtering was done using keyword-based search for terms like  
*marketing, advertising, branding, campaign, social media, promotion, sales, content strategy, digital marketing*, etc.

---

## Dataset Details

### Dataset Description
- **Curated by:** Marketeam.ai  
- **Source:** Filtered subset of `nvidia/Nemotron-Personas`  
- **Language(s):** English  
- **License:** Apache 2.0 (inherits from source dataset license)  

This dataset is designed for anyone who wants to work specifically with **Marketing Personas** for applications in AI, business research, and generative persona-driven systems.

### Dataset Sources
- **Repository:** [nvidia/Nemotron-Personas](https://huggingface.co/datasets/nvidia/Nemotron-Personas)

---

## Uses

### Direct Use
- Training AI assistants to role-play as marketing professionals  
- Generating personas for business simulations  
- Benchmarking models on marketing-related tasks  

### Out-of-Scope Use
- Should not be used as factual marketing data  
- Not intended for sensitive profiling or real-world targeting  

---

## Dataset Structure

Each record contains all original fields from `Nemotron-Personas`, filtered by keywords.  

Fields may include:
- **persona** – the name/title of the persona  
- **description** – extended background of the persona  
- **other metadata** – depending on the source dataset  

---

## Dataset Creation

### Curation Rationale
The goal was to extract a domain-specific subset (Marketing Personas) to make experimentation easier for business/AI use cases.

### Source Data
Filtered from the `nvidia/Nemotron-Personas` dataset.  

#### Data Collection and Processing
- Filtering was done using **Pandas + Python keyword search**  
- Kept only rows where `persona` or `description` matched marketing-related terms  

#### Who are the source data producers?
Original creators: **NVIDIA (Nemotron-Personas dataset)**  

#### Personal and Sensitive Information
No sensitive information included; personas are synthetic.  

---

## Bias, Risks, and Limitations

- The dataset reflects only personas that **explicitly mention marketing keywords**  
- Some relevant personas may have been excluded if they used different terminology  
- Risk of bias toward English-speaking marketing roles  

### Recommendations
Use this dataset for research and prototyping, **not** as ground-truth marketing profiles.  

---
---

![image/png](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F686bbbffa66475f08533b50e%2Fq7Xvj59zzmpNqW-cq8Tdp.png)

![image/png](/static-proxy?url=https%3A%2F%2Fcdn-uploads.huggingface.co%2Fproduction%2Fuploads%2F686bbbffa66475f08533b50e%2FvtB7YTXN-lV5xAUQ9q5IZ.png)

## Citation

If you use this dataset, please cite:  

**APA:**  
NVIDIA. (2024). *Nemotron-Personas*. Hugging Face. https://huggingface.co/datasets/nvidia/Nemotron-Personas  

Filtered subset prepared by Romi18 (2025).