Skip to content

ramiradwan/onlyfans-conversational-analytics

Repository files navigation

💬 onlyfans-conversational-analytics

Turn your chats into insights. Track conversations, trends, and performance — all in one dashboard.


📖 Overview

onlyfans-conversational-analytics helps creators make sense of their conversations by turning raw chat data into easy-to-understand insights.
Once your conversations are analyzed, you’ll see all your key metrics visualized in an interactive dashboard where you can track engagement, response times, and audience sentiment over time.


📊 Metrics

Each topic or theme includes detailed metrics you can filter by date or time range:

Metric Description
Volume Total number of conversations about a topic.
% of Total How much of your overall chat activity that topic represents.
Trend Shows whether conversation volume is growing or dropping compared to the previous period.
AHT (min) Average handling time — how long your typical chat lasts from start to finish.
% Silence Average percentage of silence time (no messages exchanged) across conversations.
Turns Average number of times the conversation switches between you and your fan.
Sentiment Average mood or tone of your fans, from 0 (negative) to 1 (positive).

💡 Why It’s Useful

  • 📊 See what topics drive the most engagement
  • 🕒 Measure your response efficiency
  • 💖 Understand your fans’ overall sentiment
  • 🧠 Identify trends and performance patterns
  • 🔧 Developer-friendly API for custom dashboards and integrations

🛠️ Getting Started

Developer setup instructions coming soon.

For now, you can connect your data pipeline and visualize your chat analytics via the provided dashboard once conversations are ingested and processed by Insights.


Technical Foundation

The system design aligns with Azure Cosmos DB’s Gremlin API (for graph storage and traversal) and vector embeddings (for semantic similarity and NLP-driven inference).
Data pipelines process text via transformer-based NLP models, extracting entities and relationships to populate a Labeled Property Graph (LPG).

This architecture supports both:

  • Creator analytics dashboards (via time-series and metric aggregation), and
  • Psychotherapy research graphs (via semantic and relational modeling).

🧠 About

Built for creators who care about connection

This project also serves as a foundation for graph-based psychotherapy research, modeling interactions and interventions as interconnected nodes and edges to study relational and dynamic patterns of change.

See Research Releases for more information!

About

provides a unified view of conversation and analysis data to help you research, monitor, and visualize insights. Once conversations are ingested and processed, the system displays detailed analytics through an interactive dashboard.

Topics

Resources

License

Stars

Watchers

Forks

Contributors