Landing Page
Retrieving a Real TikTok link
BrightShare's Analysis Page

BrightShare — Fair, Transparent, and Inclusive Rewards for TikTok Creators

Overview
Problem Statement
Features
Reward System
Tech Stack
Live Deployment
Setup & Local Installation
Usage
Demo
Screenshots
License

Overview

BrightShare is a reward distribution system that fairly allocates revenue to TikTok creators based on:

Engagement metrics
Sentiment in comment sections
Detection of AI-generated content
Inclusivity factors

By combining these into a normalized reward score, BrightShare:

Ensures fair and transparent payouts
Discourages low-quality or fraudulent content
Promotes inclusivity and ecosystem health

Creators also gain access to a dashboard showing reward calculations, engagement breakdowns, and sentiment analysis for full transparency.

Problem Statement

While TikTok monetization exists, current mechanisms suffer from:

Inconsistent/opaque rewards → many creators underpaid
System gaming & fraud → inflated metrics, AIGC spam
Toxic/low-quality content incentives → hurts community trust

BrightShare addresses these by rewarding quality, authenticity, and inclusivity.

Features

Base Value Analysis → Compute core revenue from TikTok (ads, gifts, EVI)
Sentiment Analysis → Assess positivity & toxicity in comments
Comment Section Regulation → Incentivize creators to maintain healthy, safe comment spaces
AIGC Detection → Penalize suspected AI-generated content
Inclusivity Analysis → Uplift small/underrepresented creators
Reward Allocation → Transparent score-based payouts
Creator Dashboard → Track engagement, scores, and payouts

Reward System

The final reward for a TikTok creator is calculated as:

$$ \text{Reward} = R_{base} \times M_{quality} \times M_{integrity} \times I_{index} $$

Each component is described below.

1. Base Revenue ($R_{base}$)

$$ R_{base} = \text{Ad Revenue} + \text{Gift Revenue} + \text{EVI} $$

where Engagement Value Index (EVI) is:

$$ EVI = 0.1 \cdot \frac{Likes}{Views} + 0.4 \cdot \frac{Shares}{Views} + 0.3 \cdot \frac{Comments}{Views} + 0.2 \cdot \frac{Reposts}{Views} $$

Purpose: Capture the core monetization of a video, including ads, gifts, and engagement quality.
Adjustable weights: Can be tuned to reflect platform priorities.

2. Community Quality Multiplier ($M_{quality}$)

$$ M_{quality} = 0.5 \cdot (\text{Positivity Rate}) + 0.5 \cdot (1 - \text{Toxicity Rate}) $$

Positivity via Hugging Face RoBERTa Twitter model
Toxicity via Google Perspective API

🛡️ Comment Section Regulation

BrightShare enforces creator accountability for their audience:

Weighted Influence on Rewards
- Positive, low-toxicity comment sections boost $M_{quality}$, increasing payouts.
- Toxic comment sections reduce $M_{quality}$, even with high engagement.
Creator Responsibility
- Encourages creators to moderate and foster respectful discussions.
Dynamic Penalties & Bonuses
- Extremely toxic comment sections → significant reward reduction.
- Exceptionally positive comment sections → bonus multiplier.

This ties community health directly to income, incentivizing better moderation.

3. Integrity Multiplier ($M_{integrity}$)

$$ M_{integrity} = 1 - 0.25 \cdot \min(1, P(\text{AIGC})) $$

Detection Model: Reality Defender API
Purpose: Penalize suspected AI-generated content (up to 25%).

🤖 AIGC Regulation

Authenticity Check
- Videos are scanned for AI generation probability ((P(\text{AIGC}))).
Reward Penalty Mechanism
- Proportional penalty based on probability:
  - $$(P(\text{AIGC}) = 0.5 \Rightarrow M_{integrity} = 0.875)$$
  - $$(P(\text{AIGC}) = 1 \Rightarrow M_{integrity} = 0.75) (max penalty)$$
Balancing Creativity & Fairness
- Minor AI assistance tolerated; full AI-generated content is penalized.
Transparency
- Dashboard shows AIGC risk and resulting multiplier.

Ensures human creators are fairly rewarded while discouraging content farms.

4. Inclusivity Index ($I_{index}$)

$$ I_{index} = 1 \times (1 + 0.1 \cdot \text{Small Creator}) \times (1 + 0.1 \cdot \text{Underrepresented Community}) $$

Small Creator: Follower count < _SMALL_CREATOR_THRESHOLD → +10%
Underrepresented Community: Creator from _TARGET_COUNTRIES → +10%

🌍 Promoting Inclusivity

Boosting Small Creators
- Small creators receive a 10% reward boost, leveling the playing field.
Supporting Underrepresented Communities
- Creators from target countries also receive a 10% bonus.
Multiplicative Bonuses
- If a creator qualifies for both:
  $$I_{index} = 1 \times 1.1 \times 1.1 = 1.21 \quad (\text{+21\% boost})$$
Dashboard Transparency
- Creators can see exactly how size and location affect rewards.

Encourages a diverse, globally representative creator ecosystem.

Tech Stack

Frontend

React 19 + TypeScript (Vite)
ESLint
CSS

Backend

FastAPI (Python 3.10+)
Pydantic
TikTokApi + Apify (scraping metadata & comments)
Hugging Face API (sentiment)
Google Perspective API (toxicity)
Reality Defender (deepfake/AIGC detection)
Docker for containerisation

Tooling

Node.js 18+ / npm
dotenv (env management)
Playwright (web scraping)
Railway for automated live deployment

Project Directory Skeleton

tiktok-valuereimagined/
│── backend/
│   ├── app/
│   │   ├── analysis.py
│   │   ├── main.py
│   │   ├── nlp_sentiment.py
│   │   ├── requirements.txt
│   │   ├── schemas.py
│   ├── download_model.py
│   └── railway.json
│
│── frontend-react/
│   ├── public/
│   ├── src/
│   │   ├── App.css
│   │   ├── App.tsx
│   ├── package.json
│   └── vite.config.ts
│
│── .env            # not provided please request from us if necessary
│── .gitignore
│── Dockerfile
│── railway.json
│── README.md

Live Deployment

The BrightShare frontend is now hosted on Railway and accessible to anyone without local setup.

🌐 Access the live app here: BrightShare on Railway

Fully functional dashboard and reward computation
Supports real-time engagement, sentiment, AIGC analysis, and inclusivity scoring
Ideal for quick demos or testing without local installation

⚠️ Backend API calls still rely on your configured environment variables if using a local or custom backend.

Setup & Installation (Local Installation)

The API keys are not provided for security reasons so please contact us if you want to test this system out locally. If not, it is reccomended to access our project via the live deployment site at BrightShare on Railway.

Prerequisites

Docker Desktop
Python 3.10.6
Node.js >= 18 (with npm)

1. Clone repository

git clone https://github.com/limgabriel/tiktok-valuereimagined
cd tiktok-valuereimagined

2. Backend setup

Docker Setup (reccomended)

  docker build -t tiktok-backend .

Using Python Virtual Environment

  # Create and activate virtual environment
  # macOS/Linux
  python3.10.6 -m venv .venv
  source .venv/bin/activate

  # Windows
  py -3.10.6 -m venv .venv
  .venv\Scripts\activate      

  # Install dependencies
  pip install -r backend/app/requirements.txt

  # Install Playwright for TikTok scraping
  playwright install

Optionally, if you are running it on VSCode, you could follow the following shortcuts: 1. Ctr + Shift + P 2. Type Python: Create Environment 3. Select Venv 4. Choose Python 3.10.6 5. Select to install via requirements.txt

3. Frontend setup

cd frontend-react
npm install

4. Run services

Backend

Docker

    docker run -p 8000:8000 --env-file .env tiktok-backend

Python Virtual Environment

  python backend/start.py

Frontend

cd frontend-react
npm run dev

Open local browser → http://localhost:5173/

Usage

You can use BrightShare in two ways:

Live Deployment (Recommended for demos):
- Open BrightShare on Railway
- Paste a TikTok video link
- Dashboard fetches engagement + comments
- Sentiment & AIGC detection auto-run
- Reward score is computed & displayed
Local Installation (for development or testing):
- Follow the steps in Setup & Installation