Yumi 🍅🌱

Yumi is an independent, open-source mobile application dedicated to consumer protection and nutritional analysis of food products.

Designed as a privacy-respecting alternative to services like Yuka, Yumi allows you to instantly understand the health impact of what you eat without compromising your personal data.

✨ Key Features

• 100% Privacy-First: No account required, no tracking, and no data sent to third-party servers for profiling.
• Fast Scanning: Instant recognition of barcodes (EAN-13, EAN-8, UPC, etc.) via camera or manual input.
• Adaptive Interface: Full support for system auto-theme, dark theme, AMOLED mode for energy saving, and a refined, high-contrast Dark Teal Light Theme.
• History & Favorites: Save your favorite products or browse your latest scans directly, stored locally on your device.
• Multilingual: Native support for Italian and English.

🧮 How the Algorithm Works (Health Score)

Yumi calculates a health score out of one hundred (0-100) based on a strict mathematical model that processes Open Food Facts data. The calculation follows these exact steps:

1. Base Score Definition (Nutri-Score)

The application extracts the official Nutri-Score grade (A, B, C, D, E) and its corresponding numerical base score (from -15 to +40).

• If the product is Water: It automatically receives a starting score of 100 points and does not suffer organic penalties.
• If Nutri-Score numerical data is available:
  • Solid Foods: The score is scaled proportionally to the official -15 to +40 range. The applied formula is:
    Base Score = 100 - (((NutriScore_Score + 15) / 55) * 100)
  • Beverages: The score is scaled proportionally to the specific liquids range from -20 to +40. The applied formula is:
    Base Score = 100 - (((NutriScore_Score + 20) / 60) * 100)
• If only the Letter is available (Estimated Grade):
  In the absence of a precise numerical score, a standard value is assigned based on the letter:
  A = 100 | B = 85 | C = 65 | D = 45 | E = 25.

2. Penalties for Harmful Additives

The algorithm analyzes the present food additive tags (E-number) and applies a penalty based on the scientific risk level (only the penalty of the most severe category found is applied):

• High Risk (Penalty: -15 points): Nitrites and nitrates (E249, E250, E251, E252), sulfites (E220-E228), azo and harmful dyes (E102, E104, E110, E122, E124, E129, E131, E133, E150c, E150d, E151), intense artificial sweeteners (E950, E951, E952, E954, E955, E961, E962).
• Moderate Risk (Penalty: -8 points): Phosphates (E338, E339, E340, E341, E343, E450, E451, E452), glutamates (E620-E625), flavor enhancers (E627, E631, E635), and critical emulsifiers/thickeners (E432-E436, E466, E471, E472e, E473, E475, E476, E491, E492).
• Limited Risk (Penalty: -4 points): Secondary thickeners and gelling agents (E407, E412, E414, E415, E416, E417, E425, E461), polyols/bulk sweeteners (E420, E421, E953, E965, E966, E967, E968).

3. Penalty for Lack of Organic Certification

Organic farming reduces exposure to chemical pesticides. If the product (excluding water) does not possess tags or labels certifying its organic nature (organic), it suffers a fixed penalty of -5 points.

4. Final Calculation and Safety Ceilings

The final score is calculated by subtracting the penalties from the base score:
Final Score = Base Score - Additives Penalty - Organic Penalty

To prevent low-quality products from receiving high scores simply due to the absence of additives, the algorithm applies unpassable maximum caps determined by the real Nutri-Score grade:

• If the Nutri-Score is E: the score cannot exceed 29/100
• If the Nutri-Score is D: the score cannot exceed 49/100
• If the Nutri-Score is C: the score cannot exceed 69/100

The result is then rounded to the nearest integer and strictly capped within the 0-100 range.

📸 Screenshots

🚀 Getting Started (Development)

The application is built using Flutter (Material 3) and is compatible with both Android and iOS.

Prerequisites

• Flutter SDK (version 3.22 or higher)
• Android Studio / Xcode

Installation

1. Clone the repository:

   git clone [https://github.com/your-username/yumi.git](https://github.com/your-username/yumi.git)
   
   
   

2. Navigate to the project directory:

cd yumi

3. Install dependencies:

flutter pub get

4. Run the application:

flutter run

🛠️ Built With

• Framework: Flutter
• Scanning: mobile_scanner (powered by Google ML Kit)
• Product Database: Public APIs from Open Food Facts
• Data Persistence: shared_preferences for local storage of history and settings.

📜 Required Acknowledgments and Licenses

Yumi is a transparent project built on the work of great open-source communities:

• Open Food Facts: Product data is sourced from the free and collaborative Open Food Facts database, distributed under the Open Database License (ODbL), and media assets under CC-BY-SA.
• Flutter SDK & Cupertino: Distributed under the BSD 3-Clause License.
• mobile_scanner: Distributed under the MIT License.
• shared_preferences: Distributed under the BSD 3-Clause License.
• Google ML Kit Barcode Scanning: Subject to the Google APIs Terms of Service.

---

Yumi is an independent application and is not affiliated in any way with Yuka or Open Food Facts.