📚 Word Learning Application

Effective tool for learning words and definitions based on intelligent repetition algorithms.

🚀 How to Run the Application

This project uses uv for dependency management. Python 3.12+ is required.

uv sync
uv run flet run .

Project Structure

The thin launcher src/main.py calls ft.run on the learning_app package (under src/, as expected by Flet packaging):

src/
  main.py                 # thin launcher: ft.run(learning_app.app.main)
  assets/                 # app icons / splash / bundled demo CSVs for flet build
  learning_app/
    app.py                # Flet application entrypoint
    data/                 # CSV processing, data model, file paths, demo sets, constants
    ui/
      components/         # reusable Flet controls
      screens/            # larger application screens and menus
    utils/                # small general helpers
tests/                    # tests (not packaged into the app)

📱 Android Release

You can also download the Android version of the application:

📖 Documentation

Full developer documentation available on GitHub Pages:

📋 About the Application

This application was designed with effective vocabulary and definition learning in mind. It allows you to:

• ✅ Import your own study sets
• ✅ Add built-in demo sets from Settings
• ✅ Track your progress
• ✅ Intelligently schedule repetitions
• ✅ Export data and statistics

📱 Application Interface

🧠 Learning Algorithm

The application uses an advanced learning algorithm visualized in the graph below:

Word Learning Queue

Words are queued for learning according to the following states:

State	Code	Description
Unknown	0/0/1	Words verified as unknown
Unverified	0/0/0	Words that have not yet been verified
Unconfirmed as known	1/0/0, 1/0/1, 1/1/1	Words marked as correctly answered but not yet confirmed as known
Known	1/1/0	Words answered correctly twice in a row

Queuing Process

1. First, words in the Unknown state (0/0/1) are selected
2. If there aren't enough unknown words, Unverified words (0/0/0) are added
3. If more words are still needed, Unconfirmed as known words are added

Explanation of State Codes (0/0/0)

State codes (e.g., 0/0/0) represent the following word attributes:

• First digit: whether the word was answered correctly in the last attempt (0 - no, 1 - yes)
• Second digit: whether the word was answered correctly in a row (0 - no, 1 - yes)
• Third digit: whether the word is marked as requiring learning (0 - no, 1 - yes)

📊 File Formats

The application supports two types of CSV files:

1. _words.csv Files

Contain words grouped by parts of speech:

• verb: verb
• person: person
• thing: noun
• adjective: adjective
• adverb: adverb

2. _definitions.csv Files

Contain definitions and their corresponding words:

• definition: definition
• word: corresponding word

Statistics Columns (Optional)

• correct_answers: number of correct answers (integer)
• good_answers_in_a_row: whether there were correct answers in a row (boolean)
• good_answer: whether the last answer was correct (boolean)
• word_to_learn: whether the word requires learning (boolean)

Sample CSV File

,definition,word
0,"1 number",one
1,"2 number",two

💡 Tips

• Make sure your CSV files have the correct format and columns
• Data import/export is available from the app's main menu
• From Settings, use Add demo sets to install three sample learning sets (word formation, English definitions, English–Polish). Delete them like any other set; Add again only installs missing ones
• Words marked as "learned" will appear less frequently
• The app automatically prioritizes words you find challenging
• For definitions with multiple correct answers or word forms, you can use the slash symbol (/) to separate them. For example: "go/goes/went". During a learning session, entering any one of these words will be marked as correct.

🔧 Technologies

The application was created using:

• Flet - a modern framework combining Flutter and Python
• Python - backend programming language
• Pandas - data manipulation library

---

Created with ❤️ and Flet (Flutter + Python)