https://github.com/pejpero/neuralnetwork-language

NeuralNetwork Language is a console application that uses a single-layer neural network to identify the language of a given text. It trains on data in the trainData folder and predicts the language of texts in the testData folder or manually inputted text. The application supports multiple languages as long as they are correctly labeled.
https://github.com/pejpero/neuralnetwork-language

artificial-intelligence console-application language-recognition machine-learning neural-network single-layer-perceptron text-classification training-dataset

Last synced: 3 months ago
JSON representation

NeuralNetwork Language is a console application that uses a single-layer neural network to identify the language of a given text. It trains on data in the trainData folder and predicts the language of texts in the testData folder or manually inputted text. The application supports multiple languages as long as they are correctly labeled.

• Host: GitHub
• URL: https://github.com/pejpero/neuralnetwork-language
• Owner: PejperO
• License: mit
• Created: 2023-04-14T10:25:57.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2023-04-26T14:07:30.000Z (about 2 years ago)
• Last Synced: 2025-01-12T06:27:36.123Z (4 months ago)
• Topics: artificial-intelligence, console-application, language-recognition, machine-learning, neural-network, single-layer-perceptron, text-classification, training-dataset
• Language: Java
• Homepage:
• Size: 36.1 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
## NeuralNetwork Language

A console application that, using simple artificial intelligence, aims to recognize in which language a given text is written. 

The single-layer neural network learns on a training set in the trainData folder - regardless of the number of folders (provided they are properly described). 

### Testing the files in the testData folder

The program, using the data set in the testData folder, tries to predict in which language each .txt file was written. 

    

    German test:

    0. correct

    1. correct

    2. correct

    3. correct

    4. correct

    English test:

    0. incorrect

    1. correct

    2. correct

    3. correct

    4. correct

    French test:

    0. correct

    1. correct

    2. correct

    3. correct

    4. correct

    Polish test:

    0. incorrect

    1. correct

    2. correct

    3. correct

    4. incorrect

### testing manually typed text into the console

The application also has the ability to check the text manually entered into the console.

    Enter a sentence: 

    In Boston engagierte sich Severance bald bei mehreren Organisationen mit unterschiedlichen Reformzielen.

    

    Your sentence is written in German

    

## What I Learned

* perceptron operation

* what is a single-layer neural network

* how to teach artificial intelligence

* reading all the files given in the path, regardless of their number