Abrham-0464/Dataset-Collection

This repository contains a parallel dataset for machine translation from English to Xamtanga. The dataset includes sentence pairs in both languages, which can be used to train and evaluate deep learning models for language translation. The dataset is cleaned and preprocessed, with each sentence in Xamtanga aligned with its English translation. Xamtaga to English Translation Dataset This repository contains a parallel dataset for machine translation from Xamtaga to English. It includes sentence pairs from various domains, such as religion, education, media, and history, aimed at helping develop and evaluate machine translation systems for the Xamtaga language, which is considered a low-resource language. Dataset Overview The dataset was created by collecting parallel data from several sources: Religious texts: Including the Bible and other religious documents, providing consistent translations. Elementary school textbooks: Designed for Xamtanga learners, these texts cover a wide range of simple and educational sentences. News and media: Sourced from the Amhara Mass Media Corporation (AMICO) website, covering formal communication and topics such as politics, sports, and economics. Historical and cultural documents: Including local newspapers, magazines, and government reports that reflect the culture and history of the Xamtanga-speaking community. The dataset is stored in a filtered_parallel_Xamtanga-English_sentence_pair.txt file, which contains aligned sentence pairs in both Xamtanga and English. This makes it suitable for training and evaluating machine translation models. Dataset Files filtered_parallel_Xamtanga-English_sentence_pair.txt: Contains the aligned Xamtanga-English sentence pairs. Each line represents a pair of sentences one in Xamtanga and the other in English. Usage Clone the repository:git clone github.com Navigate into the repository:cd Dataset-collection The dataset file filtered_parallel_Xamtanga-English_sentence_pair.txt is located in the root directory. You can use it directly for training and evaluating machine translation models. This dataset is available under the MIT License. Feel free to use, modify, and distribute it for research and academic purposes with proper attribution. Contributions We welcome contributions to this da …