Welcome to the Sinhala Data Cleaning and Preprocessing Library! This comprehensive Python library is designed to streamline the process of cleaning and preprocessing Sinhala text data for various natural language processing (NLP) tasks. Our library addresses the unique challenges posed by the Sinhala language, providing efficient solutions for data cleaning, text summarization, and more.
## Objective
Our primary objective is to create an all-encompassing library that enhances the quality and accessibility of Sinhala text data. Through machine learning techniques and innovative approaches, we aim to empower users across industries to efficiently process, understand, and utilize Sinhala language content.
## Key Research Questions
### Sameera W.G.G.A.S - IT20146924
- How can Singlish text be accurately and efficiently converted to Sinhala language using machine learning techniques?
- What approaches can be taken to support different variations of Singlish text and real-time conversion?
- How can integration with other systems be seamlessly achieved to enhance user experience?
### Galappaththi S.T - 20167264
- What are the key challenges in accurately converting Sinhala speech to text while maintaining grammar rules?
- How can verbal Sinhala text be effectively converted to written text while adhering to the language's specific grammar?
- What machine learning and deep learning models can be utilized for optimal performance in Sinhala language processing?
### Yasodya P.B.B - 20227586
- What are the limitations of existing tools for Sinhala text cleaning and preprocessing, and how can they be addressed?
- How can hybrid approaches combining rule-based and machine learning methods improve the accuracy and scalability of text processing?
- What strategies can be implemented to support various Sinhala text variations and promote integration with other systems?
### Wijesinghe W.R.A.S.S - 20181406
- What are the current challenges in summarizing Sinhala news articles, and how can they be overcome?
- How can the library effectively combine extractive and abstractive summarization techniques to generate high-quality summaries?
- What methods can be employed to provide users with customizable summarization lengths, formats, and translation options?
## Individual Objectives
### Sameera W.G.G.A.S - IT20146924
Develop a Python library for converting Singlish text to Sinhala language. Our library leverages deep learning models for accurate and real-time conversion, supporting various Singlish variations and integration options.
### Galappaththi S.T - 20167264
Create a comprehensive Sinhala Data Cleaning and Preprocessing Library. Focus areas include accurate Sinhala speech-to-text and verbal text-to-written text conversion, machine learning integration, and real-time performance optimization.
### Yasodya P.B.B - 20227586
Develop a robust Python library for cleaning and preprocessing Sinhala text data. Employ hybrid approaches combining rule-based and machine learning methods for accuracy and scalability. Prioritize support for various Sinhala text variations and integration with other systems.
### Wijesinghe W.R.A.S.S - 20181406
Design an advanced Sinhala text summarization library with extractive and abstractive techniques. Offer customizable summarization lengths and formats, and integrate translation features for wider accessibility of Sinhala news content.
## Contact
For inquiries or feedback, please contact us at sssbprojects@gmail.com.
---
Note: This version includes the individual research questions and objectives for each team member, providing a more detailed overview of their contributions to the project. Adapt the content to match your README's format and design.
