The Centre for Tamazight Language Processing Research (CTLPR) represents a proper educational and research resource in the fields of langugae processing and speech recognition and their applications for serving Tamazight in various areas such as manuscript reservation, language processing, machine translation, information retrieval etc... Also, playing a part in the research community of the discipline at international levels and communicating the results of our research to a wider public. The CTLPR focuses on delivering knowledge and expertise in postgraduate research level . All academic activities in our institution are carried out by academics and professionals who are committed to work in this area covering various aspects such as manuscript digitization, tamazight language processing using artificial intelligent (AI) techniques, text analysis using statistical methods etc... The major research areas of interest include:
- Tamazight Language Processing & Speech Recognition
- Machine Translation
- Information Retrieval
- Computer Aided Language Teaching/Learning Tools
- Tamazight Corpora Generation
- Tamazight text classification and analaysis
- Manuscript Digitization
- Tifinagh Development & Processing
Tifinagh processing using XML and
unicode standard ver. 4.1
[Source: http://www.w3.org/ ]
Tifinagh Keyboard [Source: www.language-keyboard.com]
Research Projects (3-5 years)
Project No. 1: Tamazight Corpora Generation (TCG)
This project aims to generate Tamazight corpora for a huge amount of Tifinagh-based text and voice data. The main sources of the generated data will broadcast news, newspapers, WWW ...etc. This data will be used to train any future artificial intelligent (AI) based solution models for automatic language processing of Tamazight text and/or voice.
Project No. 2: Tamazight Morphological Analyzer (TMA)
Project No. 3: Tamazight Part of Speech Tagger (T-POT)
Project No 4: Tamazight Documents Automatic Classification (TDAC)
This project aims to model a solution for classifying Tamazight Tifinagh-based text douments using artificial intelligence (AI) techniques. This project will make use of the corpora generated from the Project No. 1 as well as POT of Project No. 3. A framework has already been proposed as follows:
