At QCRI we are dedicated to promoting the Arabic language in the information age by conducting world-class research in Arabic language technologies.
Ensuring that the Arabic language flourishes in the digital world is a primary focal area of our research. Some of our current research projects address the challenges related to lack of content and equally important, extracting that content.
QCRI strives to become the regional and global leader in Arabic language technologies – in the areas of search, information retrieval and analysis, multilingual language processing, advanced machine translation and also leading efforts to increase and enrich Arabic language content online.
We are working hard to help close the gap caused by the lack of valuable Arabic content on the web by engaging in efforts that increase and enrich this content. Our partnership with Wikimedia Foundation marked the first step in this initiative. Through our collaboration with the Arabic Wikipedian community, the number of editors and their productivity has increased. We are working with universities and educational institutions to integrate Wikipedia into the curricula, which will also help increase and enrich online content through increasing knowledge and consumption. The initial goal is the addition of 10,000 Arabic articles of the highest quality. “Online content” is not just restricted to documents. We are also working with YouTube/Google on video content, as well as social media platforms such as Twitter. A critical component of the initiative is the creation of an outreach program to communicate with the Arabic Internet users to raise awareness and the delivery of the desired information.
QCRI’s initiatives do not only address the lack of content, it also addresses challenges in retrieving this content when it exists, making it accessible and enabling information flow across language barriers. In this regard, development is underway to process the Arabic language in the search domain such as the use of morphological word analysis, named entity recognitionand data learning technology to detect relevant content that can be used for more elaborate analysis. In addition, the development of proofing tools such as typographical checks and language identificationand the handling of different forms of the Arabic language in the form of local dialects and Arabic written using Latin characters.
A major effort at QCRI goes into improving machine translation for both text and speech. Combining a “Speech-to-Text” engine that allows the instantaneous transcription of videos with machine translation system for dealing with the Arabic language allows access to broadcast news and news distributed over the web. Future research will concentrate on applications such as lecture translation.
With our work in search and information retrieval, we have developed services that go beyond basic search functionality thus enabling a more exploratory search and in turn, better analytics of search results. We have built search functionality that is more scalable and more language-aware. Much of our work has been done in the social media domain, yet is transferable to other domains. Our expertise in natural language processing and machine translation has helped build the foundation for this research.
Bridging a gap identified in the education domain, we have established projects related to e-education, enabling people to access and learn material in a language not native to their own. The development of an e-book reader with native Arabic support for the Arabic language, as well as an assistive language tutor are examples of such tools that will have an immediate impact on society and learning.
We have worked closely and collaborated with many local and international organizations including Al Jazeera, MIT and the Qatar Supreme Education Council on our projects.
Some of achievements and focus areas include: