The main purpose of question answering is to answer automatically to those questions which are made in a natural language.
The search for answers, called in English Question Answering (QA) is a type of information retrieval. Given a certain amount of documents (such as World Wide Web), the system should be able to recover answers to questions raised in natural language. A system of question answering is one of the most complex systems around the retrieval of information. We must take into account that a system based on the question-answering is much more difficult than a normal system which is responsible for seeking some information in a much more or less large documents, since these should draw from these documents a fragment of text to respond to a question given in natural language. These systems are closely linked to the seekers web.
There can be two types of questions answered:
- The first of them are called close-domain questions and can be easily answered because they belong to a topic which can be found on the Internet without any problem.
- The second ones are called open-domain questions and are harder to answer because they rely on world knowledge.
Even if many questions can be answered due to this system, a hard work must be done in the future in order to be able of answering all of them.
References:
- Question answering (13th June 2009 at 19:00). In Wikipedia, The Free Encyclopedia. Retrieved 17.20, 19th June, 2009 from: http://en.wikipedia.org/wiki/Question_answering
- Sistemas de busqueda Question-Answering (20th April, 2006). Retrieved 17.23, 19th June, 2009 from: http://usuarios.lycos.es/questionanswering/
- PIQASso: Pisa Question Answering System (Giuseppe Attardi, Antonio Cisternino, Francesco Formica, Maria Simi, Alessandro Tommasi). Retrieved 17.28, 19th June, 2009 from: http://trec.nist.gov/pubs/trec10/papers/piqasso.pdf
