The AQUAINT Program is devoted to solving a single problem: How do we find topically relevant, semantically related, timely information in massive amounts of data in diverse languages, formats, and genres? Given the incredible amounts of information available today, merely reducing the size of the haystack is not enough; information professionals such as intelligence analysts require timely, focused answers to complex questions.
AQUAINT technology is advancing the development of components and functions that allows users to pose a series of intertwined, complex questions and obtain comprehensive answers in the context of broad information-gathering tasks. Additionally, while most information retrieval systems present only links to documents, AQUAINT is producing technology which will actually present answers to the user’s questions. This Question-Answering technology is being developed with features for managing: semantic similarity, coreference, event characterization, opinions, linguistic and social and world inferencing, redundancy, deception, and missing or contradictory information. In order to allow the analyst to guide the exploration in concert with the machine, AQUAINT technology employs: interactive question-answering, the automatic suggestion of additional paths of exploration, and the inferencing of the social context of the information search.
The AQUAINT program has long led the way in objective third-party evaluations, most notable sponsoring the Text Retrieval Conference (TREC) at the National Institute of Standards and Technology (NIST). AQUAINT tools are frequently tested by actual analysts. AQUAINT has advanced question-answering research to the point that commercial search tools identify targeted, semantic searching as the next great breakthrough in information discovery.