Natural Language Processing
Applications of Natural Language Processing: Although the language and security applications for NLP are obvious, the analytic techniques used in Natural Language Processing also have unexpected applications to projects that at first do not seem to involve a language or linguistics.
- Financial markets depend on accurate understanding by financial analysts of opinions around the world that may not always be expressed in standard language.
- Medical applications include the facilitation of clear, accurate translation between English speaking doctors and patients who speak Arabic with a colloquial dialect, particularly when the patient may be a child who cannot speak formal standard Arbabic.
- NLP algorithms can be used to deal with alternative abbreviations and geographical coordinates. For example, the Center’s work with a large public utility includes thousands of entries listing street names or locations of manholes. This information is a vital part of understanding and predicting which electrical components under which manholes have failed or, depending on location, weather, age and other conditions are most likely to fail, but old data archives may record the address in different forms. Natural language processing algorithms are needed to ensure that 210 West Main St, 210 W, Main, Corner of West Main and Oak Street, and West Main – 210.
We invite you to center-jobscs [dot] columbia [dot] edu (contact us) to learn about our work in more detail or to discuss your challenges. All our research is open to the public and we are always interested in considering new challenges. Our team is already thinking about ways to apply their work to Spanish, Chinese, and other languages.
Preserving Morphological Richness in Poor-Pivot-based Statistical Machine Translation
The proposed effort addresses the weakness of using a morphologically poor language (English) as a pivot/bridge in statistical machine translation between morphologically rich languages (Hebrew and Arabic).
CCLS receives its first international grant award. The granting agency is the Qatar National Research Fund (QNRF), part of its National Priorities Research Program (NPRP). Out of 1,400 letters of intent submitted to the current NPRP cycle, 631 proposals were considered for review, and 145 projects were awarded (for a total of $121M).
Marine Carpuat, Yuval Marton and Nizar Habash received one of the two best paper awards at TALN 2010, the 17th "Conférence sur le Traitement Automatique des Langues Naturelles" (Conference on Natural Language Processing), which was held in Montréal from July 19 to 23, 2010.
Prof. Julia Hirschberg and Dr. Owen Rambow, along with Richard Sproat of the Oregon Health and Science University, have been awarded a grant by the National Science Foundation to develop new theoretical models and technology to automatically convert descriptive text into 3D scenes representing the text’s meaning.