Bootstrapping for Numerical Open IE
Swarnadeep Saha, Harinder Pal, Mausam Mausam · 2017
We design and release BONIE, the first open numerical relation extractor, for extracting Open IE tuples where one of the arguments is a number or a quantity-unit phrase.BONIE uses bootstrapping to learn the specific dependency patterns that express numerical relations in a sentence.BONIE's novelty lies in task-specific customizations, such as inferring implicit relations, which are clear due to context such as units (for e.g., 'square kilometers' suggests area, even if the word 'area' is missing in the sentence).BONIE obtains 1.5x yield and 15 point precision gain on numerical facts over a state-of-the-art Open IE system. 1 * Most work was done when the author was a graduate student at IIT Delhi. 1 https://github.com/knowitall/openie1. Sentence: Hong Kong's labour force is 3.5 million.Open IE 4.2: (Hong Kong's labour force; is; 3.5 million) BONIE: (Hong Kong; has labour force of; 3.5 million) 2. Sentence: Microsoft has 100,000 employees.Open IE 4.2: (Microsoft; has; 100,000 employees) BONIE: (Microsoft; has number of employees; 100,000) 3. Sentence: James Valley is nearly 600 miles long.Open IE 4.2: (James Valley; is; nearly 600 miles long) BONIE: (James Valley; has length of; nearly 600 miles) 4. Sentence: Donald Trump is 70 years old.Open IE 4.2: (Donald Trump; is; 70 years old) BONIE: (Donald Trump; has age of; 70 years) 5. Sentence: James Valley has 5 sq kms of fruit orchards.Open IE 4.