Pintér, B; Vörös, G; Palotai, Z; Szabo, Z; Lőrincz, A; (2013) Determining Unintelligible Words from their Textual Contexts. Procedia - Social and Behavioral Sciences , 73 101 - 108. 10.1016/j.sbspro.2013.02.028.

Preview	PDF pinter13determining.pdf Available under License : See the attached licence file. Download (1MB)
Preview	PDF pinter13determining_presentation.pdf Available under License : See the attached licence file. Download (686kB)

Abstract

We propose a method to determine unintelligible words based on the textual context of the word determined. As there can be many different candidate words to choose from for a word, a robust, large-scale method is needed. The large scale makes the problem sensitive to spurious similarities of contexts: when the contexts of two, different words are similar. To reduce this effect, we induce structured sparsity on the words by formulating the task as a group Lasso problem. We compare this formulation to a k-nearest neighbor and a support vector machine based approach, and find that group Lasso outperforms both by a large margin. We achieve up to 75% of accuracy when determining the word from among 1000 candidate words both on the Brown corpus and on the British National Corpus. The relevance of this work is in Optical Character Recognition (OCR), where unintelligible words are often produced. Our proposed method utilizes information independent from information used in OCR and in turn, one expects that a combined approach could be very successful.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item