Probabilistic character disambiguation for reduced keyboards using small text samples

Abstract

Reduced keyboards are text typing keyboards which contain fewer than 26 alphabetic keys, and which may therefore be accessed and used by certain physically disabled persons more easily than a conventional "QWERTY" typing keyboard. Automatic character disambiguation systems enable text to be typed upon a reduced keyboard with a keying efficiency approaching one key/character, despite the fact that each key represents more than one alphabetic character. Existing disambiguation systems typically use probabilistic models of character sequences (n-grams) from representative text samples to predict the next character, and hence disambiguate among the different characters on each key. N-grams for such disambiguation models have been extracted previously from large (1 million word) text corpora. The research reported here shows that a much smaller corpus of limited domain can be used with similar results, thus facilitating development of disambiguation systems by eliminating the need for a large corpus. Four reduced keyboard layouts are compared, three of which were used in earlier research on character disambiguation in the Dutch and English languages, and a fourth based on character frequency, which achieves similar efficiency to the first three. Models of different order are compared (the "order" of the model being determined by the length of the longest n-grams in it), the principal result being that, while higher-order models (containing long n-grams) give better performance than lower-order models (which contain only short n-grams), lower-order n-grams can contribute significantly to the disambiguation performance of a higher-order model, and should therefore be included in order to maximize disambiguation efficiency.

Ambiguous keyboards for AAC

“…Minneman's work was quickly built upon, with a number of authors investigating the topic (Kondraske & Shennib 1986;Sh Levine et al 1987;Foulds 1987;Kreifeldt et al 1989; Sh Levine & Goodenough-Trepagnier 1990;Arnott 1992). Areas investigated by these authors include using syllable level disambiguation (i.e.…”

“…In addition, optimising layouts on reduced key sets can allow ambiguous text entry to approach 1 keystroke per letter (Arnott, 1992).…”

Assistive Technology and Affective Mediation

“…These keyboards prove to be extremely useful, for example, when sending SMS (Short Message Service) communications on mobile phones, as well as helping users with special needs. The usual key distribution is the T9 keyboard 1 , however, there are studies that try to obtain better distributions (Arnott & Javed, 1992;Foulds, Soede, & Van Balkom, 1987;Gardeazabal, 2000;Lesher, Moulton, & Higginbotham, 1998a;Levine, Goodenough-Trepagnier, Getschow, & Minneman, 1987). The disambiguation that the reduced keyboards achieve can also be used as a prediction system.…”

Modelling text prediction systems in low- and high-inflected languages

“…This is the case with the T9 keyboard (http://www.T9.com), which is a reduced keyboard commonly used for composing SMS messages in mobile phones. Text prediction can also be applied to other types of reduced keyboards Arnott and Javed, 1992). In the area of assistive communication, a predictor is a system which attempts to anticipate the next block of characters (letters, syllables, words, sentences, etc.)…”