Predictive text

From Wikipedia, the free encyclopedia

Predictive text is an input technology most commonly used on mobile phones, and for accessibility. The technology allows some common words to be entered by a single keypress for each letter, as opposed to the multiple keypress approach used in the older generation of mobile phones. The intent is to simplify the writing of text messages, e-mail, entries into an address book or calendar, and the like. Theoretically, the number of keystrokes per character, on average, is comparable to using a full, unambiguous keyboard, provided that all words used (including all slang, proper nouns, abbreviations, urls, foreign-language words and so on) are in the dictionary, punctuation is ignored, and that no spelling mistakes or typing mistakes are made.^[1] In practice, however, these factors are found to be crucial to speed and accuracy, as has been demonstrated in experiments with real users. ^[2]

Contents 1 Background 2 Dictionary vs. non-dictionary systems 3 History 4 Example 5 Companies and products 6 Textonyms 7 Disambiguation failure and misspelling 8 See also 8.1 Concepts 8.2 Products 8.3 Devices 9 Notes 10 External links 10.1 References 10.2 Companies 10.3 Additional reference

[edit] Background

Short message service (SMS) permits a mobile phone user to send text messages, (also called messages, SMSes, texts, and txts) as a short message. The most common system of SMS text input is referred to as "multi-tap". Using multi-tap, a key is pressed multiple times to access the list of letters on that key. For instance, pressing the"2" key once displays an "a", twice displays a "b" and three times displays a "c". To enter two successive letters that are on the same key, the user must either pause or hit a "next" button. A user can type by pressing an alphanumeric keypad without looking at the electronic equipment display. Thus, multi-tap is easy to understand, and can be used without any visual feedback. However, multi-tap is not very efficient, requiring potentially many keystrokes to enter a single letter.

In ideal circumstances, predictive text entry software generally reduces the number of key strokes a user is required to enter a word, given the limitations expressed above. The user presses the number corresponding to each letter and, as long as the word exists in the predictive text dictionary, or is correctly disambiguated by non-dictionary systems, it will appear. For instance, pressing "4663" will typically be disambiguated as the word "good", provided that a linguistic database in English is currently in use, though alternatives such as home, hood, hoof are also valid disambiguations of the sequence of key strokes.

The most widely used systems of predictive text dictionaries are T9 and iTap. The competing Eatoni products do not use a dictionary, but rather a set of statistical rules to recreate words from keystroke sequences. Each of these systems requires a linguistic database for every input language to be supported.

[edit] Dictionary vs. non-dictionary systems

Traditional disambiguation works by referencing a dictionary of commonly used words, though Eatoni offers a dictionary-less disambiguation system. In dictionary-based systems, as the user presses the number buttons, an algorithm searches the dictionary for a list of possible words that match the keypress combination, and offers up the most probable choice. The user can then confirm the selection and move on, or use a key to cycle through the possible combinations. A non-dictionary system constructs words and other sequences of letters from the statistics of word parts.

To attempt predictions of the intended result of keystrokes not yet entered, disambiguation may be combined with a word completion facility.

A disambiguation or predictive system may include a user database for storing entered words or phrases which are not well-disambiguated by the pre-supplied database. When words are entered into the user database without direct user intervention, such systems are sometimes referred to as "learning" systems. Some disambiguation systems further attempt to correct spelling, format text or perform other automatic rewrites, with the effect of either enhancing or frustrating user efforts to enter text.

[edit] History

Predictive entry of text from a telephone keypad has been known at least since the 1970s (Smith and Goodwin, 1971). Aspects of predictive text have been patented for instance by Kondraske(1985), and as a fully functional keypad to text system for communicating with deaf people via phone in 1988 (Roy Feinson #4,754,474). Predictive text was mainly used to look up names in directories over the phone, until mobile phone text messaging came into widespread use.

[edit] Example

Consider a typical phone keypad:

Suppose a user wishes to type "The". In a traditional "multi-tap" keypad entry system, it would be necessary to do the following:

• Press 8 (tuv) once to select t.
• Press 4 (ghi) twice to select h.
• Press 3 (def) twice to select e.

Meanwhile, in a phone with predictive text, it is only necessary to:

• Press 8 once to select the (tuv) group for the first character.
• Press 4 once to select the (ghi) group for the second character.
• Press 3 once to select the (def) group for the third character.

The system updates the display as each keypress is entered to show the most probable entry. In this case, predictive text reduced the number of button presses from 5 to 3. The effect is even greater with longer, more complex words.

A dictionary-based predictive system is based on hope that the desired word is in the dictionary. That hope may be misplaced if the word differs in any way from common usage—in particular, if the word is not spelled or typed correctly, is slang, or is a proper noun. In these cases, some other mechanism must be used to enter the word.

Furthermore, the simple dictionary approach fails with agglutinative languages, where a single word doesn't necessarily represent a single semantic entity.

[edit] Companies and products

Predictive text is developed and marketed in a variety of competing products. Nuance Communications's T9 is the market leader. Other products include Motorola's iTap, Eatoni Ergonomic's LetterWise, (character, rather than word-based prediction), WordWise (word-based prediction without a dictionary), EQ3 (a Qwerty-like layout compatible with regular telephone keypads); Prevalent Devices's Phraze-It; Xrgomics' TenGO (a six-key reduced QWERTY keyboard system); Adaptxt (considers language, context, grammar and semantics); CleverTexting (statistical nature of the language, dictionary less, dynamic key allocation) ; and Oizea Type (temporal ambiguity).

[edit] Textonyms

As mentioned above, the key sequence 4663 on a telephone keypad, provided with a linguistic database in English, will generally be disambiguated as the word "good". However, the same key sequence also corresponds to other words, such as "home", "gone", "hoof", "hood" and so on. Such confusions may lead to mistaken meaning even if all of the words are typed correctly and spelled correctly. For example, "Are you home?" could be rendered as "Are you good?"

Words produced by the same combination of keypresses are technically paragrams,^[1], but may be referred to as "textonyms" (or "txtonyms",) or "T9onyms" (pronounced "tynonyms").^[3], though the phenomenon has nothing to do with T9 per se and occurs in other systems.^[4]

Reportedly, textonyms may be adopted in regular speech; for example, the use of the word "book" to mean "cool" since book is debatably considered more frequent than "cool" by some predictive text systems ,^[5] "idiom" to mean "Heino" (an abbreviation for "Heineken", used in Dublin) and "Zonino!" used to mean "Woohoo!".^{[citation needed]}

[edit] Disambiguation failure and misspelling

Textonyms in which a disambiguation systems gives more than one dictionary word for a single sequence of keystrokes, are not the only issue, or even the most important issue, limiting the effectiveness of predictive text implementations. More important, according to the above references, are words for which the disambiguation produces a single, incorrect response. The system may, for example, respond with "Blairf" upon input of 252473, when the intended word was "Blaire" or "Claire" both of which correspond to the keystroke sequence, but are not, in this example, found by the predictive text system. When mis-typings or mis-spellings occur, they are very unlikely to be recognized correctly by a disambiguation system, though error correction mechanisms, such as used on the Apple iphone keyboard, may mitigate that effect.

[edit] See also

[edit] Concepts

• Multi-tap
• Assistive technology
• Autocomplete
• Word completion
• Text entry interface
• Input method editor
• Text messaging
• SMS language
• Speech-to-Text Reporter

[edit] Products

• T9 (predictive text)
• ITap
• LetterWise
• Q9 input method – method for inputting Chinese characters on a standard mobile phone keyboard
• Adaptxt - True Next Word Prediction based on context, Supports 40 languages
• CleverTexting - Statistical predictive texting
• Panini Keypad - Texting technology for 11 regional languages of India.

[edit] Devices

• iPhone
• LG VX9400
• Nokia 5510
• Texas Instruments' Avigo 10
• List of BlackBerry Products

[edit] Notes

1. ^  Paragram - a word formed by altering a letter or group of letters in another word.^{[citation needed]}

[edit] External links

[edit] References

[edit] Companies

• T9 (Nuance Communications)
• Eatoni Ergonomics
• SureType (RIM)
• PhraseExpress text prediction software
• iTap (Motorola)
• Adaptxt
• TenGO (Xrgomics)
• (WordLogic)
• Oizea Type
• CleverTexting

[edit] Additional reference

• "Alphabetic Data Entry Via the Touch-Tone Pad: A Comment", Sidney L. Smith and Nancy C. Goodwin, The Mitre Corporation, HUMAN FACTORS, 1971, 13(2) Page 189-190
• T9 training
• GSM Helpdesk Netherlands - T9 Wordinput
• New Scientist article on textonyms
• An Australian newspaper article on textonyms
• Technical notes on iTap (including lists of textonyms)