In computing, spell checker (or spell checker ) is an application program that marks words in documents that may not be spelled correctly. The spell checker may be on its own, capable of operating in a text block, or as part of a larger app, such as a word processor, email client, electronic dictionary, or search engine.
Video Spell checker
Design
The basic spell checker performs the following process:
- It scans the text and extracts the words it contains.
- Then compare each word with a list of properly spelled words (ie dictionaries). It may contain only a list of words, or it may contain additional information, such as connecting points or lexical and grammatical attributes.
- Additional steps are language dependent algorithms to handle morphology. Even for a slightly influenced language such as English, spell checkers will need to consider various forms of the same word, such as plurals, verbal forms, contractions, and possessions. For many other languages, such as featuring more complex agglutination and declarations and conjugations, this part of the process is more complicated.
It is unclear whether morphological analysis - allowing for different forms of words depending on the role of grammar - provides significant benefits for English, although its benefits for highly synthetic languages ​​such as German, Hungarian or Turkish are obvious.
In addition to these components, the program user interface will allow the user to approve or reject the replacement and modify the operation of the program.
An alternative type of spell checker uses only statistical information, such as n-grams, to recognize errors, not spelled words correctly. This approach usually requires a lot of effort to obtain adequate statistical information. The main advantages include less runtime storage and the ability to correct errors in words that are not included in the dictionary.
In some cases, the spell checker uses a misspelled list and incorrect suggestions for the misspellings; this less flexible approach is often used in paper-based correction methods, such as see also encyclopaedic entries.
The grouping algorithm has also been used for spell checking in combination with phonetic information.
Maps Spell checker
History
The study expanded back to 1957, including spell checkers for bitmap cursive writing and special applications to find records in the database regardless of incorrect entries. In 1961, Les Earnest, who led the research on novice technology, saw the need to include the first spelling checker accessing an acceptable list of 10,000 words. Ralph Gorin, a graduate student under Earnest at the time, created the first spell checker program written as an application program (not research) for general English text: Spell for DEC PDP-10 at the Stanford University Artificial Intelligence Laboratory, in February 1971 Gorin writes EJA in assembly language, for quicker action; he creates the first spelling corrector by searching for a list of words for a perfectly plausible spelling that is different from a single letter or transposition of adjacent letters and presenting it to the user. Gorin makes SPELL publicly accessible, as it did with most of the SAIL (Stanford Artificial Intelligence Laboratory) programs, and soon spread across the globe via the new ARPAnet, about ten years before personal computers began to be used in general. Spell, its algorithm and data structure inspire the Unix ispell program.
The first spelling checker was widely available on mainframe computers in the late 1970s. A group of six linguists from Georgetown University developed the first spelling checking system for IBM companies.
The first spell checker for personal computers appeared for CP/M and TRS-80 computers in 1980, followed by packages for IBM PCs after its introduction in 1981. Developers such as Maria Mariani, Random House, Soft-Art, Microlytics, Proximity, Circle Noetics , and Software Reference rushes OEM packages or end-user products into the fast-growing software market, especially for PCs but also for Apple Macintosh, VAX, and Unix. On PCs, this spell checker is a stand-alone program, many of which can be run in TSR mode from within a word processing package on a PC with sufficient memory.
However, the market for stand-alone packages is short-lived, because in the mid-1980s developers of popular word processing packages such as WordStar and WordPerfect have included spell checkers in their packages, mostly licensed from the above companies, which quickly expanded support only from English to Europe and finally even Asian languages. However, this requires an increase in sophistication in software morphology routines, especially with regard to highly agglutinative languages ​​such as Hungarian and Finnish. Although the size of word processing markets in countries like Iceland may not justify investing in the implementation of spell checkers, companies like WordPerfect are still trying to localize their software for as many national markets as possible as part of their global marketing strategy.
Firefox 2.0, a web browser, has spell check support for user-written content, such as when editing Wikitext, writing on many websites, blogs, and social networking websites. The Google Chrome, Konqueror, and Opera web browsers, Kmail email clients and Pidgin instant messaging clients also offer spell check support, transparently using GNU Aspell as their machine. Mac OS X now has a spell checking system, extending services to almost all third party packages and applications.
Function
The first spell checker is "verifier" instead of "corrector". They do not offer suggestions for spelled words. It helps for typos but not so helpful for logical or phonetic errors. The challenge that developers face is the difficulty of offering useful suggestions for misspelled words. This requires reducing the word to a skeletal form and applying pattern matching algorithms.
It may seem logical that where the spell check dictionary notices, "the bigger, the better," so that the correct words are not marked as wrong. But in practice, the optimal size for English seems to be about 90,000 entries. If there is more than this, incorrectly spelled words may be missed because they are considered wrong by others. For example, a linguist may determine based on the linguistic corpus that the word baht is more often misspelled bathe or bat than a reference to Thai currency. Therefore, it is usually more useful if some people who write about the Thai currency experience less discomfort than if the spelling mistakes of more people discussing the baths are ignored.
The first MS-DOS spell checker is mostly used in checking mode from within the word processing package. After preparing the document, the user scans the text looking for a misspelling. Then, however, batch processing is offered in packages like CoAuthor that are short-lived and allow the user to view the results after the document is processed and only correct the words that are known to be incorrect. When memory and processing power become abundant, spell checking is done in the background in an interactive way, as has happened with the Sector Software produced Spellbound Program released in 1987 and Microsoft Word since Word 95.
In recent years, spell checkers have become more sophisticated; some are now able to recognize simple grammatical errors. However, even in its best condition, they rarely capture all the errors in the text (such as homophone errors) and will mark neologism and foreign words as misspellings. However, spell checkers can be considered as a kind of foreign language writing aid that can be relied upon by foreign language learners to detect and correct their misspellings in the target language.
Spelling non-English language
English is unusual because most words used in formal writing have one spelling that can be found in a typical dictionary, with the exception of some jargon and modified words. In many languages ​​words are often combined into a combination of new words. In Germany, compound nouns are often created from other nouns that exist. Some scripts do not clearly separate one word from another, which requires a word separation algorithm. Each presents a unique challenge for non-English spelling checkers.
Context-sensitive spell checker
- Recently, research has focused on developing algorithms that are able to recognize misspelled words, even if the word itself is in the vocabulary, based on the context of the words around it. Not only does this allow words like the one in the above poem to be captured, but it reduces the damaging effects of enlarging the dictionary, allowing more words to be recognized. For example, baht in the same paragraph as Thai or Thai will not be recognized as misspelling bathe . The most common example of errors caught by such a system is homophone errors, such as bold words in the following sentence:
- They will come too sea if reel .
The most successful algorithms to date are Andrew Golding and Dan Roth "Winnow based spelling correction algorithms", published in 1999, which are able to recognize about 96% of context-sensitive spelling errors, in addition to the usual non-word misspellings. The context-sensitive spell checker appears in Microsoft Office 2007, Google Wave, and in the Ghotit Dyslexia Software context the spell checker is set for people with dyslexia.
Criticism
Several technological and computer criticisms have tried to connect spell checkers to skill-loss trends in writing, reading, and speaking. They claim that computers have made people lazy, often not correcting written work except for a simple pass through a spell checker. Proponents claim that these changes can actually benefit the community, by making writing and learning new languages ​​more accessible to ordinary people. They claim that missing skills with the use of automatic spell checkers are replaced by better skills, such as faster and more efficient research skills. Other tech supporters point out that these skills are not lost by people who often use written language in their work, such as writers and journalists.
An example of folly relying entirely on the spell checker is shown in the Spell Checker Pronunciation above. Originally composed by Dr. Jerrold H. Zar in 1991, assisted by Mark Eckman, with an original length of 225 words, and containing 123 wrongly used words. According to the most unsophisticated spell checker, the poem is valid, but most people can say at a glance that most words are used incorrectly. As a result, spell checkers are sometimes mocked as spilling chuckers or names that resemble misspellings.
Not all critics are opponents of technological progress. An article based on research by Galletta et al. reports that higher verbal skills are required for highest performance when using spell checkers. The theory states that only authors with higher verbal skills can recognize and ignore positive errors or wrong suggestions. However, it was found that those with higher skills lost performance excellence without their help in various categories of errors, underperforming such low verbs with spell-checkers turned on. The conclusion shows some evidence of loss of skills.
See also
- Matching strings that are close to
- Cupertino Effects
- Grammar checker
- Record link problem
- Spelling suggestions
- Words (Unix)
References
External links
- Spell checklist in Curlie (based on DMOZ)
- Norvig.com, "How to Write Spelling Corrector", by Peter Norvig
- BBK.ac.uk, "Computer spell checking", by Roger Mitton
- CBSNews.com, Check Spelling Crutch Curtails Correctness, by Lloyd de Vries
- NIU.edu, Prospective Pullet Suprise - Complete corrected poem
- Microsoft Word Spelling and Grammar Check Demonstration
- Grammar Checker
Source of the article : Wikipedia
0 Comments