Spelling Correction
Identify and auto-correct real-world spelling errors in context. TextRazor can understand and correct slang, typos and homophones, even in brands and other names, providing scored suggestions for any errors.
Typos in names and keywords can cause huge problems for text analysis systems, especially when working with noisy Tweets or social data. Our state-of-the-art spelling correction algorithms check your content for spelling errors and generate ranked contextual corrections, with results easily integrated into your app through a simple API call.
TextRazor's Deep Spelling Corrector takes technology a step beyond the spellcheckers that you might find in a word processor. Traditional spellcheckers rely on dictionaries and hardcoded rules. TextRazor's solution offers significant advantages:
- Detect typos in millions of real world entities like people, places and brands that aren't necessarily in the dictionary.
- Identify incorrect homophones - words with the same pronunciation but different meanings (mail vs male).
- Generate a confidence score for each correction suggestion using a deep analysis of the sentence context.
- Correct and score thousands of words per second.
TextRazor uses state-of-the-art Recurrent Neural Network language modelling to help understand whole sentences, so it is able to correct words in context and generate suggestion scores based on the grammatical 'correctness' of each word in the sentence. Trained using billions of words, TextRazor knows about slang, people's name variations, and brand names.
TextRazor's language models are updated frequently, so we're constantly learning new words as language evolves.
Language Support
Spelling Correction is currently only available in English. We plan to extend support to other languages with demand, if you are interested in spelling correction for another language please let us know.
API Calls
You can try out the Spelling Correction system with your own documents through our Online Demo. On the results page under 'sentences' you can see the suggestions column populated next to each word.
The TextRazor spelling correction system is fully integrated with the TextRazor analysis pipeline. Simply add the spelling extractor to your request, and look for the 'spellingSuggestions' field with each word in the response. There is no additional charge for enabling this analysis on your documents.
If you don't need suggestions from your results you don't need to add the spelling extractor, TextRazor is already correcting obviously misspelled entities for all your requests.
