System to compare local and global contexts to check spelling of unknown word
Publication Date: 2017-Jun-07
The IP.com Prior Art Database
System to Compare Local and Global Contexts to Check Spelling of Unknown Word
Disclosed is a spell checker system that uses the local context of the target word to create search terms, which it then uses in a search of the global context, including online dictionaries, social media, internet search results, and so forth. If the system identifies a repeated use of the word in the global context, then it determines that the word is correctly spelled.
Most spell checkers do not take in to account the spellings of words that are commonly accepted in colloquial speech but are not present in dictionaries. These can include proper nouns such as product or place names, industry-specific jargon, or slang terms. A method is needed to identify the correct spellings of previously unrecognized words in a spell-checking system. The novel contribution is a smarter spell checker system that uses word suggestions based on local and global context. Local context includes message subject, body, contact information, personal dictionary, etc. The global context includes search results from the internet, social media, online dictionaries, etc. The spell check/word suggestions identify proper nouns (for example, names of people, companies, products, street addresses, etc.) or words that may not be in a standard dictionary but are common on the worldwide web. The smart spell checker searches for the word in question among the global context results (i.e., the results that are based on the search for the keywords drawn from the local context). If the system identifies a repeated use of the word in the global context, then it determines that the word is correctly spelled. The core novelty is the system for auto-processing spell check/word suggestions by using the local context of the target word to create search terms to use in a search of the global context including online dictionaries, social media, internet search results, etc. This is particularly useful for company names, product names, or other proper nouns that may not be in a "standard" dictionary. To implement the smart spell checker system in a preferred embodiment:
1. Gather the word 2. Determine the top (approximately three) keywords of the current page
A. The current page could be (not limited to): i. blog post ii. social media post iii. email conversation iv. text message conversation
B. The keywords are determined based on (not limited to): i. the words wit...