Systems and methods for improved spell checking

Abstract

The present invention leverages iterative transformations of search query strings along with statistics extracted from search query logs and / or web data to provide possible alternative spellings for the search query strings. This provides a spell checking means that can be influenced to provide individualized suggestions for each user. By utilizing search query logs, the present invention can account for substrings not found in a lexicon but still acceptable as a search query of interest. This allows a means to provide a higher quality proposal for alternative spellings, beyond the content of the lexicon. One instance of the present invention operates at a substring level by utilizing word unigram and / or bigram statistics extracted from query logs combined with an iterative search. This provides substantially better spelling alternatives for a given query than employing only substring matching. Other instances can receive input data from sources other than a search query input.

Description

CROSS REFERENCE TO RELATED APPLICATIONS [0001] This application is a continuation of U.S. patent application Ser. No. 10 / 801,968, filed on Mar. 16, 2004, entitled “SYSTEMS AND METHODS FOR IMPROVED SPELL CHECKING”, the entirety of which is incorporated herein by reference.TECHNICAL FIELD [0002] The present invention relates generally to spell checkers, and more particularly to systems and methods for improving spell checking via utilization of query logs. BACKGROUND OF THE INVENTION [0003] Interaction with automated programs, systems, and services, has become a routine part of most people's lives—especially with the advent of the Internet. Web surfing or browsing for instance may even be the “new” national pastime for a certain segment of the population. In accordance with such systems, applications such as word processing have helped many become more efficient in their respective jobs or with their personal lives such as typing a letter or e-mail to a friend. Many automated features...

AI Technical Summary

Benefits of technology

The present invention is a system and method for improving spell checking by utilizing search query logs. It uses iterative transformations of search query strings along with statistics from search query logs and web data to provide possible alternative spellings for the search query strings. This allows for a higher quality proposal of alternative spellings, beyond the content of a lexicon. The invention can tailor its suggested alternatives based on the recent history of popular concepts / queries and also tailor its corrections for a given user based on the corresponding prior query logs. It can also be used in ordinary word processors and the like. Overall, the invention provides a superior spell checking means that can influence individualized suggestions for each user.

Problems solved by technology

As many have come to find out, if the site information or the search query is entered incorrectly, the cost in time to re-navigate can become quite high.

Browser or other search queries for information present a unique problem for spell checking applications, since the queries often consist of words that may not be found in a standard spell-checking dictionary, such as artist, product, or company names.

Another problem is that a word in a query may have been entered incorrectly, but not be spelled incorrectly (for example, “and processors” instead of “amd processors”).

As such, a standard dictionary, while suitable for spell checking in the context of word processing, may not be appropriate for type-in-line and search-query spell checking.

However, for many applications where spell checking is desired (e.g., text input provided to input boxes), a standard dictionary is not optimal for the problem.

Unfortunately, a problem with this approach is that the query logs will generally also contain a large number of input errors and return substring matches that are not relevant to a user's desired search.

These dynamic behaviors cannot be accounted for utilizing traditional dictionary and search query processing.

Images