Machine translation for query expansion

A computer-implemented method comprising:

• receiving an initial search query including a first term;
• determining that the first term is associated with a second term occurring in a translated search query including one or more context terms, the one or more context terms occurring adjacent to the second term in the translated search query, the second term being designated as a synonym for the first term when the first term occurs adjacent to the one or more context terms in search queries;
• determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query;
• in response to determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query, generating an expanded search query that includes the second term;
• obtaining search results using the expanded search query to search a collection of documents;
• and
• providing the search results in response to the initial search query.

A computer-readable storage device having stored thereon instructions that, when executed by a computer, cause the computer to perform operations comprising:

• receiving an initial search query including one or more terms;
• using statistical machine translation to translate the initial search query into a translated search query, the translated search query being different from the initial search query and being in a same natural language as the initial search query;
• identifying a first term occurring in the initial search query that was replaced by a corresponding second term occurring in the translated search query;
• determining that one or more context terms occurring adjacent to the first term in the initial search query also occur adjacent to the second term in the translated search query;
• and
• in response to determining that one or more context terms occurring adjacent to the first term in the initial search query also occur adjacent to the second term in the translated search query, designating the second term as a synonym of the first term when the first term occurs adjacent to the one or more context terms in search queries, wherein a search engine expands a particular search query to include the second term in response to a determination that the first term occurs adjacent to the one or more context terms in the particular search query.

A computer-readable storage device having stored thereon instructions that, when executed by a computer, cause the computer to perform operations comprising:

• receiving an initial search query including a first term;
• determining that the first term is associated with a second term occurring in a translated search query including one or more context terms, the one or more context terms occurring adjacent to the second term in the translated search query, the second term being designated as a synonym for the first term when the first term occurs adjacent to the one or more context terms in search queries;
• determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query;
• in response to determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query, generating an expanded search query that includes the second term;
• obtaining search results using the expanded search query to search a collection of documents;
• and
• providing the search results in response to the initial search query.

A system comprising:

• one or more computers;
• and
• a computer-readable storage device having stored thereon instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
• receiving an initial search query including one or more terms;
• using statistical machine translation to translate the initial search query into a translated search query, the translated search query being different from the initial search query and being in a same natural language as the initial search query;
• identifying a first term occurring in the initial search query that was replaced by a corresponding second term occurring in the translated search query;
• determining that one or more context terms that occur adjacent to the first term in the initial search query also occur adjacent to the second term in the translated search query;
• and
• in response to determining that one or more context terms occurring adjacent to the first term in the initial search query also occur adjacent to the second term in the translated search query, designating the second term as a synonym of the first term when the first term occurs adjacent to the one or more context terms in search queries, wherein a search engine expands a particular search query to include the second term in response to a determination that the first term occurs adjacent to the one or more context terms in the particular search query.

A system comprising:

• one or more computers;
• and
• a computer-readable storage device having stored thereon instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:
• receiving an initial search query including a first term;
• determining that the first term is associated with a second term occurring in a translated search query including one or more context terms, the one or more context terms occurring adjacent to the second term in the translated search query, the second term being designated as a synonym for the first term when the first term occurs adjacent to the one or more context terms in search queries;
• determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query;
• in response to determining that the first term occurs in the initial search query adjacent to the one or more context terms that also occur adjacent to the second term in the translated search query, generating an expanded search query that includes the second term;
• obtaining search results using the expanded search query to search a collection of documents;
• and
• providing the search results in response to the initial search query.