Abstract: A word corpus is identified and a word probability value is associated with each word in the word corpus. A sentence is identified, candidate segmentations of the sentence are determined based on the word corpus, and the associated probability value for each word in the word corpus is iteratively adjusted based on the probability values associated with the words and the candidate segmentations.

Abstract: A system forms search results clustered by address or telephone number. When clustering by address, the system may receive a search query and identify a geographical area of interest based, at least in part, on the search query. The system may identify documents that are associated with addresses located within the geographical area of interest, group the identified documents into clusters based, at least in part, on the addresses located within the geographical area of interest, and present the clusters as the search results. When clustering by telephone number, the system may receive a search query that includes at least one portion of a telephone number and identify documents that are associated with telephone numbers that match the at least one portion of the telephone number. The system may group the identified documents into clusters based on the telephone numbers included in the identified documents and present the clusters as the search results.

Abstract: Systems and methods improve search rankings for a search query by using location data associated with queries and documents related to the search query. In one aspect, a search query is received, a location score is determined, a topical score is determined, and an ordering of documents related to the search query is determined based, at least in part, on the location score and the topical score.

Abstract: Systems and methods improve search rankings for a search query by using location data associated with queries and documents related to the search query. In one aspect, a search query is received, a location score is determined, a topical score is determined, and an ordering of documents related to the search query is determined based, at least in part, on the location score and the topical score.

Abstract: Systems and methods improve search rankings for a search query by using location data associated with queries and documents related to the search query. In one aspect, a search query is received, a location score is determined, a topical score is determined, and an ordering of documents related to the search query is determined based, at least in part, on the location score and the topical score.

Abstract: A word corpus is identified and a word probability value is associated with each word in the word corpus. A sentence is identified, candidate segmentations of the sentence are determined based on the word corpus, and the associated probability value for each word in the word corpus is iteratively adjusted based on the probability values associated with the words and the candidate segmentations.

Abstract: A geocoding component generates geographic coordinate information, such as latitude and longitude values, for postal addresses. A table includes a number of rows, each corresponding to one or more addresses. The geocoding component can quickly locate a particular row in the table based on a number of input address identifiers as the intersection of the sets rows that correspond to each of the address terms. The geocoding component may operate on addresses that are received by the geocoding component or extracted from documents.

Abstract: A system forms search results clustered by address or telephone number. When clustering by address, the system may receive a search query and identify a geographical area of interest based, at least in part, on the search query. The system may identify documents that are associated with addresses located within the geographical area of interest, group the identified documents into clusters based, at least in part, on the addresses located within the geographical area of interest, and present the clusters as the search results. When clustering by telephone number, the system may receive a search query that includes at least one portion of a telephone number and identify documents that are associated with telephone numbers that match the at least one portion of the telephone number. The system may group the identified documents into clusters based on the telephone numbers included in the identified documents and present the clusters as the search results.

Abstract: Systems and methods improve search rankings for a search query by using location data associated with queries and documents related to the search query. In one aspect, a search query is received, a location score is determined, a topical score is determined, and an ordering of documents related to the search query is determined based, at least in part, on the location score and the topical score.