Systems and methods for parsing search queries
View Patent ↗Computer-implemented systems and methods are provided for parsing search queries. In accordance with some embodiments, search records including character strings are retrieved, and search query templates are generated comprising sequences of categories corresponding to character substrings of the character strings. Also, in accordance with some embodiments, search queries are parsed into character substrings and matched with a search query template. The search query template may then be used to associate categories with the character substrings. A search engine may use the categories to focus or otherwise refine a search based on parsed search query.
1. A computer-implemented method for improving the relevance of search results, comprising:
receiving a character string;
identifying, by at least one processor, a sequence of character substrings in the character string;
generating, by the at least one processor, context objects for performing a search, the context objects storing associations between categories and character substrings, the generating comprising:
retrieving, from a storage device, a first search query template including a first sequence of categories,
storing, in a first context object, an association between a first category of the first sequence and a first character substring in the sequence of character substrings,
deleting the first context object based on a comparison of a second character substring in the sequence of character substrings and a second category of the first sequence,
determining that the first character substring in the sequence of character substrings corresponds to a third category in a second sequence of categories included in a second search query template retrieved from the storage device, and
storing, in a second context object, an association between the third category of the second sequence and the first character substring;
searching indexed information using at least the second context object by comparing the third category with labels and/or metadata of the indexed information to generate search results; and
providing the search results in response to the received character string.
2. The computer-implemented method of claim 1 , further comprising
determining, based on the categories of the second search query template, whether to determine if an initial character substring in the sequence of character substrings corresponds to an initial category in the second sequence of categories, or to determine if the last character substring in the sequence of character substrings corresponds to the last category in the second sequence of categories.
3. The computer-implemented method of claim 1 , the third category representing one of a preposition, street, neighborhood, county, postal code, city, state, country, or franchise.
4. The computer-implemented method of claim 1 , wherein determining that the first character substring of the sequence of character substrings corresponds to the third category in the second sequence of categories comprises one or more of:
identifying that the first character substring matches a character pattern corresponding to the third category; and
identifying that the first character substring matches a term or phrase in a listing of terms or phrases corresponding to the third category.
5. The computer-implemented method of claim 1 , further comprising:
determining that a second character substring in the sequence of character substrings corresponds to a fourth category of the second sequence of categories;
associating the fourth category with the second character substring; and
storing in the second context object an association between the fourth category and the second character substring.
6. A computer system for improving the relevance of search results, comprising:
a memory device that stores a set of instructions; and
at least one processor that executes the set of instructions, causing the computer system to perform operations comprising:
receiving a character string;
identifying a sequence of character substrings in the character string;
generating context objects for performing a search, the context objects storing associations between categories and the character substrinqs, the generating comprising:
retrieving a first search query template including a first sequence of categories,
storing, in a first context object, an association between a first category of the first sequence and a first character substring in the sequence of character substrings,
deleting the first context object based on a comparison of a second character substring in the sequence of character substrings and a second category of the first sequence,
determining that the first character substring in the sequence of character substrings corresponds to a third category in a second sequence of categories included in a second retrieved search query template, and
storing, in a second context object, an association between the third category of the second sequence and the first character substring;
searching indexed information using at least the second context object by comparing the third category with labels and/or metadata of the indexed information to generate search results; and
providing the search results in response to the received character string.
7. The computer system of claim 6 , wherein the operations further comprise determining, based on the categories of the second retrieved search query template, whether to determine if an initial character substring in the sequence of character substrings corresponds to an initial category in the second sequence of categories, or to determine if the last character substring in the sequence of character substrings corresponds to the last category in the second sequence of categories.
8. The computer system of claim 6 , the third category representing one of a preposition, street, neighborhood, county, postal code, city, state, country, franchise, or search substring.
9. The computer system of claim 6 , the operations further comprising:
determining that a second character substring in the sequence of character substrings corresponds to a fourth category of the second sequence of categories;
associating the fourth category with the second character substring; and
storing in the second context object an association between the fourth category and the second character substring.
10. A non-transitory computer-readable medium that stores a set of instructions for improving the relevance of search results that, when executed by at least one processor of a computer system, configures the computer system to perform operations comprising:
receiving a character string;
identifying a sequence of character substrings in the character string;
generating, by the at least one processor, context objects for performing a search, the context objects storing associations between categories and the character substrings, the generating comprising:
retrieving a first search query template including a first sequence of categories,
storing, in a first context object, an association between a first category of the first sequence and a first character substring in the sequence of character substrings,
deleting the first context object based on a comparison of a second character substring in the sequence of character substrings and a second category of the first sequence,
determining that the first character substring in the sequence of character substrings corresponds to a third category in a second sequence of categories included in a second retrieved search query template, and
storing, in a second context object, an association between the third category of the second sequence and the first character substring;
searching indexed information using at least the second context object by comparing the third category with labels and/or metadata of the indexed information to generate search results; and
providing the search results in response to the received character string.
11. The computer-readable medium of claim 10 , the operations further comprising determining, based on the categories of the second retrieved search query template, whether to determine if an initial character substring in the sequence of character substrings corresponds to an initial category in the second sequence of categories, or to determine if the last character substring in the sequence of substrings corresponds to the last category in the second sequence of categories.
12. The computer-readable medium of claim 10 , the third category representing one of a preposition, street, neighborhood, county, postal code, city, state, country, or franchise.
13. The computer-readable medium of claim 11 , wherein determining that the first character substring of the sequence of character substrings corresponds to the third category in the second sequence of categories comprises one or more of:
identifying that the first character substring matches a character pattern corresponding to the third category; and
identifying that the first character substring matches a term or phrase in a listing of terms or phrase corresponding to the third category.
14. The computer-readable medium of claim 10 , the method further comprising:
determining that a second character substring in the sequence of character substrings corresponds to a fourth category of the second sequence of categories;
associating the fourth category with the second character substring; and
storing in the second context object an association between the fourth category and the second character substring.