Dynamically identifying and removing potential stopwords from a local search query
View Patent ↗Dynamic identification and removal of potential stopwords from a search query is provided. A search query comprising plural terms is received. A possible stopword among the terms that appears in a predefined list of possible stopwords is identified. Query splits are determined based on one or more query patterns, each query split comprising a location portion using a first subset of the plural terms and a subject portion using a second subset. Either the location portion or the subject portion of the query splits is modified, by removing the possible stopword from that portion. The search engine is queried with the location portion and the subject portion for each of the query splits, and scored search results are obtained for each of the query splits. The search results returned for at least one of the query splits is selected to return as a response to the search query.
1. A machine-implemented method for dynamically identifying and removing potential stopwords from a search query, the method comprising:
receiving a search query comprising plural terms;
identifying a possible stopword among the plural terms that appears in a predefined list of possible stopwords;
determining one or more query splits by matching the search query to one or more query patterns, wherein each query pattern corresponds to a respective rule for interpreting the search query, wherein each query pattern is matched to the search query based on the respective rule, and wherein each query split comprises at least one of a location portion that uses one or more of the plural terms or a subject portion that uses one or more of the plural terms;
removing the possible stopword from at least one of the location portion or the subject portion for at least one of the query splits based on a stopword removal rule, wherein the stopword removal rule specifies whether to remove the stopword from the location portion only, the subject portion only, or both the location portion and the subject portion;
querying the search engine with the location portion and the subject portion for each of the one or more query splits, and respectively obtaining scored search results for each of the one or more query splits; and
selecting the search results returned for at least one of the query splits to return as a response to the search query.
2. The method of claim 1 , wherein the selected search results correspond to a query split in which the possible stopword was removed.
3. The method of claim 1 , wherein the selected search results correspond to a query split in which the possible stopword was not removed.
4. The method of claim 1 , wherein the selected search results are the search results with the highest score.
5. The method of claim 1 , further comprising determining whether the possible stopword is more closely associated with the location portion or the subject portion of the plurality of query splits.
6. The method of claim 5 , wherein removing the possible stopword comprises removing the possible stopword from the location portion when the possible stopword is more closely associated with the subject portion.
7. The method of claim 5 , wherein removing the possible stopword comprises removing the possible stopword from the subject portion when the possible stopword is more closely associated with the location portion.
8. A system for dynamically identifying and removing potential stopwords from a search query, the system comprising:
one or more processors; and
a non-transitory machine-readable medium comprising instructions stored therein, which when executed by the processors, cause the processors to perform operations comprising:
receiving a search query comprising plural terms;
identifying a possible stopword among the plural terms that appears in a predefined list of possible stopwords;
determining one or more query splits by matching the search query to one or more query patterns, wherein each query pattern corresponds to a respective rule for interpreting the search query, wherein each query pattern is matched to the search query based on the respective rule, and wherein each query split comprises at least one of a location portion that uses one or more of the plural terms or a subject portion that uses one or more of the plural terms;
removing the possible stopword from at least one of the location portion or the subject portion for at least one of the query splits based on a stopword removal rule, wherein the stopword removal rule specifies whether to remove the stopword from the location portion only, the subject portion only, or both the location portion and the subject portion;
querying the search engine with the location portion and the subject portion for each of the one or more query splits, and respectively obtaining scored search results for each of the one or more query splits; and
selecting the search results returned for at least one of the query splits to return as a response to the search query.
9. The system of claim 8 , wherein the selected search results correspond to a query split in which the possible stopword was removed.
10. The system of claim 8 , wherein the selected search results correspond to a query split in which the possible stopword was not removed.
11. The system of claim 8 , wherein the selected search results are the search results with the highest score.
12. The system of claim 8 , wherein removing the possible stopword comprises removing the possible stopword from the location portion when the possible stopword is more closely associated with the subject portion.
13. The system of claim 8 , wherein removing the possible stopword comprises removing the possible stopword from the subject portion when the possible stopword is more closely associated with the location portion.
14. A non-transitory machine-readable medium comprising instructions stored therein, which when executed by a processor, cause the processor to perform operations comprising:
receiving a search query comprising plural terms;
identifying a possible stopword among the plural terms that appears in a predefined list of possible stopwords;
determining one or more query splits by matching the search query to one or more query patterns, wherein each query pattern corresponds to a respective rule for interpreting the search query, wherein each query pattern is matched to the search query based on the respective rule, and wherein each query split comprises at least one of a location portion that uses one or more of the plural terms or a subject portion that uses one or more of the plural terms;
removing the possible stopword from at least one of the location portion or the subject portion for at least one of the query splits based on a stopword removal rule, wherein the stopword removal rule specifies whether to remove the possible stopword from the location portion only, the subject portion only, or both the location portion and the subject portion;
querying the search engine with the location portion and the subject portion for each of the one or more query splits, and respectively obtaining scored search results for each of the one or more query splits; and
selecting the search results returned for at least one of the query splits to return as a response to the search query, wherein the selected search results are the search results with the highest score.
15. The non-transitory machine-readable medium of claim 14 , wherein the selected search results correspond to a query split in which the possible stopword was removed.
16. The non-transitory machine-readable medium of claim 14 , wherein the selected search results correspond to a query split in which the possible stopword was not removed.
17. The non-transitory machine-readable medium of claim 14 , wherein removing the possible stopword comprises:
removing the possible stopword from the location portion when the possible stopword is more closely associated with the subject portion; and
removing the possible stopword from the subject portion when the possible stopword is more closely associated with the location portion.