Systems and methods for performing geo-search and retrieval of electronic point-of-interest records using a big index
View Patent ↗Methods and systems for providing a geo-search engine capability for large point-of-interest record datasets are disclosed. These methods and systems employ a Partition-by-Query index containing key-values pairs corresponding to keys reflecting concept-ordered geo-search phrases and values reflecting ordered lists of point-of-interest record references that are responsive to the concept-ordered geo-search phrase in a corresponding key. A large Partition-by-Query index may be partitioned across multiple servers depending on the size of the index, or the size of the index may be reduced by compressing query-references pairs into clusters. The methods and systems described herein may provide suggestions and spelling corrections to the user, thereby improving the user's geo-search engine experience while meeting user expectations for geo-search quality and responsiveness.
1. A method for configuring a search engine to provide spelling corrections or suggestions to search queries, the method comprising:
generating, by a computing device, one or more residual strings associated with a first token in record corpus, wherein each of the one or more residual strings represents a variation of the first token and is associated with a first weight reflecting a degree of variation from the first token;
generating, by the computing device, a first list associated with the first token, the first list comprising the one or more residual strings and the first weights associated with the one or more residual strings;
generating, by the computing device, a second list associated with at least one of the one or more residual strings;
wherein the second list comprises one or more second tokens;
wherein each of the one or more second tokens represents a variation of the at least one of the one or more residual strings; and
wherein each of the one or more second tokens is associated, in the second list, with a second weight reflecting a degree of variation from the at least one of the one or more residual strings; and
generating, by the computing device, a third list associated with the first token, wherein the third list comprises the one or more second tokens;
wherein at least one of the one or more second tokens is associated, in the third list, with a third weight reflecting a relationship between the first token and the at least one of one or more second tokens; and
wherein the third weight is generated by a combination of the first weight and the second weight; and
determining, by the computing device, a spelling correction or suggestion to the search query based on the one or more third weights.
2. The method of claim 1 , wherein propagating tokens generating a third list further comprises discarding redundant residual strings.
3. The method of claim 2 , further comprising the step of generating, by a computing device, a bloom filter.
4. An apparatus for providing geo-search engine suggestions to users, the apparatus comprising:
at least one processor;
a network connection; and
a memory comprising instructions that, when executed by the at least one processor, perform the operations of:
generating, by the at least one processor, one or more residual strings associated with a first token in a record corpus of point of interest records, wherein each point of interest record comprises a name and a location formed from tokens;
wherein each of the one or more residual strings represents a variation of the first token and is associated with a first weight reflecting a degree of variation from the first token;
generating, by the at least one processor, a first list associated with the first token, the first list comprising the one or more residual strings and the first weights associated with the one or more residual strings;
generating, by the at least one processor, a second list associated with at least one of the one or more residual strings;
wherein the second list comprises one or more second tokens;
wherein each of the one or more second tokens represents a variation of the at least one of the one or more residual strings; and
wherein each of the one or more second tokens is associated, in the second list, with a second weight reflecting a degree of variation from the at least one of the one or more residual strings; and
generating, by the at least one processor, a third list associated with the first token, wherein the third list comprises the one or more second tokens;
wherein at least one of the one or more second tokens is associated, in the third list, with a third weight reflecting a relationship between the first token and the at least one of one or more second tokens; and
wherein the third weight is generated by a combination of the first weight and the second weight; and
determining, by the at least one processor, a geo-search suggestion to a geo-search query for a point of interest based on the one or more third weights.
5. The apparatus of claim 4 , wherein generating a third list further comprises discarding redundant residual strings.
6. The apparatus of claim 5 , the memory further comprising instructions that, when executed, perform the step of generating a bloom filter.
7. A non-transitory computer-readable medium containing instructions that, when executed, perform a method comprising the steps of:
generating, by at least one processor, one or more residual strings associated with a first token in a record corpus of point of interest records, wherein each point of interest record comprises a name and a location formed from tokens;
wherein each of the one or more residual strings represents a variation of the first token and is associated with a first weight reflecting a degree of variation from the first token;
generating, by the at least one processor, a first list associated with the first token, the first list comprising the one or more residual strings and the first weights associated with the one or more residual strings;
generating, by the at least one processor, a second list associated with at least one of the one or more residual strings;
wherein the second list comprises one or more second tokens;
wherein each of the one or more second tokens represents a variation of the at least one of the one or more residual strings; and
wherein each of the one or more second tokens is associated, in the second list, with a second weight reflecting a degree of variation from the at least one of the one or more residual strings; and
generating, by the at least one processor, a third list associated with the first token, wherein the third list comprises the one or more second tokens;
wherein at least one of the one or more second tokens is associated, in the third list, with a third weight reflecting a relationship between the first token and the at least one of one or more second tokens; and
wherein the third weight is generated by a combination of the first weight and the second weight; and
determining, by the at least one processor, a geo-search suggestion to a geo-search query for a point of interest based on the one or more third weights.
8. The non-transitory computer-readable medium of claim 7 , wherein generating a third list further comprises discarding redundant residual strings.
9. The non-transitory computer-readable medium of claim 8 further containing instructions that, when executed, perform a step of generating a bloom filter.