Smart suggestions for query refinements
In an example embodiment, a query for search results is received, the query including at least one value for one facet, a facet defining a categorical dimension for the search results. It is then determined that the facet in the query is exclusive. In response to the determination that the facet is exclusive: for each potential facet different from the facet in the query: for each potential value in the potential facet: conditional entropy gain of the value in the query and the potential value is determined. The potential value in the potential facet that has the highest conditional entropy gain is determined, as is the potential facet with the minimum maximum conditional entropy gain. Then the potential facet with the minimum maximum is input into a machine learning model, causing the machine learning model to output one or more suggested facets to add to the query.
1. A computer-implemented method, comprising:
receiving a query for search results, the query including at least one value for one facet, a facet defining a categorical dimension for the search results;
determining that the facet in the query is exclusive;
in response to the determination that the facet is exclusive:
for each potential facet different from the facet in the query:
for each potential value in the potential facet:
determining conditional entropy gain of the value in the query and the potential value;
determining the potential value in the potential facet that has the highest conditional entropy gain;
determining the potential facet with the minimum maximum conditional entropy gain; and
inputting the potential facet with the minimum maximum into a machine learning model, causing the machine learning model to output one or more suggested facets to add to the query.
2. The computerized method of claim 1 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining how close the potential value is to a location of a searcher that created the query; and
wherein the inputting includes inputting how close the potential value is to the location of the searcher into the machine learning model.
3. The computerized method of claim 1 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining whether a region of the potential value matches a region of the value; and
wherein the inputting includes inputting whether the region of the potential value matches the region of the value into the machine learning model.
4. The computerized method of claim 1 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining whether a country of the potential value matches a country of the value; and
wherein the inputting includes inputting whether the country of the potential value matches the country of the value into the machine learning model.
5. The computerized method of claim 1 , wherein the search results are member profiles in a social networking service.
6. The computerized method of claim 5 , wherein the query is a query specifying ideal candidate profiles.
7. The computerized method of claim 1 , wherein the potential facet is always different than the facet when the facet is an exclusive facet.
8. A system comprising:
a computer-readable medium having instructions stored thereon, which; when executed by a processor, cause the system to:
receive a query for search results, the query including at least one value for one facet, a facet defining a categorical dimension for the search results;
determine that the facet in the query is exclusive;
in response to the determination that the facet is exclusive:
for each potential facet different from the facet in the query:
for each potential value in the potential facet:
determine conditional entropy gain of the value in the query and the potential value;
determine the potential value in the potential facet that has the highest conditional entropy gain;
determine the potential facet with the minimum maximum conditional entropy gain; and
input the potential facet with the minimum maximum into a machine learning model, causing the machine learning model to output one or more suggested facets to add to the query.
9. The system of claim 8 , wherein the instructions further cause the system to:
for each potential value in each potential facet, determine that the potential value is a location and in response to the determining that the potential value is a location, determine how close the potential value is to a location of a searcher that created the query; and
wherein the inputting includes inputting how close the potential value is to the location of the searcher into the machine learning model.
10. The system of claim 8 , wherein the instructions further cause the system to:
for each potential value in each potential facet, determine that the potential value is a location and in response to the determining that the potential value is a location, determine whether a region of the potential value matches a region of the value; and
wherein the inputting includes inputting whether the region of the potential value matches the region of the value into the machine learning model.
11. The system of claim 8 , wherein the instructions further cause the system to:
for each potential value in each potential facet, determine that the potential value is a location and in response to the determining that the potential value is a location, determine whether a country of the potential value matches a country of the value; and
wherein the inputting includes inputting whether the country of the potential value matches the country of the value into the machine learning model.
12. The system of claim 8 , wherein the search results are member profiles in a social networking service.
13. The system of claim 12 , wherein the query is a query specifying ideal candidate profiles.
14. The system of claim 8 , wherein the potential facet is always different than the facet if the facet is an exclusive facet.
15. A non-transitory machine-readable storage medium comprising instructions, which when implemented by one or more machines, cause the one or more machines to perform operations comprising:
receiving a query for search results, the query including at least one value for one facet, a facet defining a categorical dimension for the search results;
determining that the facet in the query is exclusive;
in response to the determination that the facet is exclusive:
for each potential facet different from the facet in the query:
for each potential value in the potential facet:
determining conditional entropy gain of the value in the query and the potential value;
determining the potential value in the potential facet that has the highest conditional entropy gain;
determining the potential facet with the minimum maximum conditional entropy gain; and
inputting the potential facet with the minimum maximum into a machine learning model, causing the machine learning model to output one or more suggested facets to add to the query.
16. The non-transitory machine-readable storage medium of claim 15 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining how close the potential value is to a location of a searcher that created the query; and
wherein the inputting includes inputting how close the potential value is to the location of the searcher into the machine learning model.
17. The non-transitory machine-readable storage medium of claim 15 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining whether a region of the potential value matches a region of the value; and
wherein the inputting includes inputting whether the region of the potential value matches the region of the value into the machine learning model.
18. The non-transitory machine-readable storage medium of claim 15 , further comprising:
for each potential value in each potential facet, determining that the potential value is a location and in response to the determining that the potential value is a location, determining whether a country of the potential value matches a country of the value; and
wherein the inputting includes inputting whether the country of the potential value matches the country of the value into the machine learning model.
19. The non-transitory machine-readable storage medium of claim 15 , wherein the search results are member profiles in a social networking service.
20. The non-transitory machine-readable storage medium of claim 19 , wherein the query is a query specifying ideal candidate profiles.