Approximate template matching for natural language queries
Systems and methods provide a media guidance application that recognizes a plurality of natural language search queries for identifying a set of search results. For example, a user may want to determine when the Yankees are playing their next baseball game. The user may structure their query in multiple ways, such as, “When are the Yankees playing?” “What time is the Yankees game?” “When is the next Yankees baseball game?” The user would expect the same result, a description of when the Yankees are playing, regardless of how the query is structured. The systems and methods enable a user to use a plurality of search queries when searching for items or information to get desired results.
1. A method for performing a search, the method comprising:
receiving, via an input device, an input query comprising a first word and a reference word;
identifying a first reference corpus that comprises the reference word;
calculating, based on the first reference corpus, a first value indicative of a first likelihood of the first word occurring with the reference word;
retrieving a search query template associated with the reference word, wherein the search query template comprises a second word;
calculating, based on a second reference corpus that comprises the reference word, a second value indicative of a second likelihood of the second word occurring with the reference word;
calculating a difference value based on subtracting the first value indicative of the first likelihood from the second value indicative of the second likelihood;
selecting, based on the difference, the search query template;
retrieving search results based on the selected search query template and the reference word; and
generating for display the retrieved search results.
2. The method of claim 1 , wherein selecting, based on the difference, the search query template comprises:
determining whether the difference is less than a threshold difference; and
in response to determining that the difference is less than the threshold difference, selecting the search query template.
3. The method of claim 2 , wherein the difference is a first difference, further comprising:
in response to determining that the first difference is not less than the threshold difference, selecting a third word in the input query and a fourth word in the search query template;
determining a third likelihood of the third word occurring with the reference word;
determining a fourth likelihood of the fourth word occurring with the reference word;
computing a second difference between the third likelihood and the fourth likelihood; and
in response to determining that the second difference is less than the threshold difference, selecting the search query template for retrieving the search results.
4. The method of claim 2 , wherein the search query template is a first search query template and the difference is a first difference, further comprising:
in response to determining that the first difference is not less than the threshold difference, retrieving a second search query template, wherein the second search query template comprises a third word associated with a third likelihood of the third word occurring with the reference word;
computing a second difference between the first likelihood and the third likelihood; and
in response to determining that the second difference is less than the threshold difference, selecting the second search query template for retrieving the search results.
5. The method of claim 1 , wherein calculating the first value indicative of the first likelihood comprises computing a first representation vector corresponding to a degree to which the first word occurs with the reference word in the first reference corpus, and wherein calculating the second value indicative of the second likelihood comprises computing a second representation vector corresponding to a degree to which the second word occurs with the reference word in the second reference corpus.
6. The method of claim 5 , wherein calculating the difference value comprises computing a distance between the first representation vector and the second representation vector, and wherein selecting the search query template comprises: determining whether the distance is less than a threshold distance; and
in response to determining that the distance is less than the threshold distance, selecting the search query template.
7. The method of claim 6 , wherein computing the distance between the first representation vector and the second representation vector comprises:
comparing a first value in the first representation vector with a corresponding second value in the second representation vector to determine whether the first value and the second value differ by less than a threshold value;
in response to determining that the values differ by less than the threshold value, leaving the distance unchanged; and
in response to determining that the values do not differ by less than the threshold value, computing a degree to which the values are dissimilar and incrementing the distance based on the degree to which the values are dissimilar.
8. The method of claim 6 , wherein computing the distance between the first representation vector and the second representation vector is based on an orientation of the first representation vector with respect to the second representation vector in a vector space, such that the distance is a measure of an angle between the first representation vector and the second representation vector.
9. The method of claim 1 , further comprising:
subsequent to receiving the input query, identifying the reference word by:
identifying a part of speech associated with a candidate reference word in the input query; and
in response to determining that the candidate reference word is a subject of the input query, identifying the candidate reference word as the reference word.
10. The method of claim 1 , further comprising:
comparing the first word to a list of stop words to determine whether the first word matches a stop word of the list of stop words; and
in response to determining that the first word matches the stop word, selecting a third word of the input query instead of the first word to determine a likelihood of the third word occurring with the reference word.
11. A system for performing a search, the system comprising:
control circuitry configured to:
receive an input query comprising a first word and a reference word;
identify a first reference corpus that comprises the reference word;
calculate, based on the first reference corpus, a first value indicative of a first likelihood of the first word occurring with the reference word;
retrieve a search query template associated with the reference word, wherein the search query template comprises a second word;
calculate, based on a second reference corpus that comprises the reference word, a second value indicative of a second likelihood of the second word occurring with the reference word;
calculate a difference value based on subtracting the first value indicative of the first likelihood from the second value indicative of the second likelihood;
select, based on the difference, the search query template;
retrieve search results based on the selected search query template and the reference word; and
generate for display the retrieved search results.
12. The system of claim 11 , wherein the control circuitry is further configured to select, based on the difference, the search query template by:
determining whether the difference is less than a threshold difference; and
in response to determining that the difference is less than the threshold difference, selecting the search query template.
13. The system of claim 12 , wherein the difference is a first difference, wherein the control circuitry is further configured to:
in response to determining that the first difference is not less than the threshold difference, select a third word in the input query and a fourth word in the search query template;
determine a third likelihood of the third word occurring with the reference word;
determine a fourth likelihood of the fourth word occurring with the reference word;
compute a second difference between the third likelihood and the fourth likelihood; and
in response to determining that the second difference is less than the threshold difference, select the search query template for retrieving the search results.
14. The system of claim 12 , wherein the search query template is a first search query template and the difference is a first difference, wherein the control circuitry is further configured to:
in response to determining that the first difference is not less than the threshold difference, retrieve a second search query template, wherein the second search query template comprises a third word associated with a third likelihood of the third word occurring with the reference word;
compute a second difference between the first likelihood and the third likelihood; and
in response to determining that the second difference is less than the threshold difference, select the second search query template for retrieving the search results.
15. The system of claim 11 , wherein the control circuitry is further configured to calculate the first value indicative of the first likelihood by computing a first representation vector corresponding to a degree to which the first word occurs with the reference word in the first reference corpus, and wherein the control circuitry is further configured to calculate the second value indicative of the second likelihood by computing a second representation vector corresponding to a degree to which the second word occurs with the reference word in the second reference corpus.
16. The system of claim 15 , wherein the control circuitry is further configured to calculate the difference value by computing a distance between the first representation vector and the second representation vector, and wherein the control circuitry is further configured to select the search query template by:
determining whether the distance is less than a threshold distance; and
in response to determining that the distance is less than the threshold distance, selecting the search query template.
17. The system of claim 16 , wherein the control circuitry is further configured to compute the distance between the first representation vector and the second representation vector by:
comparing a first value in the first representation vector with a corresponding second value in the second representation vector to determine whether the first value and the second value differ by less than a threshold value;
in response to determining that the values differ by less than the threshold value, leaving the distance unchanged; and
in response to determining that the values do not differ by less than the threshold value, computing a degree to which the values are dissimilar and incrementing the distance based on the degree to which the values are dissimilar.
18. The system of claim 16 , wherein computing the distance between the first representation vector and the second representation vector is based on an orientation of the first representation vector with respect to the second representation vector in a vector space, such that the distance is a measure of an angle between the first representation vector and the second representation vector.
19. The system of claim 11 , wherein the control circuitry is further configured to:
subsequent to receiving the input query, identify the reference word by:
identifying a part of speech associated with a candidate reference word in the input query; and
in response to determining that the candidate reference word is a subject of the input query, identifying the candidate reference word as the reference word.
20. The system of claim 11 , wherein the control circuitry is further configured to:
compare the first word to a list of stop words to determine whether the first word matches a stop word of the list of stop words; and
in response to determining that the first word matches the stop word, select a third word of the input query instead of the first word to determine a likelihood of the third word occurring with the reference word.