Specialized search system and method for matching a student to a tutor
A system and method for matching students to tutors is disclosed. In particular, a user submits a query and in response a plurality of tutors are identified responsive to the search query. The identified tutors are then ranked based on relevance to the search query. The identified tutors are then presented to the user in an order based on the relevance score assigned to each tutor.
1. A computer-implemented method performed by at least one processor comprising the steps of:
receiving a search query from a user;
parsing the search query to produce one or more query primitives, wherein parsing the search query includes removing any stop words from the search query;
accessing a database including a table of tutor profiles, each tutor profile in the table of tutor profiles corresponding to a specific tutor, and the database including additional tutor information regarding the specific tutors, the additional tutor information including a track record of a specific tutor to convert a prospective student to an actual student;
obtaining a plurality of tutor profiles from the database that are responsive to the search query;
extracting from each of the tutor profiles at least one subject and one or more related words wherein said one or more related words are related to the at least one subject;
computing a relevance score for each of the tutor profiles using the at least one subject, the one or more related words, the one or more query primitives, and the track record of the specific tutor to convert a prospective student to an actual student;
wherein the step of computing comprises determining the equivalence of the one or more query primitives with the at least one subject and the one or more related words, and assigning a relevance score to a tutor profile based on how alike the at least one subject and the one or more related words are to the one or more query primitives and based on the track record of the specific tutor to convert a prospective student to an actual student; and
presenting the tutor profiles to a user so that tutor profiles with the highest relevance scores are presented to the user first.
2. The method of claim 1 , further comprising a step of accessing a historical search database including a historical search table relating historical search terms to subjects that students actually received lessons in; and wherein the step of extracting comprises querying the historical search table, comparing the one or more query primitives to the historical search terms, and when a hit occurs, adding a subject that students actually received lessons in to the one or more query primitives.
3. The method of claim 1 , wherein the step of generating the list of query primitives comprises identifying a part of speech of a particular word in a sub-query.
4. The method of claim 3 , wherein, when the particular word is identified as a pure adjective, the step of generating the list of query primitives further comprises eliminating the pure adjective.
5. The method of claim 2 , wherein the table relating historical search terms to subjects includes more than one occurrence of a particular search term.
6. The method of claim 5 , wherein the particular search term occurs at least once with a first resultant subject and at least once with a second resultant subject different from the first resultant subject.
7. The method of claim 5 , wherein the particular search term occurs more than once with the same resultant subject.
8. The method of claim 1 , wherein the stop words include conjunctions.
9. A computer-implemented method performed by at least one processor comprising the steps of:
receiving a search query from a user;
parsing the search query to produce one or more query primitives, wherein parsing the search query includes removing any stop words from the search query;
accessing a database including a table of tutor profiles;
obtaining a plurality of tutor profiles from the database that are responsive to the search query;
extracting from each of the tutor profiles at least one subject and one or more related words wherein said one or more related words are related to the at least one subject and further wherein the one or more related words are related to the at least one subject using historical search terms;
computing a relevance score for each of the tutor profiles using the at least one subject, the one or more related words, and the one or more query primitives;
wherein the step of computing comprises determining an equivalence of the one or more query primitives with the at least one subject and the one or more related words, and assigning a higher relevance score to a tutor profiles when the at least one subject and the one or more related words of the tutor profile are more like the one or more query primitives than those of a different tutor profile are like the one or more query primitives; and
presenting the tutor profiles to a user so that tutor profiles with the highest relevance scores are presented to the user first.
10. The method of claim 9 further comprising a step of accessing a historical search database including a historical search table relating historical search terms to subjects that students actually received lessons in; and wherein the step of extracting comprises querying the historical search table, comparing the one or more query primitives to the historical search terms, and when a hit occurs, adding a subject that students actually received lessons in to the one or more query primitives.
11. The method of claim 9 , wherein the step of generating the list of query primitives comprises identifying a part of speech of a particular word in a sub-query.
12. The method of claim 11 , wherein, when the particular word is identified as a pure adjective, the step of generating the list of query primitives further comprises eliminating the pure adjective.