Search experience management system
A system and method to train one or more components of a search algorithm based on feedback received from a user system. The system and method provide, via an interface, a first search query comprising one or more terms and a first query response generated by a component of the search algorithm. The system and method receive, via the interface, a label corresponding to the query response. An override action is generated in view of the label. In response to a second query comprising the one or more terms, generating a second query response in accordance with the override action.
1 . A method comprising:
receiving, at a computer system at a first time, a first search query from a user;
generating, via at least one processor of the computer system executing a search algorithm, a first query response comprising a first sub response and a second sub response in response to the first search query, wherein the first sub response is generated by a first component of the search algorithm and the second sub response is generated by a second component of the search algorithm;
receiving, at the computer system from the user, a first and a second feedback each corresponding to the first and the second sub response;
associating, via the at least one processor, each of the first and the second feedback with the first and the second component of the search algorithm;
generating, via the at least one processor in response to the first and the second feedback, an override action corresponding to at least one of the first component and the second component and with one or more terms of the first search query, the override action indicating that a second query response generated by the corresponding at least one of the first component and the second component in response to a second search query including the one or more terms is to be overridden;
storing, via the at least one processor, the override action in a data store comprising an override cache;
training, via the at least one processor executing a machine learning algorithm using the first and the second feedback as input, a model associated with the first component and the second component, wherein the training of the first component is based on the first feedback corresponding to the first sub response and the training of the second component is based on the second feedback corresponding to the second sub response;
modifying, via the at least one processor using the model, the search algorithm such that the first and the second component are updated, resulting in a modified search algorithm;
receiving, at the computer system at a second time which is distinct from the first time, the second search query comprising the one or more terms of the first search query, the second search query being equivalent to the first search query; and
generating, via at least one processor of the computer system executing the modified search algorithm and in accordance with the override action, the second query response in response to the second search query, the second query response being distinct from the first query response.
2 . The method of claim 1 , wherein the first feedback comprises one of an accept label, a rejection label, an override label, or a revise label.
3 . The method of claim 1 , wherein the first component of the search algorithm comprises a portion of the search algorithm configured to perform a filter function of the search algorithm.
4 . The method of claim 1 , wherein the first component comprises a natural language processing (NLP) filter component configured to perform NLP filtering associated with the first search query.
5 . The method of claim 1 , the first feedback comprises a revise label.
6 . The method of claim 1 , wherein the second search query is identical to the first search query.
7 . A system comprising:
at least one processor; and
a non-transitory computer-readable storage medium having instructions stored which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
receiving, at a first time, a first search query from a user;
generating, by executing a search algorithm, a first query response comprising a first sub response and a second sub response in response to the first search query, wherein the first sub response is generated by a first component of the search algorithm and the second sub response is generated by a second component of the search algorithm;
receiving, from the user, a first and a second feedback each corresponding to the first query response;
associating each of the first and the second feedback with the first and the second component of the search algorithm;
generating, via the at least one processor in response to the first and the second feedback, an override action corresponding to at least one of the first component and the second component and with one or more terms of the first search query, the override action indicating that a second query response generated by the corresponding at least one of the first component and the second component in response to a second search query including the one or more terms is to be overridden;
storing, via the at least one processor, the override action in a data store comprising an override cache;
training, by executing a machine learning algorithm using the first and the second feedback as input, a model associated with the first and the second component, wherein the training of the first component is based on the first feedback corresponding to the first sub response and the training of the second component is based on the second feedback corresponding to the second sub response;
modifying, using the model, the search algorithm such that the first and the second component are updated, resulting in a modified search algorithm;
receiving, at a second time which is distinct from the first time, the second search query comprising the one or more terms of the first search query, the second search query being equivalent to the first search query; and
generating, by executing the modified search algorithm and in accordance with the override action, the second query response in response to the second search query, the second query response being distinct from the first query response.
8 . A non-transitory computer readable storage medium having instructions stored that, when executed by a processing device, cause the processing device to perform operations comprising:
receiving, at a first time, a first search query from a user;
generating, by executing a search algorithm, a first query response comprising a first sub response and a second sub response in response to the first search query, wherein the first sub response is generated by a first component of the search algorithm and the second sub response is generated by a second component of the search algorithm;
receiving, from the user, a first and a second feedback each corresponding to the first and the second sub response;
associating each of the first and the second feedback with the first component and the second component of the search algorithm;
generating, via the at least one processor in response to the first and the second feedback, an override action corresponding to at least one of the first component and the second component and with one or more terms of the first search query, the override action indicating that a second query response generated by the corresponding at least one of the first component and the second component in response to a second search query including the one or more terms is to be overridden;
storing, via the at least one processor, the override action in a data store comprising an override cache;
training, by executing a machine learning algorithm using the first and the second feedback as input, a model associated with the and the second component, wherein the training of the first component is based on the first feedback corresponding to the first sub response and the training of the second component is based on the second feedback corresponding to the second sub response;
modifying, using the model, the search algorithm such that the first and the second component are updated, resulting in a modified search algorithm;
receiving, at a second time which is distinct from the first time, the second search query comprising the one or more terms of the first search query, the second search query being equivalent to the first search query; and
generating, by executing the modified search algorithm and in accordance with the override action, the second query response in response to the second search query, the second query response being distinct from the first query response.
9 . The non-transitory computer readable storage medium of claim 8 wherein the first feedback comprises one of an accept label, a rejection label, an override label, or a revise label.
10 . The non-transitory computer readable storage medium of claim 8 , wherein the first component of the search algorithm comprises a portion of the search algorithm configured to perform a filter function of the search algorithm.
11 . The non-transitory computer readable storage medium of claim 8 , wherein the first component comprises a natural language processing (NLP) filter component configured to perform NLP filtering.
12 . The non-transitory computer readable storage medium of claim 8 , wherein the first feedback comprises a revise label.