SYSTEMS AND METHODS FOR INTENT RESPONSE SOLICITATION AND PROCESSING
Disclosed embodiments provide a framework to solicit and evaluate responses from different systems and other users to the intents communicated by requesting users. In response to obtaining an intent, an intent messaging service provides the intent to a selected system to solicit a response to the intent. In response to obtaining an intent response, the intent messaging service prohibits the system from generating further intent responses and provides the obtained intent response to the requesting user. The intent messaging service establishes a communications session between the requesting user and the system in response to another request corresponding to the intent. This allows for additional responses to be provided to the requesting user by the system identified by the intent messaging service.
1 . A computer-implemented method comprising:
obtaining an intent, wherein the intent is associated with a requesting user;
generating in real-time a dataset, wherein the dataset includes sample intents, sample outputs, and feedback corresponding to the sample outputs, wherein the sample outputs correspond to features associated with a set of other users that are provided with the sample intents, and wherein the dataset is generated in real-time as the feedback is obtained;
dynamically training in real-time a machine learning algorithm to identify different users for soliciting responses to different intents, wherein the machine learning algorithm is dynamically trained using the dataset as the feedback is obtained;
processing the intent through the machine learning algorithm to identify one or more users for soliciting responses to the intent;
obtaining in real-time an intent response, wherein the intent response is associated with a particular user from the one or more users;
providing the intent response, wherein the intent response is provided as a result of a determination that the intent response is relevant to the intent;
facilitating a communications session between the requesting user and the particular user, wherein the communications session is facilitated in response to input corresponding to the intent response;
dynamically updating the dataset in real-time using the intent and features corresponding to particular user; and
updating in real-time the machine learning algorithm using the dynamically updated dataset.
2 . The computer-implemented method of claim 1 , wherein the intent is provided without identifying information associated with the requesting user, and wherein when the communications session is facilitated, the identifying information associated with the requesting user is provided.
3 . The computer-implemented method of claim 1 , further comprising:
performing a semantic analysis of a request associated with the requesting user to extract the intent from the request.
4 . The computer-implemented method of claim 1 , further comprising:
identifying additional information for supplementing the intent; and
submitting one or more prompts for the additional information, wherein when the one or more prompts are presented through a computing device associated with the requesting user, the additional information is provided.
5 . The computer-implemented method of claim 1 , further comprising:
transmitting a set of executable instructions, wherein when the set of executable instructions are received by an application associated with the particular user, the application automatically prohibits the particular user from submitting additional intent responses.
6 . The computer-implemented method of claim 1 , further comprising:
obtaining in real-time a different intent response, wherein the different intent response is associated with a different user from the one or more users;
determining that the different intent response is irrelevant to the intent; and
dynamically updating the dataset in real-time such that a likelihood of the different user being identified by machine learning algorithm for similar intents is reduced.
7 . The computer-implemented method of claim 1 , further comprising:
dynamically training in real-time a classification algorithm to classify intent responses as being either relevant or irrelevant to corresponding intents, wherein the classification algorithm is dynamically trained using a dataset of input intents, known responses, and classifications corresponding to the known responses; and
processing the intent and the intent response through the classification algorithm, wherein when the intent and the intent response are processed, the classification algorithm classifies the intent response as being relevant to the intent.
8 . A system, comprising:
one or more processors; and
memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to:
obtain an intent, wherein the intent is associated with a requesting user;
generate in real-time a dataset, wherein the dataset includes sample intents, sample outputs, and feedback corresponding to the sample outputs, wherein the sample outputs correspond to features associated with a set of other users that are provided with the sample intents, and wherein the dataset is generated in real-time as the feedback is obtained;
dynamically train in real-time a machine learning algorithm to identify different users for soliciting responses to different intents, wherein the machine learning algorithm is dynamically trained using the dataset as the feedback is obtained;
process the intent through the machine learning algorithm to identify one or more users for soliciting responses to the intent;
obtain in real-time an intent response, wherein the intent response is associated with a particular user from the one or more users;
provide the intent response, wherein the intent response is provided as a result of a determination that the intent response is relevant to the intent;
facilitate a communications session between the requesting user and the particular user, wherein the communications session is facilitated in response to input corresponding to the intent response;
dynamically update the dataset in real-time using the intent and features corresponding to particular user; and
update in real-time the machine learning algorithm using the dynamically updated dataset.
9 . The system of claim 8 , wherein the intent is provided without identifying information associated with the requesting user, and wherein when the communications session is facilitated, the identifying information associated with the requesting user is provided.
10 . The system of claim 8 , wherein the instructions further cause the system to:
perform a semantic analysis of a request associated with the requesting user to extract the intent from the request.
11 . The system of claim 8 , wherein the instructions further cause the system to:
identify additional information for supplementing the intent; and
submit one or more prompts for the additional information, wherein when the one or more prompts are presented through a computing device associated with the requesting user, the additional information is provided.
12 . The system of claim 8 , wherein the instructions further cause the system to:
transmit a set of executable instructions, wherein when the set of executable instructions are received by an application associated with the particular user, the application automatically prohibits the particular user from submitting additional intent responses.
13 . The system of claim 8 , wherein the instructions further cause the system to:
obtain in real-time a different intent response, wherein the different intent response is associated with a different user from the one or more users;
determine that the different intent response is irrelevant to the intent; and
dynamically update the dataset in real-time such that a likelihood of the different user being identified by machine learning algorithm for similar intents is reduced.
14 . The system of claim 8 , wherein the instructions further cause the system to:
dynamically train in real-time a classification algorithm to classify intent responses as being either relevant or irrelevant to corresponding intents, wherein the classification algorithm is dynamically trained using a dataset of input intents, known responses, and classifications corresponding to the known responses; and
process the intent and the intent response through the classification algorithm, wherein when the intent and the intent response are processed, the classification algorithm classifies the intent response as being relevant to the intent.
15 . A non-transitory, computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to:
obtain an intent, wherein the intent is associated with a requesting user;
generate in real-time a dataset, wherein the dataset includes sample intents, sample outputs, and feedback corresponding to the sample outputs, wherein the sample outputs correspond to features associated with a set of other users that are provided with the sample intents, and wherein the dataset is generated in real-time as the feedback is obtained;
dynamically train in real-time a machine learning algorithm to identify different users for soliciting responses to different intents, wherein the machine learning algorithm is dynamically trained using the dataset as the feedback is obtained;
process the intent through the machine learning algorithm to identify one or more users for soliciting responses to the intent;
obtain in real-time an intent response, wherein the intent response is associated with a particular user from the one or more users;
provide the intent response, wherein the intent response is provided as a result of a determination that the intent response is relevant to the intent;
facilitate a communications session between the requesting user and the particular user, wherein the communications session is facilitated in response to input corresponding to the intent response;
dynamically update the dataset in real-time using the intent and features corresponding to particular user; and
update in real-time the machine learning algorithm using the dynamically updated dataset.
16 . The non-transitory, computer-readable storage medium of claim 15 , wherein the intent is provided without identifying information associated with the requesting user, and wherein when the communications session is facilitated, the identifying information associated with the requesting user is provided.
17 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
perform a semantic analysis of a request associated with the requesting user to extract the intent from the request.
18 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
identify additional information for supplementing the intent; and
submit one or more prompts for the additional information, wherein when the one or more prompts are presented through a computing device associated with the requesting user, the additional information is provided.
19 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
transmit a particular set of executable instructions, wherein when the particular set of executable instructions are received by an application associated with the particular user, the application automatically prohibits the particular user from submitting additional intent responses.
20 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
obtain in real-time a different intent response, wherein the different intent response is associated with a different user from the one or more users;
determine that the different intent response is irrelevant to the intent; and
dynamically update the dataset in real-time such that a likelihood of the different user being identified by machine learning algorithm for similar intents is reduced.
21 . The non-transitory, computer-readable storage medium of claim 15 , wherein the executable instructions further cause the computer system to:
dynamically train in real-time a classification algorithm to classify intent responses as being either relevant or irrelevant to corresponding intents, wherein the classification algorithm is dynamically trained using a dataset of input intents, known responses, and classifications corresponding to the known responses; and
process the intent and the intent response through the classification algorithm, wherein when the intent and the intent response are processed, the classification algorithm classifies the intent response as being relevant to the intent.