US-20260129009-A1 - INTENT PREDICTION FOR DIALOGUE GENERATION

US20260129009A1US 20260129009 A1US20260129009 A1US 20260129009A1US-20260129009-A1

Abstract

In certain embodiments, intent prediction and dialogue generation may be facilitated. In some embodiments, a chat initiation request may be obtained from a user. The latest activity information associated with the user may be provided to a prediction model to obtain a first set of predicted intents of the user. For each intent of the first set of predicted intents, a candidate question may be selected from a question set based on the candidate question matching the intent. In some embodiments, the candidate questions may be simultaneously presented on the chat interface.

Inventors

Victor Alvarez Miranda
Rui Zhang
Vinay Igure
Scott Karp
Erik Mueller
Tanushree Luke
Kunlaya Soiaporn

Assignees

CAPITAL ONE SERVICES, LLC

Dates

Publication Date: 20260507
Application Date: 20251230

Claims (20)

1 . A system comprising: one or more processors and non-transitory media storing instructions that, when executed by the one or more processors, cause operations comprising: starting a communication session; presenting, in a first dialogue in the communication session, based on user information associated with a user, a set of first dialogue items comprising a first plurality of selectable questions regarding one or more user accounts of the user; and in response to receiving input corresponding to selection of a first dialogue item of the set of first dialogue items, and while presenting the first plurality of selectable questions in a feed of the communication session as part of the first dialogue, presenting, in the feed of the communication session, a set of second dialogue items comprising a second plurality of selectable questions in a first person point of view of the user.
2 . The system of claim 1 , wherein presenting the set of second dialogue items comprises, in response to receiving the input corresponding to selection of the first dialogue item, and while the first plurality of selectable questions remains presented in the feed of the communication session as part of the first dialogue, presenting, in a second dialogue in the feed of the communication session, the second plurality of selectable questions in the user's first person point of view based on intents derived via a model from (i) the input corresponding to selection of the first dialogue item and (ii) the user information.
3 . The system of claim 1 , wherein presenting the set of first dialogue items comprises presenting, in the first dialogue in the feed of the communication session, one or more selectable questions inquiring about one or more financial accounts of the user.
4 . The system of claim 1 , wherein presenting the set of first dialogue items comprises, based on intents derived via a model from the user information, presenting, in the first dialogue in the feed of the communication session, one or more selectable questions in the user's first person point of view.
5 . A method comprising: in connection with a communication session, presenting, as at least part of a dialogue in the communication session, a first set of dialogue items comprising a first plurality of selectable options regarding one or more user accounts of a user; receiving an input corresponding to selection of a first dialogue item of the first set of dialogue items; and while presenting the first plurality of selectable options in a feed of the communication session, and based on the input corresponding to selection of the first dialogue item, presenting, in the feed of the communication session, a set of second dialogue items comprising a second plurality of selectable options in a first person point of view of the user.
6 . The method of claim 5 , wherein: presenting the set of second dialogue items comprises, while the first plurality of selectable options remains presented in the feed of the communication session as part of a first dialogue, presenting, in a second dialogue in the feed of the communication session, the second plurality of selectable options in the user's first person point of view based on one or more intents derived via a model from (i) the input corresponding to selection of the first dialogue item and (ii) user information associated with the user; the first dialogue is presented in the feed before a given dialogue of the user is presented in the feed; and the second dialogue is presented in the feed after the given dialogue of the user is presented in the feed.
7 . The method of claim 5 , wherein presenting the set of first dialogue items comprises presenting, in a first uninterrupted dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
8 . The method of claim 5 , wherein presenting the set of first dialogue items comprises presenting, as at least part of a welcome dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
9 . The method of claim 5 , wherein presenting the set of first dialogue items comprises presenting, as at least part of an uninterrupted introduction dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
10 . The method of claim 5 , wherein presenting the set of first dialogue items comprises presenting, in a first dialogue in the feed of the communication session, a plurality of first selectable questions regarding the one or more user accounts of the user.
11 . The method of claim 5 , wherein presenting the set of first dialogue items comprises presenting, in a first dialogue in the feed of the communication session, a plurality of first selectable statements regarding the one or more user accounts of the user.
12 . The method of claim 5 , wherein presenting the set of first dialogue items comprises, based on one or more intents derived via one or more neural networks from user information associated with the user, presenting, as at least part of the dialogue in the communication session, the set of first dialogue items comprising the first plurality of selectable options regarding the one or more user accounts of the user.
13 . One or more non-transitory computer-readable media comprising instructions that, when executed by one or more processors, cause operations comprising: presenting, as at least part of a dialogue in a communication session, a set of first dialogue items comprising a first plurality of selectable options related to one or more user accounts of a user; receiving input corresponding to selection of a first dialogue item of the set of first dialogue items; and while presenting the first plurality of selectable options in a feed of the communication session, and based on the input corresponding to selection of the first dialogue item, presenting, in the feed of the communication session, a set of second dialogue items comprising a second plurality of selectable options in a first person point of view of the user.
14 . The one or more non-transitory computer-readable media of claim 13 , wherein: presenting the set of second dialogue items comprises, while the first plurality of selectable options remains presented in the feed of the communication session as part of a first dialogue, presenting, in a second dialogue in the feed of the communication session, the second plurality of selectable options in the user's first person point of view based on one or more intents derived via a model from (i) the input corresponding to selection of the first dialogue item and (ii) user information associated with the user; the first dialogue is presented in the feed before a given dialogue of the user is presented in the feed; and the second dialogue is presented in the feed after the given dialogue of the user is presented in the feed.
15 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises presenting, in a first uninterrupted dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
16 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises presenting, as at least part of a welcome dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
17 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises presenting, as at least part of an uninterrupted introduction dialogue in the feed of the communication session, the first plurality of selectable options based on one or more intents derived via a model from user information associated with the user.
18 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises presenting, in a first dialogue in the feed of the communication session, a plurality of first selectable questions related to the one or more user accounts of the user.
19 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises presenting, in a first dialogue in the feed of the communication session, a plurality of first selectable statements related to the one or more user accounts of the user.
20 . The one or more non-transitory computer-readable media of claim 13 , wherein presenting the set of first dialogue items comprises, based on one or more intents derived via one or more neural networks from user information associated with the user, presenting, as at least part of the dialogue in the communication session, the set of first dialogue items comprising the first plurality of selectable options related to the one or more user accounts of the user.

Description

RELATED APPLICATION(S) This application is a continuation of U.S. patent application Ser. No. 18/652,785, filed May 1, 2024, which is a continuation of U.S. patent application Ser. No. 18/167,801, filed Feb. 10, 2023, which is a continuation of U.S. patent application Ser. No. 17/543,106, filed Dec. 6, 2021, which is a continuation of U.S. patent application Ser. No. 16/821,406, filed Mar. 17, 2020, which is a continuation of U.S. Ser. No. 16/821,008, filed Mar. 17, 2020, which claims the benefit of priority of U.S. Provisional Application No. 62/942,588, entitled “Pre-Chat Intent Prediction for Dialogue Generation,” filed Dec. 2, 2019, which is hereby incorporated by reference herein in its entirety. BACKGROUND OF THE INVENTION Chatbots (e.g., a form of automated conversional artificial intelligence) enable a human user to message or chat with a computer that “talks” like a human and, in some instances, get answers without necessitating human interaction or independent searches by the user. For example, a chatbot may obtain context from the questions submitted by the user or answers (or other comments) provided by the user during a chat session and then propose solutions or answers to the user. Often, however, the user does not know what questions to ask the chatbot or how to phrase questions to the chatbot, which causes friction with respect to the user's interactions with a chatbot service. SUMMARY OF THE INVENTION Aspects of the invention relate to methods, apparatuses, and/or systems for facilitating intent prediction, dialogue generation based on such intent prediction, or training or configuration of neural networks or other prediction models to facilitate dialogue generation. In some embodiments, a chat initiation request may be obtained from a user. The latest activity information associated with the user may be provided to a prediction model to obtain predicted intents of the user. For each intent of the current intents, a candidate question may be selected from a question set based on the candidate question matching the intent. In some embodiments, upon loading of the chat interface or within ten seconds of the chat initiation request, the candidate questions may be simultaneously presented on the chat interface. A user selection of a first question of the candidate questions may be obtained via the chat interface responsive to the presentation of the candidate questions. In this way, for example, the user is able to choose one or more of the presented questions to submit as the user's own question without needing to come up the question (or the phrasing of the question) or to understand how to phrase a question on the chat interface. In some embodiments, based on the user selection and the first question matching a first intent of the predicted intents, the first intent may be provided as reference feedback for the prediction model. As an example, the first intent may be used to update one or more configurations of the prediction model (e.g., weights, biases, or other parameters of the prediction model). In this way, for example, the prediction model may be trained or configured to generate more accurate predictions. Various other aspects, features, and advantages of the invention will be apparent through the detailed description of the invention and the drawings attached hereto. It is also to be understood that both the foregoing general description and the following detailed description are examples and not restrictive of the scope of the invention. As used in the specification and in the claims, the singular forms of “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. In addition, as used in the specification and the claims, the term “or” means “and/or” unless the context clearly dictates otherwise. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows a system for facilitating conversational artificial intelligence, in accordance with one or more embodiments. FIG. 2 shows a machine learning model configured to facilitate a conversation, in accordance with one or more embodiments. FIG. 3 shows chat interfaces with respective lists of selectable dialogue items, in accordance with one or more embodiments. FIG. 4 shows components of a personal assistance platform and their interactions with one another, in accordance with one or more embodiments. FIG. 5 shows a chart indicating various presentation orderings for dialogue items, in accordance with one or more embodiments. FIG. 6 shows a flowchart of a method of facilitating pre-chat intent prediction and dialogue generation based on such intent prediction, in accordance with one or more embodiments. FIG. 7 shows a flowchart of a method of providing additional dialogue items in response to negative feedback related to presented dialogue items, in accordance with one or more embodiments. FIG. 8 shows a flowchart of a method of generating category-based sets of predicted intents to facilitate dialogue generation, in