17643204. ENHANCING INPUT TO CONVERSATIONAL AGENTS THROUGH FEEDBACK simplified abstract (INTERNATIONAL BUSINESS MACHINES CORPORATION)

ENHANCING INPUT TO CONVERSATIONAL AGENTS THROUGH FEEDBACK

Organization Name

INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor(s)

Jason Andrew Telner of Lackawanna NY (US)

Jon Gerard Temple of Southbury CT (US)

Tyler Waite of Lake Washington WA (US)

Donna Thelander of Kailua Kona HI (US)

Mark Willard Malatesta of Ortonville MI (US)

William G. Dusch of Morrisville NC (US)

ENHANCING INPUT TO CONVERSATIONAL AGENTS THROUGH FEEDBACK - A simplified explanation of the abstract

This abstract first appeared for US patent application 17643204 titled 'ENHANCING INPUT TO CONVERSATIONAL AGENTS THROUGH FEEDBACK

Simplified Explanation

The patent application describes a method, system, and program for improving user input to a conversational agent. Here are the key points:

• The method involves using a natural language processing algorithm to identify a word in a user's query to the conversational agent.
• The method then determines a set of keywords associated with different intents of the conversational agent from a database.
• A semantic similarity score is calculated between each set of keywords and the identified word.
• If the identified word is not found in any set of keywords and the semantic similarity score is above a certain threshold, the identified word is displayed to the user on a screen with a distinct indication.

Potential applications of this technology:

• Enhancing the accuracy and effectiveness of conversational agents or chatbots.
• Improving user experience by providing relevant suggestions or alternatives when the user's input is not recognized.
• Assisting users in finding the right information or navigating through complex systems.

Problems solved by this technology:

• Inaccurate or incomplete understanding of user input by conversational agents.
• Difficulty in handling queries or requests that do not match predefined keywords or intents.
• Lack of feedback or guidance for users when their input is not recognized.

Benefits of this technology:

• Improved user satisfaction and engagement with conversational agents.
• Increased efficiency and effectiveness of conversational agents in understanding and responding to user queries.
• Enhanced user experience through personalized and relevant suggestions or alternatives.

Original Abstract Submitted

A computer-implemented method, a computer system and a computer program product enhance user input to a conversational agent. The method includes identifying a word using a natural language processing algorithm. The word is in a query from a user to the conversational agent. The method also includes determining a set of keywords related to each of a plurality of intents associated with the conversational agent within a database. In addition, the method includes calculating a semantic similarity score between each determined set of keywords and the identified word. Lastly, the method includes displaying the identified word on a screen to the user when the identified word is not found in each determined set of keywords and when the semantic similarity score is above a threshold. The display includes a distinct indication of the identified word.