18486441. CONVERSATIONAL DOCUMENT QUESTION ANSWERING simplified abstract (Oracle International Corporation)

CONVERSATIONAL DOCUMENT QUESTION ANSWERING

Organization Name

Oracle International Corporation

Inventor(s)

Xu Zhong of Melbourne (AU)

Thanh Long Duong of Seabrook (AU)

Mark Edward Johnson of Sydney (AU)

Charles Woodrow Dickstein of New York NY (US)

King-Hwa Lee of Bellevue WA (US)

Xin Xu of San Jose CA (US)

Srinivasa Phani Kumar Gadde of Fremont CA (US)

Vishal Vishnoi of Redwood City CA (US)

Christopher Kennewick of Kirkland WA (US)

Balakota Srinivas Vinnakota of Sunnyvale CA (US)

Raefer Christopher Gabriel of San Jose CA (US)

CONVERSATIONAL DOCUMENT QUESTION ANSWERING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18486441 titled 'CONVERSATIONAL DOCUMENT QUESTION ANSWERING

Simplified Explanation

The patent application describes techniques for integrating document question answering in an artificial intelligence-based platform, such as a chatbot system. The process involves receiving a user query, rewriting the query with specific descriptors, computing an embedding vector for the rewritten query, retrieving textual passages from a document store using the embedding vector, determining answers within the passages, and returning the answers to the user.

• Receiving user query and rewriting with specific descriptors
• Computing embedding vector for rewritten query
• Retrieving textual passages from document store using embedding vector
• Determining answers within passages
• Returning answers to user

Potential Applications

This technology could be applied in customer service chatbots, virtual assistants, and information retrieval systems.

Problems Solved

This technology streamlines the process of retrieving specific information from documents, making it easier for users to find answers to their queries.

Benefits

The benefits of this technology include improved user experience, faster response times, and more accurate answers to user queries.

Potential Commercial Applications

Potential commercial applications of this technology include customer support chatbots, knowledge management systems, and search engines.

Possible Prior Art

One possible prior art for this technology is the use of natural language processing techniques in question answering systems. Another prior art could be the use of embeddings for information retrieval in search engines.

How does this technology handle multi-turn conversations in chatbot systems?

This article does not address how the system manages multi-turn conversations where the user may ask follow-up questions or provide additional context.

What types of documents are supported for question answering in this system?

The article does not specify the types of documents that can be used for question answering, such as text documents, PDFs, or web pages.

Original Abstract Submitted

Techniques are disclosed herein for integrating document question answering in an artificial intelligence-based platform, such as a chatbot system. The techniques include receiving a query from a user, rewriting the query to include one or more specific descriptors, computing an embedding vector for the rewritten query, retrieving one or more textual passages from a document store utilizing the embedding vector for the rewritten query, determining one or more answers to the rewritten query within the one or more textual passages, and returning the one or more answers.