AutoSummary Entity Extraction
Feature Release
This is the announcement for an upcoming ASAPP feature. Your ASAPP account team will provide a target release date and can direct you to more detailed information as needed.
Overview
AutoSummary is being extended with a new feature, Entity Extraction.
The entity extraction feature identifies and extracts key entities---such as Product names, dates, $ amounts, competitor names---from the conversations and text. The feature is designed to enhance data retrieval and Targeted, custom analytics for downstream use and to optimize operations.
This feature is fully customizable. Customers can request the extraction of specific keywords tailored to their domain and line of business, enabling the extraction of a wide range of insights from conversations.
Use and Impact
Users need the flexibility to target and extract analytics for downstream use and to optimize operations, and these data points differ across companies and over time. Entity extraction can be useful in several key use cases:
- Quick Information Retrieval: Agents can quickly extract critical information such as order numbers, $ amount, competitor name and product details from conversations, improving response times and customer satisfaction.
- Insight Extraction: Agents can extract specific keywords and entities from large volumes of text, facilitating detailed reporting, trends and patterns that inform decision-making.
The Entity Extraction feature aims to deliver significant benefits to the enterprises:
- Faster Information Retrieval (after-call wrap time): By automatically extracting key entities from conversations, users can quickly locate relevant information without manually sifting through chat/conversation.
- Informed Actions: With quick access to important elements of a conversation, users can make more informed decisions, leading to better outcomes in customer support
How it Works
When agents finish a conversation, AutoSummary analyzes the conversations, and extracts key entities for insights which are relevant to the conversation. These key entities can be retrieved via the Entities API, or within the Conversation Explorer.
Work with your ASAPP team to configure the entities you want to extract. ASAPP also provides a number of common out of the box entities that you may use or modify from.
Some examples of the types of questions you could ask:
- Date mentioned by the customer as the start of the issue
- The type of product the call described (e.g. iPhone vs Android)
- $ amounts that are important to the issue
- Last-four digits of the account number (if not redacted in the transcript)
- A competitor name
This upgrade comes with no changes to the API specs; Work with your ASAPP team to configure the same both custom and default entities that you want to enable.
Below are the example to be provided to get the entities enabled
- Entity name: ‘Product’
- Description: The product that customer is talking about
- Example (Product Name): ‘Nike Dunk Low Retro’
FAQs
-
Can I customize the entities that are extracted?
Yes, you can customize the feature to extract specific keywords and entities relevant to your domain and business needs, ensuring the extracted data is tailored to your unique requirements.
-
How do you keep the entities up to date?
Entities are kept up to date through a combination of automated updates and user feedback on a regular interval.
Was this page helpful?