Labeling text with categories
Efficiently analyze large volumes of text and assign each sentence to a predefined category by using the text categorization feature. With text categorization, you can quickly react to what your customers are saying and aptly address their inquiries or concerns.
The text categorization feature assigns the analyzed text to one or more predefined categories. Pega Platform provides you with three types of text categorization, depending on what you want to detect:
- Topic detection
- Topic detection determines the underlying topic of a single piece of text or an entire document to efficiently route an incoming customer query to the right agent.
For example, in a chat window on the Emu Airlines website, a customer writes I want to book a ticket from London to Tokyo. Topic detection automatically assigns the message to the Booking a flight category based on similar customer messages and the book and ticket keywords. The company application routes all messages from the Booking a flight category to the corresponding customer service team. As a result, Emu Airlines reduces the response time to customer queries and improves the quality of their customer service.
Topic detection supports both machine learning and keyword-based categorization.
For more information, see Detecting the topics of text fragments.
- Intent detection
- Intent detection analyzes a unit of text, for example, a comment on your company social media profile, to determine the intent of the author. For example, intent detection classifies the sentence How do I create an account at uPlusTelco? as an inquiry.
Intent detection supports machine learning categorization.
For more information, see Recognizing user intent.
- Sentiment detection
- Sentiment detection recognizes the feelings (attitudes, emotions, opinions) that characterize a unit of text and then assigns the analyzed text to one of the following sentiment categories: positive, neutral, or negative. For example, the sentence I am happy with uPlusTelco's customer service. is determined as positive.
Sentiment detection supports machine-learning categorization.
- Detecting the topics of text fragments
Efficiently categorize and rout customer inquiries to the corresponding customer service consultant with topic detection. Topic detection scans a piece of text and determines the underlying topic, and then automatically assigns the text to a predefined category.
- Recognizing user intent
Create intent analysis models to enable your application to detect the ideas that users express through written communication. For example, you can use an intent model when you want your chatbot to understand and respond when someone asks for help.
- Determining the emotional tone of text
Sentiment analysis determines whether the opinion that the writer expressed in a piece of text is positive, neutral, or negative. Knowledge about customers' sentiments can be very important because customers often share their opinions, reactions, and attitudes toward products and services in social media or communicate directly through chat channels.
- Analyzing natural language
Effortlessly analyze and extract meaningful information from large volumes of text with the use of text analytics. Based on your findings, you can further improve business performance and customer experience.
- Text analytics accuracy measures
Models predict an outcome, which might or might not match the actual outcome. The following measures are used to examine the performance of text analytics models. When you create a sentiment or classification model, you can analyze the results by using the performance measures that are described below.
- Testing text analytics models
You can perform ad-hoc testing of text analytics models that you created and analyze their performance in real-time, on real-life data.