5 Things to Know About Watson Natural Language Classifier
MarcelaAdanRochester 2700048UPS Visits (17375)
Quite often, in university classes, conferences, and cognitive computing hackathons I’ve been asked by students, IBM clients and developers "What is Natural Language Classifier?”, "What are the typical use cases of Natural Language Classifier ?”, “How is Natural Language Classifier related to cognitive computing or AI?” and “How do I get started using the Watson Natural Language Classifier service?”.
Here are five questions I often get from students and developers about the Watson Natural Language Classifier service:
1. What is Natural Language Classifier?
The Natural Language Classifier service applies cognitive computing techniques to return best matching predefined classes for short text inputs, such as a sentence or phrase. It has the ability to classify phrases that are expressed in natural language into categories. Natural Language Classifier is based on Natural Language Processing (NLP) technology. NLP is a field of computer science, artificial intelligence (AI) and computational linguistics concerned with the interactions between computers and human (natural) languages as depicted by Figure 1. The Natural Language Understanding service provides the ability to classify phrases that are expressed in natural language into categories and returns a complete corresponding classification with a confidence score.
2. What are examples of what Natural Language Classifier can do?
Consider the following questions: “When can you meet me?” or “When are you free?” or “Can you meet me at 2:00 PM?” or “Are you busy this afternoon?”. NLC can determine that they are all ways of asking about “setting up an appointment”. Short phrases can be found in online discussion forums, emails, Twitter feeds, SMS messages, and electronic forms. Using Watson APIs you can send text from these sources to a trained natural language classifier. The classifier will return its prediction of class that best captures what is being expressed in that text. Based on the predicted class you can trigger your application to take the appropriate action such as provide an answer to a question, suggest a relevant product based on expressed interest or forward the text to an appropriate human expert who can help.
3. What are typical Natural Language Classifier use cases?
When the amount of information that you want to categorize is very large, consider using the Watson Natural Language Classifier. You can use Natural Language Classifier in several practical applications, for example:
4. How do I use Natural Language Classifier?
Figure 2 provides an overview of the four steps that are included in the process of creating and using the classifier.
5. How can I get started with Natural Language Classifier?
To get started with Natural Language Classifier sign up for a tria
Marcelo Mota Manhaes is a Certified IT Specialist in IBM Global Technology Services®, IBM Brazil. Marcelo is an IT Delivery Architect; his areas of expertise include cloud computing, software automation tools, business analytics, and cognitive computing. Marcelo has over 20 years of experience in the IT industry. He led several projects to design and build cognitive solutions such as an application that helps managers to evaluate the performance of their employees and a question-answering system that uses Watson Natural Language Classifier (NLC), Retrieve and Rank, and Language Translator to enable IBM Knowledge Center users to find technical information by asking questions in their native language. Marcelo teaches cloud computing and cognitive systems at the Universidade Positivo. He is the author of several IBM Redbooks publications. Marcelo holds a B.S. in Computer Science from Universidade Federal do Paraná—UFPR and an M.S. in Computer Science from Universidade Tecnológica Federal do Paraná—UTFPR.