A New Way To Classify: Watson Natural Language Classifier & Watson Studio
Watson Natural Language Classifier
Independent of industry vertical or use case, natural language processing is becoming central to all user workflows. From developer to data scientist, the need for a simple workflow is essential.
At the root of all natural language processing is text classification.
Watson Natural Language Classifier is the enterprise solution for training, customizing, and deploying natural language classifiers at scale.
Today we are happy to extend Watson Natural Language Classifier to Watson Studio. While maintaining existing API functionality, you will now be able to leverage Watson Natural Language Classifier with a new easy to use interface in Watson Studio.
A Frictionless User Journey
Keeping in mind the end user while redesigning the tooling was important. Common pain points in the old toolkit were gathered and addressed to create an improved experience for Watson Studio.
Watson Studio caters to your full user experience. In one environment, you will be able to train, build, and test your classifiers. Additionally, you can import other services, notebooks, frameworks, and models into your projects.
Note: Old toolkit users will be auto migrated; however, you must download your training data and switch to the new tooling by July 31st. Read More
Existing Service Capabilities & Users
- Classify Natural Language Text: Train classifiers using IBM Deep Learning as a Service
- Multilingual Support: English, Arabic, French, German, Italian, Japanese, Korean, Portuguese (Brazilian), and Spanish
- Multi-category classification: List results for many classes for given text input
- Multi-phrase classification: Classify up to 30 separate text inputs in a single API request
All users will need to create or login to an existing instance of Watson Natural Language Classifier. Create an instance
Upon reaching the dashboard you will be able to access the API or click the launch tool button. Clicking the Launch Tool button will take you to the Watson Studio tooling allows users to use the service in a non-programmatic way.
- Read how to build your own Natural Language Classifier on Watson Studio
What People Are Making?
From financial risk to service solutions, enterprises around the globe are using Watson Natural Language Classifier to classify data.
Below are some sample apps that you can build in less than a day. Each application comes with a demo, GitHub repo, video tutorial, and business use case.
- Phishing E-Mail Detection: Demo
- E-Commerce Product Categorization: Demo
- Job Description Categorization: Fork the code
- Health Injury ICD-10 Classification: Video Tutorial
Zia Mohammad is a product manager on the Watson Language Translator and Natural Language Classifier team. His passion centers around: AI, emerging technologies, space, and neuroengineering. Feel free to reach out or comment below!