Use cases and sample notebooks

Following are real-world use cases and sample code to help you immediately begin to derive value from the Watson Natural Language Processing Library for Embed.

Note: To use these notebooks you need to create a Watson Studio instance, ensure you have a NLP compatible Watson Studio runtime, and import the notebook into Watson Studio.

Analyze movie reviews using Watson NLP

Sentiment analysis classifies the sentiment of movie reviews into positive, negative or neutral sentiment. You will use the Sentiment model to analyze the sentiment for a complete review text, as well as for individual sentences.

See the sample Jupyter notebook

Classify emotions in tweets using Watson NLP

Emotion analysis classifies the emotions of tweets into sadness, joy, anger, fear, or disgust. The dataset contains over seven thousand movie quotations from tweets. Because some quotations have multiple sentences, they can be regarded as documents.

See the sample Jupyter notebook

Extract entities in hotel reviews using Watson NLP

Perform entity extration and keyword phrase extraction from the text of customer reviews. The dataset contains 515,000 customer reviews and scoring of 1493 luxury hotels across Europe. A review contains the customer’s narrative description of their experience. This could be used, for example to route a complaint to the appropriate staff member, or to detect common issues within a facility.

See the sample Jupyter notebook