Use a few lines of code to easily analyze sentiment, extract key phrases, detect topics, and detect language for any kind of text.
Find out what users think of your brand or topic by analyzing any text using sentiment analysis. You are now easily able to monitor the perception of your brand or topic over time.
Sentiment score is generated using classification techniques, and returns a score between 0 and 1. The input features to the classifier include n-grams, features generated from part-of-speech tags, and embedded words.
Key Phrase Extraction
Automatically extract key phrases to quickly identify the main points. We employ techniques from Microsoft Office's sophisticated Natural Language Processing toolkit.
For example, for the input text ‘The food was delicious and there were wonderful staff’, the service returns the main talking points: ‘food’ and ‘wonderful staff’.
Topic Detection (To be deprecated August 24, 2017)
This service returns topics that have been detected in multiple text articles. The service is designed to work well for short, human written text such as reviews and user feedback. It can help you to understand the main issues or suggestions that customers are mentioning.
The service can be used to detect which language the input text is written in. 120 languages are supported.
|Language||Language code||Sentiment||Key phrases||Topics|
* indicates language support in preview
Scoring large number of records
You can send several records in a single call, but keep in mind the following limits: Max request size: 1MB (for sentiment, KP and language) Max number of documents per request: 1000 Max size of a single document: 10KB
If you need more than 1000 records analyzed, break up your content into several calls. You can make up to 100 API calls per minute -- effectively allowing you to score 100,000 records per minute.