Amazon Comprehend

Amazon Comprehend is a natural-language processing (NLP) service from Amazon Web Services (AWS) that uses machine learning to uncover valuable insights and connections in structured data and text.

Most businesses have a wealth of untapped information inside many documents spread across different departments of the company. For example, the valuable insights found in customer support requests, emails, customer surveys, call transcripts, financial forms, and more can significantly be leveraged to improve customers’ experiences.

For instance, you could extract insights from customer surveys to improve your products. But extracting insights from these documents can be expensive, hard to scale, and may take hours.

That’s where Amazon Comprehend comes in. The tool is designed to simplify this business process by using machine learning to help them uncover and better understand unseen information.

This allows you to get more out of your business documents, communication materials, and stored data to serve your customers better and drive success. Business need not miss out on critical information because they can’t uncover them.

Amazon Comprehend is trained to identify essential elements in data using a dozen natural language processing features such as Entity Recognition, Custom Entity Recognition, Key phrase Extraction, Sentiment Analysis, Custom Classification, and more APIs that enable it to uncover and record valuable information, including language, people, places, etc., from everywhere.

The best part is that anyone can get up and running quickly with no machine learning experience required. Simply integrate natural language processing into your applications and provide the location of the source document or text. The APIs will output entities, key phrases, sentiment, and language in a JSON format, which you can use in your applications.

Amazon Comprehend can also automatically organize these text files in a category by relevant terms or topics. This can help you accurately detect customer sentiment in your content and then make real-time mediation decisions based on these findings.

You can even train it to label documents with topics or tags to find using rule-based tagging and keyword search. And even beyond that, the tool can accurately classify documents. If your content contains personally identifiable information, Amazon Comprehend can locate it. You can quickly redact or mask it with the PII Identification and Redaction to keep your privacy.

Key features

• Classify and extract entities from your documents
• Use Custom Classification to categorize inbound requests automatically
• Analyze the overall sentiment of a text (Positive, Negative, Neutral, or Mixed)
• Detect and redact personally identifiable information
• Identify text written in over 100 languages