Semantic Analysis

Semantic analysis API helps users cluster similar articles by understanding the relatedness between different content and streamlines research by eliminating redundant text contents. Semantic analysis API can help bloggers, publishing and media houses to write more engaging stories by retrieving similar articles from the past quickly, and news aggregators to combine similar news from different sources to reduce clutter in the feeds of their readers.

Try the demo

enter a text

You might also be interested in these APIs: Sentiment Analysis, Intent Analysis, Keyword Extractor

Why our Semantic Analysis API ?


Provide normalized confidence score on a scale of 0 to 5 to support varying levels of similarity thresholds (very similar to completely dissimilar).


Returns similarity score between two documents instantly and suited to build applications like recommendation engines, chatbots or semantic search.


Komprehend semantic similarity algorithm can be trained in an unsupervised fashion allowing you to customize it on your dataset and improve the accuracy.

How Our Semantic Analysis API Works?

Semantic similarity API understands relatedness between different pieces of text. It helps in comparing the structure and meaning of the text which can be used to extract similar text and phrases from corpus.

Our API converts textual information to its corresponding document embeddings and the cosine similarity between the two embeddings is scaled to provide the result. The document embeddings are made using Recursive Auto Encoders. These encoders try to reconstruct the given sentences to determine their respective document embeddings.Semantic Similarity API provides a score on a range of 0-5 (0-Not similar, 5-Almost same)

use cases

Creating more compelling content through semantic publishing

Semantic analysis API can help bloggers, publishing and media houses in building recommendation engines. By analyzing the similarity of current articles with other articles in the archive, publishers can surface more contextual content to the users and improve key metrics like CTR, Time on Page, etc.

Chatbot Applications

Semantic API can be used to build Chatbots by analyzing the similarity of a user’s query with an existing database of question-answer pairs (FAQs). Such a technique of building chatbots can scale well to eventually support complex queries and can be an alternative to intent-based chatbot platform like Dialogflow.

Get Started

Custom Solutions

Want to train your own custom model? Contact Sales to get started

Contact Sales