You are here

Home » Using Our API

Introduction

The TextWise API builds on the power of our Semantic Gist® technology. First introduced to the public in early 2008, our legacy Trainable Semantic Vectors (TSV) technology allowed applications to utilize the meaning of text to power highly-relevant Similarity Searches between query content and indexed content. In 2012, TextWise introduced a significantly improved semantic technology, Semantic Gist®. This technology, based on language modeling and neural networks, has been transparently integrated into the TextWise API.

Semantic Gist® also supports additional semantic services such as Concept Tagging and Categorization, using our simple yet powerful API.

The Content Matching service offers:

  • Support for licensing custom indexes to perform Similarity Search operations against your custom content.
  • Support for facets in the index and match services. Facets are simply attributes of an item that exist within the index and provide a mechanism to restrict the search space of the index based on specific values.
     

As documents are added to a semantic index, the language and concepts used in the corpus may evolve and diverge from the original index. This semantic drift can eventually affect the quality of query results. To compensate for this drift, the Semantic Gist® matching system takes advantage of its unsupervised learning capability and self-evolves its sematic model to compensate for this drift. This assures high quality sematic search over a constantly changing set of documents.

To learn more about our new Semantic Gist® Technology, go to the technology section of this website.

Getting started with the API is as simple as registering with TextWise.com. Once your registration is confirmed, you will receive an API token. Using this token, you can start making calls to any of the services defined below and begin integrating Semantic Gist® technology into your applications immediately.

Available Services

  • Concept Service -- Extracts the key concepts from the referenced content, ordered by weight.
  • Category Service -- Identifies the ODP categories, for the referenced content, ordered by weight.
  • Match Service -- Finds the similarity between the content you provide and the content contained within one of our indexes - currently News and Blogs, Wikipedia articles, Amazon products, Photos and YouTube® videos, or one of your custom indexes. The service then returns the most similar matches.
  • Index Service -- Provides you with the ability to store your own content within a licensed custom index and use the Match Service call for performing Similarity Search on that content.
  • Filter Service -- Removes specific formatting (e.g. HTML) from the provided content.