Developer API

Unprecedented Value from Text in your Solution

Embed advanced search, analytical and discovery capabilities to extract information previously undiscovered in text and turn it into valuable insights for many types of business users.

Download: Factsheet – KMX for OEM

Less Seeking, More Finding

Publishers, software vendors and enterprises are facing increased demand to deploy advanced search and visualization technologies to support their users to do less seeking and more finding in unstructured data sources like email, application notes, blogs, social media, content management systems, research and patent libraries.

Co-developed and used by the world’s most demanding enterprises in technology, chemical, life sciences and government KMX is a leading platform for delivering fast and powerful analytics, clear visual insights and advanced management support for information professionals.

KMX for OEM core model and supported functions (summary)

KMX API core model and potential functions (summary)

Embed to Empower your Users

CTOs and product managers can focus on developing their content-based application or subscription-based service by differentiating through the power of KMX resulting in a lower implementation cost and faster time-to-market.

At the core of the KMX Platform are best-of-breed big data text analytics functions like clustering, auto-classification and visualization. This enables OEM partners to quickly and seamlessly embed advanced search and discovery capabilities to enhance value and market acceptance of your own product and service offerings.

KMX is using state-of-art proprietary machine learning algorithms to extract and recall information previously hidden in text, turning it into valuable insights that can be used by all types of business users. Learn-by-example works without a lot of pre-work or set-up allowing KMX to classify every element in every variable.

The KMX  REST API provide deep insights and understanding of text, which not only enables users to discover previously unknown patterns and relationships in your data but also supports workflow based analytical processes like alerting and data enrichment.

API for Advanced Text Analytics & Visualization

KMX for OEM provide partners with the option to select features they need to augment their own product line or service offering by adding a specific set of functionality-including search, clustering and auto-classification. Such flexibility reduces partners’ overall costs and empowers them to drive the future of their own solution.

Treparel KMX provides an open platform for enabling organizations to leverage the power of KMX within existing technology environments. KMX enables software developers to access and extend KMX Text Analytics technology using standards for data access and integration. The KMX REST API offers access through web services and XML formats to enable rapid integration that is scalable, secure and flexible.

Text Acquisition & Pre-processing

KMX delivers fast and efficient access to you data and advanced search on that data through native connections or Excel and XML based importers. KMX facilitates developing connectors to directly retrieve and index third-party data repositories. KMX automatically pre-processes the raw unstructured data and applies stemming, stop word removal and tokenization before it generates high dimensional sparse vectors.


Clustering functionality categorizes unstructured information into thematic groups, identifying the major topics in documents. It can recommend structure for further classification.

Clustering provides users unsupervised analytics and automatically identifies inherent themes or information clusters. KMX clustering creates order out of chaos and delivers instant, high-level visibility into the data. Through a dynamic hierarchical topic view into search results it enables users to quickly focus on annotated subjects rather than scrolling through long results lists.


KMX provides users to do supervised analytics to help them automatically categorize large sets of documents. The Classification process can use a small number of documents sets for learn-by-example categorization. By sorting the content of documents by topic, relevancy and keywords users can apply their own models or rules for classification. It reliably detects fully or partially identical texts, and can purge or filter stored data.

Building classifiers is easy and supported by cross validation to calculate the precision and recall of the analyzed results.

By using the Support Vector Machine algorithm KMX classification removes time-consuming reliance on manual intervention required by traditional rule-based methods where you first need to build and apply taxonomies. It overcomes issues caused by the exponential growth of unstructured data through its ability to automatically, consistently and accurately classification of data.

Note: Classification also represents the terms:

  • Predictive Coding or Suggestive Coding
  • Technology-assisted review (TAR) or Automated Review
  • Auto-Categorization
  • Automated Relevance Designation
  • Relevance Ranking


Advanced visual knowledge discovery for displaying, exporting and sharing data results, ranked document lists, labeled and enriched data or interactive visualizations. Terms can be extracted to use in building thesauri or taxonomies.

KMX for OEM offers different functions to cross validate and improve the classification performance. Tagging, annotating and ranking the results support users and developers to improve relevance. KMX supports collaboration between users by sharing result sets, visualizations and classification models. Server based alerting signals new or changed relevant documents.

How Do I Leverage KMX In My Organization?

Organizations leverage KMX within existing application environments like existing content-based services, analytics and business intelligence workflows or can become part of existing unstructured data analysis pipelines.

Contact us for a call or personal meeting to discuss in more detail how KMX can be applied in your application or content-based service offering.

Comments are closed.