Semantic and lexical technologies

Services

Driving data-driven applications

Areas of Practice


CORPUS DEVELOPMENT & DATA ENGINEERING

The foundation of linguistic computing is linguistic data. Digital text is abundant in the electronic age, but every application has distinctive needs. We can scale your data-collection pipeline, annotate and transform existing corpora to find the meaningful signal in the modern information firehose.


MACHINE LEARNING

A machine learning model is a combination of algorithms, data, and learned parameters. Each of these components must be chosen or engineered with discernment and understanding of the problem at hand. Given data and a set of questions, MDCCLV can collaborate with you to systematically answer any questions you might put to the data. Call it 'data science' if you want, but expect testable hypotheses and rigorous methodology.


CORPUS DEVELOPMENT & DATA ENGINEERING

Collecting data is one thing; finding the best data to meet an information need is another. Adept in both classical information retrieval theory and modern powerhouses (Elasticsearch/Lucene, Sphinx), MDCCLV can optimize your search workflows to provide answers quickly.


DIGITAL PUBLISHING & WEB DEVELOPMENT

With infinite time and infinite data, you can make an information product do anything. In the real world with finite resources, how do you prioritize to do make the most distinctive and engaging experience that is consistent with the goals of your organization? With an adeptly experimental approach, MDCCLV can build it with you.


INFORMATION EXTRACTION

Language is a treasure trove of information, encoding both hard data (information to be extracted) and embedded cultural clues (sentiment analysis and its ilk). Off-the-shelf packages can answer standard questions, but when your problems are unique, MDCCLV can tune standard algorithms to focus on specific needs.


MULTIMODAL DIGITAL STRATEGY

In a multimodal web of mobile, desktop, and wearable computing, any content is a product and any product is content. User experience and service design are essential considerations in any project, from the largest CMS to the smallest brochure — or even in the voice of a Twitter account. With a background in innovative editorial and data products, MDCCLV can help you create something that makes a dent in the universe.

 
 
How we group words into kinds will depend on the aim of the classification — and on our own inclination
— Wittgenstein, Philosophical Investigations, §17