Services Offered

Alias-i is an appropriate vendor for customers who need specialized applications that are not met by other vendors (see our competition page). We are a a good choice if you have internal staff with solid Java skills and curiosity about computational linguistics. Our offerings include commercial licensing of LingPipe, consulting services as well as basic R&D efforts for our customers.

Licensing LingPipe

LingPipe is available under licensing terms that range from free to perpetual server licenses. The distinctions between the licenses turn on uses that LingPipe can be used for, extent of indemnification, guarantees and support. See our download page for an overview of the standard licenses. Contact us if the available licenses do not meet your needs.

Consulting Services

Building NLP systems is a bit of a black art when it comes to deciding feasibility, how to encode desired processing into the available tools and system tuning. Our preferred model for working with customers is to teach them how to build NLP systems rather than leave them dependent on us for system development.

A commonly occurring arrangement for classification style projects proceeds as follows.

  1. Customer provides 10 examples of manually processed text for which the customer desires an automated solution.
  2. Alias-i provides feasibility based on examples.
  3. Customer provides additional examples based on Alias-i feedback on applicability of state-of-the-art to business needs. Objective function defined for desired performance goals.
  4. Initial prototype developed by Alias-i in conjunction with customer developers with a shared svn archive.
  5. Additional training/evaluation data applied
  6. Acceptance testing on linguistic goals
  7. Hand off to customer developers with occasional interaction with Alias-i
  8. Customer is responsible for integration of LingPipe functionality

There is also a a more detailed blog post.

The goal of this process is to incrementally teach the customer developers how to do NLP for the problems they care about as a by product of annotation and shared coding. It has been very successful.

We have also done a range of information extraction projects, spelling correction and customized database linking applications.

Research and Development

We offer development of customized extensions of the LingPipe API for customers desiring capabilities not currently in the suite of tools. For example we developed query spell checking at the request of Thomson-Reuters WestLaw.

We have a history of government research contracts and grants. The company was founded with DARPA funding under the TIDES program and has more recently received two grants from the U.S. National Institutes of Health (NIH) for linking text mentions to gene and protein databases.

We internally fund collaborations with academic institutions as well. We have an ongoing collaboration with Harvard Medical School to create the world's best indexed research data collection for the autism literature using non-specialists as annotators, as part of the BioNotate project.

Company History

Alias-i was founded by Breck Baldwin 1999 under the name Baldwin Language Technologies. The original funding source was a Defense Advanced Research Project Agency (DARPA) research grant under the Translingual Information Detection, Extraction and Summarization (TIDES) program.

In early 2002, Baldwin Language Technologies relocated to New York from Philadelphia and began doing business as Alias-i. In 2003 we released the LingPipe tool kit and it has become the focus of our development. We offer consulting services around LingPipe as well as basic R&D services for government, commercial and academic customers.


Breck Baldwin, President

Breck Baldwin received his Ph.D. in computer science in 1995 from the University of Pennsylvania. In the time between his thesis on coreference resolution and evaluation and founding Alias-i in 1999, Breck worked on DARPA-funded projects through UPenn.

Bob Carpenter, Software Architect

Bob Carpenter received his Ph.D. in cognitive and computer science in 1989 from the University of Edinburgh. Between 1988 and 1996, he was a post-doc through associate professor at Carnegie Mellon University. He worked in the multimedia communications research group at Bell Laboratories between 1996 and 2000, then at the speech-recognition startup SpeechWorks, between 2000 and joining Alias-i in early 2002.