SemtiKamols - Natural Language Processing

Natural Language Processing (NLP) technology lately is emerging as one of the new application areas for the Grid computing. The computational challenges in this area are related to the deep semantic analysis of natural language texts, which are highly ambiguous by their very nature. The relevant semantic Knowledge Representation and Reasoning technology (the same one used also in Semantic Web and Description Logic languages like OWL-DL) is lately reaching the maturity level, where deep semantic analysis of large text corpuses is becoming theoretically possible.
Unfortunately, the computational complexity of the involved techniques is known to be
worst-case NP-complete, meaning that their solution requires large computational resources.
IMCS at Univerity of Latvia is conducting research in this area in the framework of
the National Research Programme project "SemtiKamols". For computationally intensive
tasks the project started with developing a local xgrid environment, but lately some of the
tasks are being ported also to the BalticGrid (EGEE gLite) environment.

"SemtiKamols" webpages are in Latvian, you can find more information in English in the Publications page.

DNLP: a prototype-grade system for distributing Prolog predicates

DNLP is a simple system to distribute a Prolog predicate amongst participating nodes. The system is a prototype for prolog-mpi, and is largely obsolete. DNLP allows a single predicate to take advantage of MPI environments, like grids, clusters, or multiprocessors, although the operational parameters are very constrained (as befits a prototype). The original goal of DNLP was to distribute an existing natural language processing application without modifying the Prolog code. This goal was met.

More information in the DNLP webpage.

Prolog MPI: a powerful Prolog environment for grid supercomputing

Prolog-MPI is a system for allowing prolog applications to take advantage of MPI environments, like grids, clusters, or multiprocessors. It must be used as a standalone application: instead of launching the prolog executable, one launches pl-mpi. The executing context is then free to use the exported predicates for distributing data.

More information - the Prolog-MPI webpage.