SystemT - overview


New!

We are publicly releasing Version 1.0 of the Universal Proposition Banks for multilingual semantic role labeling!

SystemT Online Classes: [here

Hiring Now!

Multiple positions available. Email your resume to Laura Chiticariu: chiti {at} us{.}ibm{.}com

 

Highlights

  • State-of-the-art AQL language for expressing NLP algorithms, optimizer and runtime engine for execution at scale, and easy to use user interface [demo]
  • Publications in top NLP, database systems, hardware and HCI conferences
  • Winner of multiple IBM Corporate Awards for its contributions to IBM products and clients
  • Currently taught in multiple universities
  • SystemT explained in 5 minutes

 

Recent Events [more]

  • We are giving a talk on Crosslingual Text Analytics at the Natural Language and Dialog Systems Lab, UC Santa Cruz, in May 2017.
  • Demo paper on learning extractors from examples accepted at SIGMOD 2017
  • We are teaching a lecture on SystemT at NYU Abu Dhabi in February 2017
  • Research paper on learning extactors from examples accepted by CHI 2017
  • We are giving a talk on Declarative Information Extraction and Multilingual SRL at the Stanford Logic Seminar in January 2017.
  • Two papers on Multilingual SRL and our Multilingual Information Extraction demo acepted at COLING 2016 [video]
  • We presented our semi-automatic approach for generating propositional banks for low-resource languages at EMNLP 2016
  • We demonstrated our multilingual Semantic Role Labeler at ACL 2016 [video] [paper]

 

SystemT

Information extraction (IE) refers to the task of extracting structured information from unstructured or semi-structured data. In recent years, IE has become increasingly important to a wide array of enterprise applications, ranging from Business Intelligence to Data-as-a-Service. Such applications drive the following main requirements for IE systems: accuracy, productivity, scalability, expressiviity, transparency, and customizability.

SystemT, a declarative IE system, has been designed and developed to address these requirements. It is based on the basic principle underlying relational database technology: complete separation of specification from execution. SystemT uses a declarative rule language, AQL, and an optimizer that generates high-performance algebraic execution plans for AQL rules. It makes IE orders of magnitude more scalable and easy to use, maintain and customize.

SystemT ships today with multiple products across 4 IBM Software Brands. Furthermore, SystemT is used in multiple ongoing research projects and being taught in universities. Our ongoing research and development efforts focus on making SystemT more usable for both technical and business users, and continuing enhancing its core functionalities based on natural language processing, machine learning, and database technology.

 

 

 

 

 

 

 

 




Awards

2013 - IBM Research Outstanding Technical Accomplishment Award

2013 - IBM Research A-Level Accomplishment Award

2010 - IBM Research A-Level Accomplishment Award

2008 - IBM Research A-Level Accomplishment Award