Matteo Grella plays the role of scientific responsible and coordinator for those projects where his oustanding skills acquired in the cognitive sciences and computational linguistics -language understanding, knowledge representation, automated reasoning, are crucial. He strenghtened his competences thanks to the projects and collaborations with research institutions including the Interaction Models Group of the University of Turin, the e-Lite Research Group of Politecnico di Torino, the ILC CNR of Pisa, the ART Group of University of Rome 'Tor Vergata' and FBK of the Autonomous Province of Trento. He is member of the Italian Association for Artificial Intelligence and the Association "Senso Comune".




Turin, Italy / Konstanz, Germany

Current Position:

Associate Director at Advanced Analytics GmbH


My research interests are in natural language processing, particularly on syntactic and semantic parsing algorithms and related machine learning systems.


I like to work with brilliant people who knows things I don't. My current team consists of Simone Cangialosi, Michele Luche and Edoardo Brambilla.


It may be necessary sometime to design and develop data/algorithm which are “aside” or “cross” the core technology focus of the projects. In these cases I’m glad to release the resources developed, as open source contribute.

Previous Projects

DAMANTIC from the foundation of Parsit's technological background offers cutting-edge text analytics software capable of understanding texts written in natural languages (Italian, English, Spanish, French and German) and tackle the Big Data challenges, offering advanced technological solutions.

MACHINE READING through simple API allows anyone to extract knowledge from unstructured text with little human effort. Machine Reading brings the power of Natural Language Processing (NLP) to many different destinations, helping break down the language barriers for man and machine, developers, webmasters and businesses.

VERBABOX is a portal of products and services aims to promote to large-scale the technology of computational linguistics.

PARSIT is his first iniziative to give industrial and commercial value to the results of research of Matteo Grella, Marco Nicola and Daniel Christen in the NLP field.


Vocabolario da Corpus: Sfide e Prototipi (Vocabulary from Corpus: Challenges and Prototypes)

Vittorio Coletti, Matteo Grella (Accademia della Crusca, 2017)

Tecnologie Semantiche Applicate ai Big Data (Semantic Technologies Applied to Big Data)

Matteo Grella (Salone del Libro - Book to the future, 2013)


Non-Projective Dependency Parsing via Latent Heads Representation (LHR)

Matteo Grella and Simone Cangialosi (, 2018)

Taking Advantage of BiLSTM Encoding To Handle Punctuation In Dependency Parsing: A Brief Idea

Matteo Grella (, 2018)

Notes About a More Aware Dependency Parser

Matteo Grella (, 2015)

Sentiment analysis via dependency parsing

Luigi Di Caro, Matteo Grella (Computer Standards & Interfaces 2012)

Parsit at Evalita 2011 Dependency Parsing Task

Matteo Grella, Marco Nicola (Springer 2011)

Rule-Based Creation of TimeML Documents from Dependency Trees

Livio Robaldo, Tommaso Caselli, Matteo Grella (AI*IA 2011)

From Italian Text to TimeML Document via Dependency Parsing

Livio Robaldo, Tommaso Caselli, Irene Russo, Matteo Grella (CICLing 2011)

Experiments with a Constraint-based Dependency Parser

Matteo Grella, Marco Nicola, and Daniel Christen (Evalita 2009)