You are here

Research Studentship: Extracting High Quality Knowledge from Text

Data & Knowledge Management (DKM) is a research unit joining the Artificial Intelligence area of the Center for Information Technology - Fondazione Bruno Kessler.

Data and knowledge management (DKM) comprises a range of practices used by organisations such as enterprises, communities of interest, social groups and single individuals, to create, represent, share, and exploit knowledge is any type of domain. DKM is an interdisciplinary research field on the border between computer science, artificial intelligence, businesses, philosophy, and mathematics.

The group aims to foster research and technology for developing methodologies and tools to support the elicitation of knowledge from data and experts, its encoding in computer interpretable formats, the integration of heterogeneous knowledge bases, and the development of intelligent inference and decision support algorithms.

Job Description:

The selected candidate will contribute to the ongoing DKM research line on “extracting high quality knowledge about entities and events from textual documents”. This line focuses on two main research directions:

●     To improve the quality of the knowledge extracted from text exploiting background knowledge and inference: State-of-the-art knowledge extraction architectures combine complementary NLP tools, each of which is specialized in one particular task (e.g., named entity recognition and classification, entity linking, semantic role labeling). However, these tools typically work independently, solving ambiguous cases without taking into account the output of the other tools. Consequently, the combined output can be inconsistent or contradictory. The goal of this activity is to develop techniques for the coordination of the output of NLP extraction tools, in the light of background (logical and statistical) knowledge, in order to maximize the quality and coherence of the resulting knowledge;

●     To infer implicit knowledge from what explicitly contained in unstructured content: Goal of this activity is to develop techniques to derive (implicit) knowledge following from what explicitly mentioned and extracted from a document, analogously to the cognitive process that humans typically perform when reading and interpreting a text.

The selected candidate is expected to contribute to the development of (some of) the aforementioned techniques, leveraging state-of-the-art Artificial Intelligence paradigms such as (but not limited to) Deep Learning and Logical Reasoning.

Tutor: Marco Rospocher

Research Unit: DKM (

FBK International PhD Program: 


The ideal candidate would have:

●     Bachelor or MSc degree in Computer Science, ICT or Mathematics

●     Solid programming skills

●     Willingness to study new, challenging research topics and technologies

●     Commitment to work in a research-driven environment

●     Problem solving attitude

Candidates must have a MSc degree (or planning to get one before October 31st, 2018). Interested candidates should consider that opportunities to start a funded Ph.D. (subject to a public selection) on a topic related to the research activities described in this call will likely be available at the end of the position.



Type of contract: Fixed Term Contract – part time

Start date:  spring 2018

End date: 31st of October 2018

Gross annual salary (full time): about 33.300€

Work Place: Povo, Trento (Italy)

Benefits: flexi-time, company subsidized cafeteria or meal vouchers, internal car park, welcome office support for visa formalities, accommodation, social security, etc., reductions on bank account opening fees, public transportation, sport, language course fees. More info at


Application: Job Offers @ FBK.

Application deadline: 23rd of March 2018

Please read the Guidelines for Selection before completing your application.

For further information or technical issues regarding the application, please contact the Human Resources Service at .