direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Salar Mohtaj

Lupe

Research Field

- Natural Language Processing
- Machine Learning
- Information Retrieval

 Research Topics

- Text quality assessment

Biography

Salar is a researcher at the Quality and Usability Lab at Technische Universitat Berlin working on natural language processing applications in crowdsourcing.
His main interest is NLP and machine learning models. He received his bachelor degree from Shahrood University of Technology and master in information technology from AmirKabir University of Technology (Tehran Polytechnic).


Address

Quality and Usability Lab
Deutsche Telekom Laboratories
TU Berlin
Ernst-Reuter-Platz 7
D-10587 Berlin, Germany

Email: salar.mohtaj[at]tu-berlin.de
Tel:  +49 30 8353 58394

 

 

Publications:

Using External Knowledge Bases and Coreference Resolution for Detecting Check-Worthy Statements
Zitatschlüssel mohtaj2019a
Autor Mohtaj, Salar and Himmelsbach, Tilo and Woloszyn, Vinicius and Möller, Sebastian
Buchtitel Working Notes of (CLEF) Conference and Labs of the Evaluation Forum, Lugano, Switzerland, September 9-12, 2019.
Seiten 1–8
Jahr 2019
Adresse RWTH Aachen, Ahornstr. 55, 52056 Aachen, Germany
Monat sep
Notiz online
Verlag CEUR-WS.org
Serie CEUR Workshop Proceedings
Wie herausgegeben Fullpaper
Zusammenfassung With the proliferation of online information sources, it hasbecome more and more difficult to judge the trustworthiness of a state-ment on the Web. Nevertheless, recent advances in natural languageprocessing allow us to analyze information more objectively according tocertain criteria - e.g. whether a proposition is factual or opinative, or eventhe authority or credibility of an author in a certain topic. In this paper,we formulated a ranking schema that can be employed in textual claimsfor speeding up the human fact-checking process. Our experiments haveshown that our proposed method statistically outperformed the base-line. Additionally, this work describes a multilingual data set of claimscollected from several fact-check websites, which was used to fine-tuningour model.
Link zur Publikation Link zur Originalpublikation Download Bibtex Eintrag

Zusatzinformationen / Extras

Direktzugang

Schnellnavigation zur Seite über Nummerneingabe