Current Position

  • Research Associate in Open-Domain Statistical Spoken Dialogue Systems.
    Member of the Dialogue Systems Group.
    Machine Intelligence Laboratory, Information Engineering Division.
    Department of Engineering at the University of Cambridge, UK.

Academic Degrees


Research Interests

  • Natural Language Processing, Dialogue Systems

    • Development of complete frameworks and toolkits for rapid prototyping of dialogue systems and for supporting dialogs in serious games.
    • Data collection (i.e. Wizard-of-OZ experiments) and automatic semantic annotation of dialogue corpora.
    • Application of Machine Learning techniques for interpretation and question answer in situated dialogues.
    • Development of NLP tools for processing romance languages (e.g. Italian, French and Spanish).

  • Knowledge-representation Application of description logics knowledge-representation and inference systems to the medical domain.

  • Biomedical Text Mining and Healthcare dialogue systems. Application of Information Extraction tecniques for mining clinical reports and biomedical literature.


Academic and Industrial Experience

  • Université de Lorraine Postdoctoral Researcher. Projects ContNomina (entity recognition) and Empathic (sentiment analysis). SYNALP Team at Loria.

  • Temporary Associated Professor and Researcher, Attaché temporaire d'enseignement et de recherche (ATER). Université de Lorraine.

  • INRIA Postdoctoral Position. Project Emospeech. Synalp Team. LORIA. Nancy France.

  • CNRS Postdoctoral Position. PORTMEDIA Project. Talaris Team. LORIA. Nancy France.

  • Doctoral scholarship. Laboratory for Biomedical Informatics. University of Pavia. Italy.

  • CSIC Predoctoral Scholarship. Bioinformatics support, development and design of web applications. CNB Proteomics Facility and PROTEORED. Centro Nacional de Biotecnologia. Madrid Spain.

  • Software analyst and developer. Bussiness and critical-operation products for telecomunications and public service companies.
    Open International Systems. Cali Colombia.

  • Developer in projects for Human Resources Management for Gillette de Colombia S.A.




Cambridge University Engineering Department, Room BN3-011

Trumpington Street

Cambridge CB2 1PZ


Telephone:(+44) 1223 765 758





    Information Extraction & Sentiment Analysis

  • Lina Maria Rojas-Barahona. Deep Learning for Sentiment Analysis. Language and Linguistics Compass. 10.12 (2016): 701-719.
  • Lina M. Rojas-Barahona and Christophe Cerisara. Weakly Supervised Discriminative Training of Linear Models for Natural Language Processing. In proceedings of the 3rd International Conference on Statistical Language and Speech Processing. SLSP 2015. Budapest, Hungary.
  • Lina M. Rojas-Barahona and Christophe Cerisara. Enhanced discriminative models with tree kernels and unsupervised training for entity detection. In proceedings of the 6th International Conference on Information Systems and Economic Intelligence. SIIE 2015. Hammamet,Tunisia.
  • Dialogue Systems and Language Understanding

  • L. Rojas-Barahona, M. Gasic, N. Mrksic, P.-H. Su, S. Ultes, T.-H. Wen and S. Young. Exploiting Sentence and Context Representations in Deep Neural Models for Spoken Language Understanding. In proceedings of the 26th International Conference on Computational Linguistics. Coling 2016. Osaka, Japan.
  • Lina M. Rojas-Barahona and Christophe Cerisara. Bayesian Inverse Reinforcement Learning for Modeling Conversational Agents in a Virtual Environment. In proceedings of the 15th International Conference on Intelligent Text Processing and Computational Linguistics. LNCS Springer. CICLING 2014, Kathmandu, Nepal.
  • Claire Gardent and Lina M. Rojas-Barahona. Using Paraphrases and Lexical Semantics to Improve the Accuracy and the Robustness of Supervised Models in Situated Dialogue Systems. In proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, EMNLP 2013, Seattle, USA.
  • Alejandra Lorenzo, Lina Rojas-Barahona and Christophe Cerisara. Unsupervised structured semantic inference for spoken dialog reservation tasks . Proceedings of the 14th annual SIGdial Meeting on Discourse and Dialogue. SIGDIAL 2013, Metz France.
  • Claire Gardent, Alejandra Lorenzo, Laura Perez-Beltrachini and Lina M. Rojas-Barahona. Weakly and Strongly Constrained Dialogues for Language Learning. Proceedings of the 14th annual SIGdial Meeting on Discourse and Dialogue. SIGDIAL 2013, Metz France.
  • Lina Maria Rojas-Barahona and Claire Gardent. What Should I Do Now? Supporting Progress in a Serious Game. Proceedings of the 16th workshop on the semantics and pragmatics of dialogue. SEINEDIAL 2012, Paris France.
  • Lina Maria Rojas-Barahona, Alejandra Lorenzo and Claire Gardent. An End-to-End Evalutation of Two Situated Dialog Systems. Proceedings of the 13th annual SIGdial Meeting on Discourse and Dialogue. SIGDIAL 2012, Seoul Korea.
  • Lina Maria Rojas-Barahona, Alejandra Lorenzo and Claire Gardent. Building and Exploiting a Corpus of Dialog Interactions between French Speaking Virtual and Human Agents. The eighth international conference on Language Resources and Evaluation. LREC 2012 , Istanbul, Turkey.
  • Fabrice Lefevre; Djamel Mostefa; Laurent Besacier; Yannick Esteve; Matthieu Quignard; Nathalie Camelin; Benoit Favre; Bassam Jabaian; Lina Maria Rojas-Barahona. Leveraging study of robustness and portability of spoken language understanding systems across languages and domains: the PORTMEDIA corpora. The eighth international conference on Language Resources and Evaluation. LREC 2012, Istanbul, Turkey.
  • Treveur Bretaudiere; Samuel Cruz-Lara; Lina Maria Rojas Barahona. Associating Automatic Natural Language Processing to Serious Games and Virtual Worlds. The Journal of Virtual Worlds Research, Virtual Worlds Institute, Inc, 2011, MPEG-V and Other Standards, 4 (3).
  • Lina Maria Rojas-Barahona and Matthieu Quignard. An Incremental Architecture for the Semantic Annotation of Dialogue Corpora with High-Level Structures. A case study for the MEDIA corpus. Proceedings of the 12th annual SIGdial Meeting on Discourse and Dialogue. SIGDIAL 2011, Portland Oregon. June 2011.
  • Lina Maria Rojas-Barahona, Thierry Bazillon, Matthieu Quignard and Fabrice Lefevre. Using MMIL for the High Level Semantic Annotation of the French MEDIA Dialogue Corpus. Proceedings of the 9th International Conference on Computational Semantics, IWCS 2011. Oxford, January 2011.
  • Alexandre Denis, Lina M. Rojas-Barahona and Matthieu Quignard. Extending MMIL Semantic Representation: Experiments in Dialogue Systems and Semantic Annotation of Corpora. In proceedings of the Fifth Joint ISO-ACL/SIGSEM Workshop on Interoperable Semantic Annotation. ISA-5, Hong Kong, January 2010.
  • Biomedical Informatics and Natural Language Processing

  • C. Larizza, M. Gabetta, L. Rojas Barahona, G. Milani, E. Guaschino, G. Sances, C. Cereda, R. Bellazzi. Extraction of Clinical Information from Clinical Reports: an Application to the Study of Medication Overuse Headaches in Italy. In proceedings of the AMIA Summit on Translational Bioinformatics, San Francisco, March 2010.
  • Silvana Quaglini, Toni Giorgino, Lina M. Rojas-Barahona, Ezio Caffi, Mauro De Vito, Alessandra Persico and Anna cavallini. An Automated Voice Response System for Anticoagulant Therapy Management.Proceedings of the XXII International Conference of the European Federation for Medical Informatics, MIE09, (2009)
  • Lina Maria Rojas-Barahona, Silvana Quaglini and Mario Stefanelli.HomeNL: Homecare Assistance in Natural Language. An Intelligent Conversational Agent for Hypertensive Patients Management. Proceedings of the 12th Conference on Artificial Intelligence in Medicine, AIME09 ,ed Carlo Combi, Yuval Shahar, Ameen Abu-Hanna. LNCS. Springer (2009) [pdf]
  • L.M. Rojas-Barahona, T. Giorgino. Adaptable dialog architecture and runtime engine (AdaRTE): A framework for rapid prototyping of health dialog systems. International Journal of Medical Informatics. Medinfo2007, IJMI vol. 78s ,pag. 56 - 68, (2009)[pdf]
  • Lina Maria Rojas-Barahona. Employing Combinatory Categorial Grammars in the Domain of Hypertensive Patients Management.Technical Report (2008). [pdf]
  • L. M. Rojas-Barahona, T. Giorgino. AdaRTE: Adaptable Dialogue Architecture and Runtime Engine. A New Architecture for Health-Care Dialogue Systems. In proceeddings of Medinfo 2007 vol. 12 ,n. Pt 2 ,pag. 1063 - 1067, (2007).
  • L.M Rojas-Barahona. Adapting Combinatory Categorial Grammars in a Framework for Health Care Dialogue Systems. Proceedings of the 11th Workshop on the Semantics and Pragmatics of Dialogue Decalog 2007 ,ed Ron Arstein and Laurie Vieupag. 187 - 188, (2007).
  • L.M Rojas-Barahona, T. Giorgino. AdaRTE: An Extensible and Adaptable Architecture for Dialog Systems. Bridging the Gap: Academic and Industrial Research in Dialog Technologies Workshop Proceedings ,Association for Computational Linguistics,pag. 64 - 67, NAACL 2007.

Academic Year 2013-2014

  • Lecturer
    • Natural Language Processing (NLP). Ingénierie des langues. Licence Sciences de la Cognition. UFR Mathématique-Informatique, Université de Lorraine.

Academic Year 2012-2013

Academic Year 2010-2011

  • Tutor
    • Langage JAVA Informatique 2. L1. L'Ecole des Mines, Nancy.