[Post-Doc 2019] Formal and statistical modeling of dialogue application to pathological data

One postdoc position (1 year) in Natural Language Processing / Machine
Learning is open in the Semagramme and SyNaLP team at LORIA and at
ATILF. This position will be funded by the impact projet OLKI (Open
Language and Knowledge for Citizens) from Université de Lorraine.

Read the pdf of the call: postdocOlkiCall

Informations

- Starting date: fall 2019
- Duration: 1 year

- Deadline for application: September 09th, 2109 (Interviews end of
  September)

- Location: Nancy, France (1h30 from Paris by train)

- Salary: around 2,000 euros per month net income

- Informal inquiries can be sent by email to Maxime Amblard
  (maxime.amblard@loria.fr) and Chloé Braud (chloe.braud@loria.fr).

- Application: CV, motivation letter, PhD evaluation, master TOR and
  support letter(s)

Keywords

NLP, Discourse and Dialogue, machine learning, logic, corpora, natural
language, pathology

Scientific project

Modeling interaction is a crucial step for Natural Language Processing
(NLP), which requires the development of automatic tools able to
simulate these exchanges. A typical example is chatbots and all the
services based on them. But Dialogue Models face two types of
difficulties.

(I) The first issue concerns the availability of resources and models
that can analyze and process dialogues. Modelling dialogues is very
hard, in particular because conversations high- light particular uses
such as the relationship between questions and answers. Recent models
show that this issue is beyond semantics understanding, and even more
so beyond discourse. Another perspective is to use Machine Learning
approaches in order to identify dialogical relations and dialogical
interactions. Generally speaking discursive analysis aims at building
a structure representing the semantic links between sentences.

(II) The second is that dialogue models must be coordinated with
pragmatic inferences at a higher level. In this case, we can refer to
linguistic models of dialogue, or to models that capture conceptual
links. In particular, we propose to apply these approaches to the SLAM
project corpus. The SLAM project links thought and language
disorders. This corpus consists of transcriptions of interviews with
schizophrenic patients. Exchanges with these patients contain
interactions that are difficult to interpret semantically and
interactively.

Candidate skill

The post-doctoral fellow must have a PhD in Computational
Linguistics/NLP, Computer Science or related fields, with a good
programming skills. He/She must be fluent in English and have
demonstrated its ability to publish at the highest international
level. Note that Knowledge of French is NOT a requirement.

Supervision of students is possible, if wanted.

Supervisors

Maxime Amblard, MCF HDR, is a specialist in computational
linguistics. He is interested in the use of logical and formal tools
to model the semantics and pragmatics of natural language. His recent
research focuses on a corpus of pathological uses of the language and
dialogue's representation. http://members.loria.fr/mamblard/

Chloé Braud, CR CNRS, is a computer linguist specializing in discourse
parsing. The strategies she implements are based on machine learning
for domain adaptation tasks in a multilingual, cross-domain
context. https://chloebt.github.io/

Michel Musiol, PR is a psycholinguist. He is interested in the
analysis of the manifestation of thought disorders through
language. He studied the particularities of interviews with
shizophrenic patients. http://michel.musiol.free.fr/

The skills of the different supervisors are complementary for the
subject, both on the formal, digital and psycholinguistic aspects.

Contacts

Maxime Amblard, LORIA, Team Semagramme, maxime.amblard@loria.fr

Chloé Braud, LORIA, Team SyNaLP, chloe.braud@loria.fr