Nadia Okinina

Implementation of textual and audio data transformation pipelines for an English and French conversational agent training and evaluation. Ingestion of audios and metadata into a Mongo DB, update, quality maintenance. Technical environment: Python, MongoDB, AWS, Git Tools: MongoDB Compass, Airtable, PyCharm, Postman, Snowflake, sox, Praat, Audacity

Web development: implementation of the web site https://demonext.llf-paris.fr/ Processing of textual data coming from lexicographic resources and its integration into the Demonext derivational database of French. Technical environment: Python, PHP, MariaDB, Javascript, jQuery, Cytoscape, CSS, Bootstrap, PHPUnit, Jasmine, XML, Git

Multilingual learner corpora management: ◦ annotation format definition ◦ development of a transcription and annotation tool ◦ cleaning, reformating ◦ quality insurance Web development: • development of a tool to search keywords in journal articles in Italian and German in Python (Flask) https://github.com/nadezdaalexandrovna/keywords-docker • development of a textual complexity measurement tool in Java https://kommul.eurac.edu/ctapWebApp/, https://github.com/commul/ctap • development of a corpus transcription and annotation tool in Ruby on Rails and Javascript https://github.com/nadezdaalexandrovna/transcanno Development of Pepper manipulators for XML annotations transformation in Java https://github.com/nadezdaalexandrovna/transcanno-pepper-module, https://github.com/commul/CoreNLPPepper Research on • textual complexity of language learner texts • language learners’ error correction • automatic tagging of ancient Italian texts Writing scientific articles in English and Italian, participating in conferences and workshops. Technical environment: Python (Flask), ROR, Javascript, CSS, XML, Git (Github/Gitlab), Docker, Java, Maven, UIMA, GWT, Moses, Pepper, TreeTagger, Tint, Stanford Parser, Open NLP, SMOR, BabelNet

Implementation of anti-spam algorithms and tools: • regular expressions writing • web development Technical environment: Python, PHP, ROR (Ruby On Rails), Javascript, jQuery, AngularJS, CSS, SASS, Bootstrap, SVN

Automatic classification of Wikipedia articles according to the Prolexbase proper nouns ontology types. Technical environment: Java, Weka, SVM

Creation of the textual content of an English learning platform based on the interaction between the learner and animated avatars.

Web development: test and integration of the Xerox Incremental Parser into the Ami Enterprise Intelligence monitoring tool.