Takwa Kochbati

Takwa Kochbati is a lead data scientist at YZR. Takwa has also worked as a PhD student at CEA - Commissariat à l'énergie atomique et aux énergies alternatives from November 2018 to October 2021. Her research focuses on bridging the gap between natural language system requirements and preliminary architecture design models in the context of complex systems.

During their research, they had the opportunity to deal with a wide range of machine learning (ML) and natural language processing (NLP) techniques. Particularly, they worked on text semantic similarity extraction from both word-level and statement-level using word embeddings (word2vec, doc2vec…) as well as text clustering (k-means, hierarchical agglomerative clustering algorithm HAC..). Another part of their research focuses on implementing specific NLP heuristics to extract relevant information from text.

Kochbati has acquired skills in POS tagging, word embedding, named entity recognition, text semantic similarity, text clustering, and summarization. Takwa is experienced in using NLTK, Tensorflow, pandas, numpy, StanfordCoreNLP, Scikit-learn, Gensim, Spacy, and other libraries.

Dr. Takwa Kochbati has a PhD in Computer Science from the Université Paris-Saclay, and a degree in Software Engineering and Information Systems from the Ecole Nationale des Sciences de l'Informatique (ENSI). Takwa also has a background in Mathematics and Physics from the Institut Préparatoire Aux Etudes d'Ingénieurs de Tunis (IPEIT). In addition, they have completed three certification courses on TensorFlow from Coursera: Fine Tune BERT for Text Classification, Transfer Learning for NLP with TensorFlow Hub, and Tweet Emotion Recognition.