Computer Systems Engineer with a master's degree in Computer Science and a Ph.D. in Computational Linguistics. I recently joined as a postdoc for the CLTL group at the Vrije Universiteit Amsterdam where I work with Prof. Dr. Antske Fokkens and Prof. Dr. Piek Vossen , particularly collaborating in the InTaVia project for European Cultural Heritage.
Development of NLP models and text mining strategies for automatic processing, integration and dynamic language analysis of biographies and other historiographical texts. This work is part of the InTaVia project, which aims to address major research challenges and bridge the semantic gap between large object databases, biography databases, and present-day users across Europe.
Implementation of SOTA lemmatizers and part-of-speech taggers for large-scale (~50 billion tokens) German resources.
My research concerns on finding effective methods for creating more training data for the task of Semantic Role Labeling (SRL) in several languages (but particularly German). I approach SRL as a sequence classification task and also as a sequence generation task. Through my research I have worked with Recurrent Neural Networks , LSTM's , Sequence-to-Sequence models and multilingual Neural Language Models such as ELMo and BERT. I implemented my research code in Pytorch and also used state-of-the-art frameworks such as SpaCy , Transformers and AllenNLP when building more complex models.
I collaborated on a project for constructing a multilingual fine-grained entity classifier
I focused on implementing dashboards for business analytics using Tableau . Some of my projects included developing machine learning algorithms (clustering and classification) using Python and Scikit-learn . My latest project was a demo for Customer Sentiment Analysis on Twitter.
Mobile App prototype design and development. I used Appcelerator Titanium for cross-platform apps and also implemented a couple of native apps using Swift (for iPhone) and Java (for Android). As a web developer I used frameworks such as Wordpress. When building projects from scratch, I focused on the Backend (mainly NodeJS and MongoDB), however I also produced some code for the Frontend (mainly using React.)
I participated as an assistant for financial consulting projects such as: the creation of an automatic executive report software on Microsoft Excel, exploitation of Bank Risk Management databases using SAS, and analysis for the improvement of the Credit-card system performance and documentation on BBVA.
Daza, A. and Frank, A. (2020). To be presented at EMNLP 2020 - Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing. On-line Conference .[PDF]
Daza, A. and Frank, A. (2019). Published at EMNLP-IJCNLP 2019 - Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing, Hong Kong, China .[PDF]
Daza, A. and Frank, A. (2018). Published at ACL 2018 - Proceedings of the 3rd Workshop on Representation Learning for NLP (RepL4NLP), Melbourne, Australia.[PDF]
Daza A., Calvo H., Figueroa-Nazuno J. (2016). Published at NAACL 2016 - Workshop on Computational Linguistics for Literature, San Diego, California.[PDF]
When I am not in front of a computer I enjoy reading, both non-fiction (science, history, philosophy) and fiction: Borges, Dostoevsky, Mishima and Rulfo are some of my favorite authors. If you want to checkout what I read, feel free to visit my GoodReads profile and friend me there!
I am also a big football fan, particularly I enjoy watching the Champions League after a nice day at work.
I love traveling and getting to know new cultures and languages (I speak Spanish and English fluently and have learned some Portuguese, German and lately Dutch). Photography and Movies are also pleasant hobbies for the weekend.