Kepa J. Rodriguez

During my study I got a broad experience in translation from English and German text to Spanish. I have some additional experience in Italian to Spanish translation.

Experience in translations from Spanish, English and German to Basque.

During my study and professional journey I was active in tasks which required research and development in areas of AI. Most relevant were Natural Language Processing, Content Extraction, Content Generation, Machine Learning and Knowledge Representation.
An additional area of work consisted of the conception of the integration of AI components into infrastructures.

During my PhD I developed machine learning models for content extraction tasks, as named entity extraction, coreference resolution and ontology population. That included the use of standard tools and libraries like Scikit-Learn, Weka, CRF++, SVMLight, etc. As part of the work I developed and implemented workflows for text annotation and data quality assessment.

I my further work I developed statistical language models for text categorization, search and metadata quality assessment. I used self developed tools and existing SVM and boosting libraries.

I have experience in the implementation of models with Tensorflow and Pytorch in the following NLP areas:
- Named entity recognition in several languages.
- Text classification
- Relation detection and classification in biomedical text.

Outside of NLP, I am active in Moodle analytics with Tensorflow.

Experience in different areas of language processing:
- Models to extract structured information from text.
- Text classification.
- Relation detection and classification
- Coreference.
- Knowledge representation: especially controlled vocabularies for archival institutions and libraries.
- Corpus data development.
- Question generation from text.
- Named entity extraction.
Integration of NLP components in projects, as syntactic analyzers, named entity recognition libraries, and other modules of NLP libraries.

During my Phd I wrote most of the software used in my research in Java. This development included linguistic text analysis and feature engineering for machine learning components.

During my work at European projects I have collaborated actively in the development of Java code, including modules used for the integration of metadata of historical archival collections into the portal, parts of the search engine and semantic data interpretation and transformation.

An additional challenge which included Java development was a attempt to build a company in Israel. In this context Java was used to analize English text, extract and classify information and generate short texts.

Development of scripts for extraction and formatting of data in different projects.

I am actually active in the development of Python scripts and short programs for data extraction, modelling and transformation, natural language processing and machine learning.
Data extraction, transformation and modelling includes the use of libraries to work with XML, CSV, JSON, PDF and other standard formats.
For NLP and Machine Learning it includes the use of libraries like NLTK, Scikit-Learn, Tesorflow and PyTorch.