profile picture

Tiago Almeida

PhD student @ IEETA-UA

Education

PhD candidate in Informatics Engineering

Aveiro University
2020 - present

MSc in Informatics Engineering

Aveiro University (18.9)
2017 - 2019

BSc in Informatics Engineering

Aveiro University (17.4)
2014 - 2017

Experiences

PhD Scholarship

FCT
2021 - Present

Research grant

IEETA - Bioinformatics
2019 - 2021

Research integration grant

IEETA - Bioinformatics
2018 - 2019

Awards

Top-team at BC8 track 2 (SympTEMIST) - Top-competitor (2nd places) at the BC8 SympTEMIST track (https://zenodo.org/records/10104547)
BioASQ 2023 Winner - Challenge winner at the 2023 BioASQ (http://bioasq.org/)
BioASQ 2022 Winner - Challenge winner at the 2022 BioASQ (http://bioasq.org/)
Best-team at BC7 track 2 (NLM-Chem Track) - Top-competitor (1st places) at the BC7 NLM-Chem Track (https://academic.oup.com/database/article/doi/10.1093/database/baad005/7071696?login=false)
BioASQ 2021 Winner - Challenge winner at the 2021 BioASQ (http://bioasq.org/)
DGES Merit Award - Awarded by DGES to the top 25 students (2018/2019) of the Aveiro University
BioASQ 2020 Winner - Challenge winner at the 2020 BioASQ (http://bioasq.org/)
Prémio BPI - Awarded by the Aveiro University to the best graduate student of a bachelor or integrated master's degree in engineering that is not awarded with specific prize.
NEI mention - Best student of BSc in Informatics Engineering - Awarded by NEI (Núcleo de Estudantes de Informática).

Publications

  • BIT.UA at Biocreative VIII track 1: A joint model for relation classification and novelty detection
    • Tiago Almeida, Richard Jonker, João Almeida, Sérgio Matos
    Proceedings of the BioCreative VIII Challenge Evaluation Workshop, 2023
  • Team BIT.UA @ BC8 SympTEMIST Track: A Two-Step Pipeline for Discovering and Normalizing Clinical Symptoms in Spanish.
    • Richard Jonker, Tiago Almeida, Sérgio Matos, João Almeida
    Proceedings of the BioCreative VIII Challenge Evaluation Workshop, 2023
  • BIT. UA at MedProcNER: discovering medical procedures in Spanish using transformer models with MCRF and augmentation
    • Tiago Almeida, Richard Jonker, Roshan Poudel, Jorge M. Silva, Sérgio Matos
    Proceedings of the Working Notes of CLEF 2023.
  • BIT. UA at BioASQ 11B: Two-Stage IR with Synthetic Training and Zero-Shot Answer Generation
    • Tiago Almeida, Richard Jonker, Roshan Poudel, Jorge M. Silva, Sérgio Matos
    Proceedings of the Working Notes of CLEF 2023.
  • Automatic extraction of medication mentions from tweets—overview of the biocreative VII shared task 3 competition
    • Davy Weissenbacher, Karen O’Connor, Siddharth Rawal, Yu Zhang, Richard Tzong-Han Tsai, Timothy Miller, Dongfang Xu, Carol Anderson, Bo Liu, Qing Han, Jinfeng Zhang, Igor Kulev, Berkay Köprü, Raul Rodriguez-Esteban, Elif Ozkirimli, Ammer Ayach, Roland Roller, Stephen Piccolo, Peijin Han, VG Vinod Vydiswaran, Ramya Tekumalla, Juan M Banda, Parsa Bagherzadeh, Sabine Bergler, João F Silva, Tiago Almeida, Paloma Martinez, Renzo Rivera-Zavala, Chen-Kai Wang, Hong-Jie Dai, Luis Alberto Robles Hernandez, Graciela Gonzalez-Hernandez
    Journal Database Volume 2023.
  • Deep Learning solutions based on fixed contextualized embeddings from PubMedBERT on BioASQ 10b and traditional IR in Synergy
    • Tiago Almeida, André Pinho, Rodrigo Pereira, Sérgio Matos
    Proceedings of the Working Notes of CLEF 2022.
  • Chemical identification and indexing in PubMed full-text articles using deep learning and heuristics
    • Tiago Almeida, Rui Antunes, João Figueira Silva, João Almeida, Sérgio Matos
    Journal Database Volume 2022.
  • Chemical detection and indexing in PubMed full text articles using deep learning and rule-based methods
    • Tiago Almeida, Rui Antunes, João Figueira Silva, João Almeida, Sérgio Matos
    Proceedings of the BioCreative VII Challenge Evaluation Workshop, Online event, p. 210-213, November 2021
  • Chemical–protein relation extraction in PubMed abstracts using BERT and neural networks
    • Rui Antunes, Tiago Almeida, João Figueira Silva, Sérgio Matos
    Proceedings of the BioCreative VII Challenge Evaluation Workshop, Online event, p. 210-213, November 2021
  • Drug mention recognition in Twitter posts using a deep learning approach
    • João Figueira Silva, Tiago Almeida, Rui Antunes, João Almeida, Sérgio Matos
    Proceedings of the BioCreative VII Challenge Evaluation Workshop, Online event, p. 210-213, November 2021
  • Universal passage weighting mechanism (UPWM) in BioASQ 9b
    • Tiago Almeida & Sérgio Matos
    Proceedings of the Working Notes of CLEF 2021
  • BioASQ Synergy: A strong and simple baseline rooted in relevance feedback
    • Tiago Almeida & Sérgio Matos
    Proceedings of the Working Notes of CLEF 2021
  • Benchmarking a transformer-FREE model for ad-hoc retrieval
    • Tiago Almeida & Sérgio Matos
    16th Conference of the European Chapter of the Association for Computational Linguistics
  • BIT. UA@ TREC 2020 Precision Medicine Track
    • Tiago Almeida & Sérgio Matos
    Text Retrieval Conference 2020
  • BIT. UA@ TREC 2020 Deep Learning Track
    • Tiago Almeida & Sérgio Matos
    Text Retrieval Conference 2020
  • Evaluating a lightweight neural reranking model for biomedical question answering
    • Tiago Almeida & Sérgio Matos
    26th Portuguese Conference on Pattern Recognition
  • BIT.UA at BioASQ 8 Lightweight neural document ranking with zero-shot snippet retrieval
    • Tiago Almeida & Sérgio Matos
    BioASQ workshop 2020 - CLEF 2020
  • Frugal neural reranking: evaluation on the Covid-19 literature
    • Tiago Almeida & Sérgio Matos
    1st Workshop onNLP for COVID-19 (Part 2) at EMNLP 202
  • Calling attention to passages for biomedical question answering
    • Tiago Almeida & Sérgio Matos
    ECIR 2020, Advances in Information Retrieval pp 69-77
  • Neural-based snippet extraction for biomedical question answering.
    • Tiago Almeida & Sérgio Matos
    25th Portuguese Conference on Pattern Recognition (2019), 79–80
  • Microtransactions in the Company’s and the Player’s Perspective: A Manual and Automatic Analysis
    • Gusmão, P., Almeida, T., Lopes, F., Muryn, Y., Martins, J., & Au-Yong-Oliveira, M.
    New Knowledge in Information Systems and Technologies. WorldCIST’19 2019 (pp. 440–451)

    Projects

    European Health Data and Evidence Network
    Ref: info:eu-repo/grantAgreement/EC/H2020/806968/EU
    Funding: European Commission (Horizon 2020 Framework Programme (H2020))
    Participation: Masters research grant ( 2019/09/01 - )
    European Medical Information Framework
    Ref: info:eu-repo/grantAgreement/EC/FP7/115372/EU
    Funding: European Commission (SEVENTH FRAMEWORK PROGRAMME (FP7))
    Participation: Research integration grant ( 2018/04/01 - 2019/03/31)

    Public software

    easyHPC - (UnderDevelopment) Easy script-based tool for installing an configure an HPC-based system (Users, Work balancer, Network Storage, GPU access) over a cluster network.
    Midas - MIDAS MultI-framework DataloAderS. MIDAS leverages the power of the DLPack protocol to export highly efficient tf.data.Dataset to other Deep Learning frameworks, like JAX or PyTorch.
    Polus - Functional deep learning training framework based on tensorflow, focused on research and performance
    Covid Search - Deep Learning powerd Search engine over the CORD-19 corpus. With purpose to facilitate the biomedical search about covid-19.
    Multidimesional RNN library with Keras and Tensorflow - Open source library that exposes multidimensional recurrent neural networks following a Keras API
    SuperCool - A three.js game inspered on SuperHot made for the VI class, jointly done with Yuriy Muryn
    BioASQ 7b document ranking - Adaptation of the DeepRank model with attention layers for the BioASQ 7b document retrieval task. It also supports an unsupervised way of snippet extraction.