Search Results for author: Carlos Soares

Found 23 papers, 11 papers with code

Lag Selection for Univariate Time Series Forecasting using Deep Learning: An Empirical Study

1 code implementation18 May 2024 José Leites, Vitor Cerqueira, Carlos Soares

Cross-validation approaches show the best performance for lag selection, but this performance is comparable with simple heuristics.

Time Series Univariate Time Series Forecasting

Time Series Data Augmentation as an Imbalanced Learning Problem

1 code implementation29 Apr 2024 Vitor Cerqueira, Nuno Moniz, Ricardo Inácio, Carlos Soares

We use these techniques to create synthetic time series observations and improve the accuracy of forecasting models.

Data Augmentation Time Series

Kernel Corrector LSTM

no code implementations28 Apr 2024 Rodrigo Tuna, Yassine Baghoussi, Carlos Soares, João Mendes-Moreira

We empirically evaluate the forecasting accuracy and the training time of the new algorithm and compare it with cLSTM and LSTM.

On-the-fly Data Augmentation for Forecasting with Deep Learning

no code implementations25 Apr 2024 Vitor Cerqueira, Moisés Santos, Yassine Baghoussi, Carlos Soares

We validated the proposed approach using a state-of-the-art deep learning forecasting method and 8 benchmark datasets containing a total of 75797 time series.

Data Augmentation Synthetic Data Generation +1

tsMorph: generation of semi-synthetic time series to understand algorithm performance

no code implementations3 Dec 2023 Moisés Santos, André de Carvalho, Carlos Soares

In this paper, we demonstrate the utility of tsMorph by assessing the performance of the Long Short-Term Memory Network forecasting algorithm.

Time Series Time Series Forecasting

Meta-aprendizado para otimizacao de parametros de redes neurais

no code implementations10 Jul 2021 Tarsicio Lucas, Teresa Ludermir, Ricardo Prudencio, Carlos Soares

In the current work, we performed a case study using meta-learning to choose the number of hidden nodes for MLP networks, which is an important parameter to be defined aiming a good networks performance.

Meta-Learning regression

Model Compression for Dynamic Forecast Combination

1 code implementation5 Apr 2021 Vitor Cerqueira, Luis Torgo, Carlos Soares, Albert Bifet

In this paper, we leverage the idea of model compression to address this problem in time series forecasting tasks.

Model Compression Time Series +1

Model Selection for Time Series Forecasting: Empirical Analysis of Different Estimators

1 code implementation1 Apr 2021 Vitor Cerqueira, Luis Torgo, Carlos Soares

We address this issue and compare a set of estimation methods for model selection in time series forecasting tasks.

Model Selection Time Series +1

Promoting Fairness through Hyperparameter Optimization

2 code implementations23 Mar 2021 André F. Cruz, Pedro Saleiro, Catarina Belém, Carlos Soares, Pedro Bizarro

Considerable research effort has been guided towards algorithmic fairness but real-world adoption of bias reduction techniques is still scarce.

Fairness Fraud Detection +1

u-cf2vec: Representation Learning for Personalized Algorithm Selection in Recommender Systems

no code implementations9 Mar 2021 Tomas Sousa-Pereira, Tiago Cunha, Carlos Soares

In the meta level, the meta learning model will use the metafeatures to extract knowledge that will be used to predict the best algorithm for each user.

Collaborative Filtering Meta-Learning +2

VEST: Automatic Feature Engineering for Forecasting

3 code implementations14 Oct 2020 Vitor Cerqueira, Nuno Moniz, Carlos Soares

Time series forecasting is a challenging task with applications in a wide range of domains.

Feature Engineering feature selection +3

A Bandit-Based Algorithm for Fairness-Aware Hyperparameter Optimization

no code implementations7 Oct 2020 André F. Cruz, Pedro Saleiro, Catarina Belém, Carlos Soares, Pedro Bizarro

Hence, coupled with the lack of tools for ML practitioners, real-world adoption of bias reduction methods is still scarce.

Decision Making Fairness +2

Machine Learning vs Statistical Methods for Time Series Forecasting: Size Matters

1 code implementation29 Sep 2019 Vitor Cerqueira, Luis Torgo, Carlos Soares

Using a learning curve method, our results suggest that machine learning methods improve their relative predictive performance as the sample size grows.

BIG-bench Machine Learning Time Series +2

Characterizing classification datasets: a study of meta-features for meta-learning

2 code implementations30 Aug 2018 Adriano Rivolli, Luís P. F. Garcia, Carlos Soares, Joaquin Vanschoren, André C. P. L. F. de Carvalho

These characterizations, also called meta-features, describe properties of the data which are predictive for the performance of machine learning algorithms trained on them.

BIG-bench Machine Learning General Classification +1

Algorithm Selection for Collaborative Filtering: the influence of graph metafeatures and multicriteria metatargets

3 code implementations23 Jul 2018 Tiago Cunha, Carlos Soares, André C. P. L. F. de Carvalho

However, the results have shown that the feature selection procedure used to create the comprehensive metafeatures is is not effective, since there is no gain in predictive performance.

Collaborative Filtering feature selection

Smart energy management as a means towards improved energy efficiency

no code implementations8 Feb 2018 Dylan te Lindert, Cláudio Rebelo de Sá, Carlos Soares, Arno J. Knobbe

The costs associated with refrigerator equipment often represent more than half of the total energy costs in supermarkets.

energy management Management

Learning Word Embeddings from the Portuguese Twitter Stream: A Study of some Practical Aspects

no code implementations4 Sep 2017 Pedro Saleiro, Luís Sarmento, Eduarda Mendes Rodrigues, Carlos Soares, Eugénio Oliveira

Using a single GPU, we were able to scale up vocabulary size from 2048 words embedded and 500K training examples to 32768 words over 10M training examples while keeping a stable validation loss and approximately linear trend on training time per epoch.

Learning Word Embeddings Playing the Game of 2048

FEUP at SemEval-2017 Task 5: Predicting Sentiment Polarity and Intensity with Financial Word Embeddings

2 code implementations SEMEVAL 2017 Pedro Saleiro, Eduarda Mendes Rodrigues, Carlos Soares, Eugénio Oliveira

This paper presents the approach developed at the Faculty of Engineering of University of Porto, to participate in SemEval 2017, Task 5: Fine-grained Sentiment Analysis on Financial Microblogs and News.

Sentiment Analysis Word Embeddings

Cannot find the paper you are looking for? You can Submit a new open access paper.