Percorrer por data de Publicação, começado por "2024-08-12"
A mostrar 1 - 2 de 2
Resultados por página
Opções de ordenação
- Multimodal sentiment classifier framework for different scene contextsPublication . Cardoso, Pedro; Rodrigues, Joao; de Matos, Nelson Manuel da SilvaSentiment analysis (SA) is an effective method for determining public opinion. Social media posts have been the subject of much research, due to the platforms’ enormous and diversified user bases that regularly share thoughts on nearly any subject. However, on posts composed by a text–image pair, the written description may or may not convey the same sentiment as the image. The present study uses machine learning models for the automatic sentiment evaluation of pairs of text and image(s). The sentiments derived from the image and text are evaluated independently and merged (or not) to form the overall sentiment, returning the sentiment of the post and the discrepancy between the sentiments represented by the text–image pair. The image sentiment classification is divided into four categories—“indoor” (IND), “man-made outdoors” (OMM), “non-man-made outdoors” (ONMM), and “indoor/outdoor with persons in the background” (IOwPB)—and then ensembled into an image sentiment classification model (ISC), that can be compared with a holistic image sentiment classifier (HISC), showing that the ISC achieves better results than the HISC. For the Flickr sub-data set, the sentiment classification of images achieved an accuracy of 68.50% for IND, 83.20% for OMM, 84.50% for ONMM, 84.80% for IOwPB, and 76.45% for ISC, compared to 65.97% for the HISC. For the text sentiment classification, in a sub-data set of B-T4SA, an accuracy of 92.10% was achieved. Finally, the text–image combination, in the authors’ private data set, achieved an accuracy of 78.84%.
- Exploring the relationship between clean energy indices and oil prices: a ten-day window approachPublication . Dias, Rui; Galvão, Rosa; Cruz, Sandra P.; Irfan, Mohammad; Teixeira, Nuno; Gonçalves, SidalinaThis paper aims to assess the comovements between clean energy indices, namely the Clean Energy Fuels (CLNE), Nasdaq Clean Edge Green Energy (CELS), S&P Global Clean Energy (SPGTCLEN), TISDALE Clean Energy (TCEC.CN), Wilderhill (ECO), West Texas Intermediate (WTI) stock indices, over the period from 1 January 2018 to 23 November 2023. We used 10-day windows to analyse the duration and nature of the shocks. Granger causality tests revealed that 20 of the 30 possible pairs showed significant movements, with the WTI influencing all the clean energy indices, highlighting its global importance. CELS also showed a robust influence on all pairs, while SPGTCLEN had a significant but less far-reaching influence. The CLNE and ECO indices showed limited influences, suggesting the potential for diversification, the TCEC.CN proved to be independent and a determining factor for portfolio diversification. The Impulse Response Functions (IRF) confirmed significant movements between CELS, SPGTCLEN and WTI, reflecting the market's response to policies and adjustments in expectations. Fluctuations in oil prices substantially affect clean energy indices, highlighting the interconnectedness and volatility of these markets. In conclusion, these results indicate that despite the growth of clean energy, the sector is still influenced by fluctuations in the fossil fuel market.