Sentilo: Frame-Based Sentiment Analysis

Recupero, Diego Reforgiato; Presutti, Valentina; Consoli, Sergio; Gangemi, Aldo; Nuzzolese, Andrea Giovanni

doi:10.1007/s12559-014-9302-z

Cited by 58 publications

(14 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many of these models take into account negations and intensifiers when computing the sentiment of the text, i.e., improving the results of dictionary-based sentiment exercises. As an example, the paper by [57] sets up a tool to extract opinions from a text by also taking into account the structure of sentences and the semantic relations between words.…”

Section: The Narrative About the Economy As A Shadow Forecast: An Analysis Using The Bank Of Spain Quarterly Reportsmentioning

confidence: 99%

New Data Sources for Central Banks

Ghirelli

Hurtado

Pérez

et al. 2021

Data Science for Economics and Finance

View full text Add to dashboard Cite

Central banks use structured data (micro and macro) to monitor and forecast economic activity. Recent technological developments have unveiled the potential of exploiting new sources of data to enhance the economic and statistical analyses of central banks (CBs). These sources are typically more granular and available at a higher frequency than traditional ones and cover structured (e.g., credit card transactions) and unstructured (e.g., newspaper articles, social media posts, or Google Trends) sources. They pose significant challenges from the data management and storage and security and confidentiality points of view. This chapter discusses the advantages and the challenges that CBs face in using new sources of data to carry out their functions. In addition, it describes a few successful case studies in which new data sources have been incorporated by CBs to improve their economic and forecasting analyses.

show abstract

Section: The Narrative About the Economy As A Shadow Forecast: An Analysis Using The Bank Of Spain Quarterly Reportsmentioning

confidence: 99%

New Data Sources for Central Banks

Ghirelli

Hurtado

Pérez

et al. 2021

Data Science for Economics and Finance

View full text Add to dashboard Cite

show abstract

“…Recently, automatically generating a sentence description for an image, has attracted much research attention in artificial intelligence. This problem, known as image captioning, plays an important role in computer vision, i.e., enabling computers to understand images, which can be exploited in wide applications, such as video tracking [28][29][30][31][32], cross-view retrieval [10,39], sentiment analysis [36,52], childhood education [54], and visual impairment rehabilitation [11]. However, image captioning is a challenging task due to the coverage of both computer vision and natural language processing technologies.…”

Section: Introductionmentioning

confidence: 99%

Neural Image Caption Generation with Weighted Training and Reference

et al. 2018

View full text Add to dashboard Cite

Image captioning, which aims to automatically generate a sentence description for an image, has attracted much research attention in cognitive computing. The task is rather challenging, since it requires cognitively combining the techniques from both computer vision and natural language processing domains. Existing CNN-RNN framework-based methods suffer from two main problems: in the training phase, all the words of captions are treated equally without considering the importance of different words; in the caption generation phase, the semantic objects or scenes might be misrecognized. In our paper, we propose a method based on the encoder-decoder framework, named Reference based Long Short Term Memory (R-LSTM), aiming to lead the model to generate a more descriptive sentence for the given image by introducing reference information. Specifically, we assign different weights to the words according to the correlation between words and images during the training phase. We additionally maximize the consensus score between the captions generated by the captioning model and the reference information from the neighboring images of the target image, which can reduce the misrecognition problem. We have conducted extensive experiments and comparisons on the benchmark datasets MS COCO and Flickr30k. The results show that the proposed approach can outperform the state-of-the-art approaches on all metrics, especially achieving a 10.37% improvement in terms of CIDEr on MS COCO. By analyzing the quality of the generated captions, we come to a conclusion that through the introduction of reference information, our model can learn the key information of images and generate more trivial and relevant words for images.

show abstract

“…Moreover, while a great deal of effort has been directed towards the extraction of frames from natural language, not many systems process frames further, to solve downstream tasks in NLP and AI -an example is Sentilo (Recupero et al, 2015), a sentiment analysis system built on top of the frame-based machine reading tool FRED by Presutti et al (2012).…”

Section: Introductionmentioning

confidence: 99%

Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics

2018

View full text Add to dashboard Cite

Sentilo: Frame-Based Sentiment Analysis

Cited by 58 publications

References 33 publications

New Data Sources for Central Banks

New Data Sources for Central Banks

Neural Image Caption Generation with Weighted Training and Reference

Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics

Contact Info

Product

Resources

About