FineNews: fine-grained semantic sentiment analysis on financial microblogs and news
- PDF / 1,090,407 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 77 Downloads / 183 Views
ORIGINAL ARTICLE
FineNews: fine‑grained semantic sentiment analysis on financial microblogs and news Amna Dridi1 · Mattia Atzeni1 · Diego Reforgiato Recupero1 Received: 1 March 2017 / Accepted: 7 March 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract In this paper, a fine-grained supervised approach is proposed to identify bullish and bearish sentiments associated with companies and stocks, by predicting a real-valued score between − 1 and + 1. We propose a supervised approach learned by using several feature sets, consisting of lexical features, semantic features and a combination of lexical and semantic features. Our study reveals that semantic features, most notably BabelNet synsets and semantic frames, can be successfully applied for Sentiment Analysis within the financial domain to achieve better results. Moreover, a comparative study has been conducted between our supervised approach and unsupervised approaches. The obtained experimental results show how our approach outperforms the others. Keywords Sentiment analysis · Financial domain · Microblogs · News
1 Introduction User-generated data in blogs and social networks have recently become a valuable resource for mining user sentiments to the end of capturing the “pulse” of stock markets [1]. Therefore, Sentiment Analysis in the financial domain is becoming more and more a big concern for businesses, organizations and marketing researchers, mainly due to the high subjectivity of this content, as users express freely their opinions, contrary to news articles which are known by their objectivity and implicit opinions [2]. Both lexicon-based [1, 3] and machine learning methods [2, 4] have been used in Sentiment Analysis within the financial domain. Most of lexicon-based methods have focused on the coarse-grained analysis of sentiment expressed in the text. However, coarse-grained methods are insufficient for the detection and polarity classification of * Diego Reforgiato Recupero [email protected] Amna Dridi [email protected] Mattia Atzeni [email protected] 1
Department of Mathematics and Computer Science, University of Cagliari, Via Ospedale 72, 09124 Cagliari, Italy
sentiment expressed about companies in financial news text, as not all expressions of sentiment are related to the company of interest [5]. To tackle this problem, machine learning techniques have been recently proposed [2, 5, 6], mainly investigating a fine-grained schema to pinpoint the particular phrases which express sentiment and analyze these sentiment expressions in a fine-grained manner. Both approaches of research in Sentiment Analysis in the financial domain are still too much focused on word occurrence methods and they seldom even use WordNet [7], ignoring consequently advancements of techniques in semantics. However, semantics is crucial for text classification problems. From this perspective, this work lies at the intersection of NLP, Semantic Web and Sentiment Analysis which are recently being increasingly researched for many emergin
Data Loading...
