A Fine-Grained Annotated Corpus for Target-Based Opinion Analysis of Economic and Financial Narratives

Patrick Paroubek, Jiahui Sophie HU

Abstract:

In this paper about aspect-based sentiment
analysis (ABSA), we present the first version
of a fine-grained annotated corpus for target-
based opinion analysis (TBOA) to analyze eco-
nomic activities or financial markets. We have
annotated, at an intra-sentential level, a corpus
of sentences extracted from documents repre-
sentative of financial analysts’ most-read mate-
rials by considering how financial actors com-
municate about the evolution of event trends
and analyze related publications (news, offi-
cial communications, etc.). Since we focus on
identifying the expressions of opinions related
to the economy and financial markets, we an-
notated the sentences that contain at least one
subjective expression about a domain-specific
term. Candidate sentences for annotations
were randomly chosen from texts of special-
ized press and professional information chan-
nels over a period ranging from 1986 to 2021.
Our annotation scheme relies on various lin-
guistic markers like domain-specific vocabu-
lary, syntactic structures, and rhetorical rela-
tions to explicitly describe the author’s sub-
jective stance. We investigated and evaluated
the recourse to automatic pre-annotation with
existing natural language processing technolo-
gies to alleviate the annotation workload. Our
aim is to propose a corpus usable on the one
hand as training material for the automatic de-
tection of the opinions expressed on an exten-
sive range of domain-specific aspects and on
the other hand as a gold standard for evalua-
tion TBOA.
In this paper, we present our pre-annotation
models and evaluations of their performance,
introduce our annotation scheme and report on
the main characteristics of our corpus.