Complaint and Severity Identification From Online Financial Content

Abstract

The automatic detection of financial complaints can benefit businesses and online merchants. Compared to manually tagged complaints, they can use this information to monitor and address issues and effectively route them to appropriate teams. This can also promote greater transparency and accountability when dealing with consumer financial products and services, strengthening the firm's brand value. In linguistic studies, complaints have been classified into severity categories based on the level of risk the complainant is prepared to accept. Furthermore, since emotions influence every speech act, an individual's emotional state considerably impacts the complaint expression. In this paper, we introduce a Financial Complaints resource, a collection of annotated complaints arising between financial institutions and consumers expressed in English on Twitter. The dataset has been enriched with the associated emotion, sentiment, and complaint severity classes. The dataset comprises 3149 complaint and 3133 non-complaint instances spanning over ten domains (e.g., credit cards, mortgages, etc.). For a comprehensive evaluation of our dataset, we develop a multi-task framework for complaint detection and severity classification with emotion recognition and sentiment classification as the additional tasks and compare it with several existing baselines 1 .

Shortchanged: Uncovering and Analyzing Intimate Partner Financial Abuse in Consumer Complaints

“…The subsequent use of our workflow intends to mitigate this issue, but still relies on the results of proximity-based keyword-based approaches as a reference set which may limit how new patterns of IPFA are discovered. Like other HCI studies based on self-reported data, complainants may overstate certain aspects of their financial history due to social desirability bias [91] or to elicit empathy and alleviate financial burdens [49]. People with negative experiences may be more inclined to write formal complaints [59], while others may fear repercussions or societal stigma [44,90].…”

Reimagining Complaint Analysis: Adopting Seq2Path for a Generative Text-to-Text Framework

“…This necessitates a more efficient complaint detection approach, as exhibited in recent studies (Preotiuc-Pietro et al, 2019;Singh et al, 2022a) that have successfully automated the classification of binary complaints and their associated severity levels. Emotion recognition (ER) and sentiment recognition (SR) play vital roles in understanding the affective aspects of customer complaints Singh et al, 2023a). In this transition towards automation, inspiration can be drawn from multitask learning (Caruana, 1997)-an approach that mirrors our inherent human ability to learn multiple tasks simultaneously and transfer knowledge across them.…”