WWW'18 Open Challenge: Financial Opinion Mining and Question Answering

By Macedo Maia, Siegfried Handschuh, André Freitas, Brian Davis, Ross McDermott, Manel Zarrouk, A. Balahur. The Web Conference, 2018

452 citations recorded by the source corpus.

The growing maturity of Natural Language Processing (NLP) techniques and resources is dramatically changing the landscape of many application domains which are dependent on the analysis of unstructured data at scale. The finance domain, with its reliance on the interpretation of multiple unstructured and structured data sources and its demand for fast and comprehensive decision making is already emerging as a primary ground for the experimentation of NLP, Web Mining and Information Retrieval (IR) techniques for the automatic analysis of financial news and opinions online. This challenge focuses on advancing the state-of-the-art of aspect-based sentiment analysis and opinion-based Question Answering for the financial domain.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/7191680b572ee7145f1a9d95ff11ab1ff44259f3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Precise Zero-Shot Dense Retrieval without Relevance Labels (2022, unknown): “Contriever FT shows performance advantages on FiQA and DBPedia.”
Promptagator: Few-shot Dense Retrieval From 8 Examples (2022, background): “, 2018) are long compositional questions, while queries in FiQA (Maia et al., 2018) are short financial questions.”
BloombergGPT: A Large Language Model for Finance (2023, background): “FiQA SA (Maia et al., 2018): The second sentiment analysis task is to predict the aspectspecific sentiment in English financial news and microblog headlines, which were published as a part of the 2018 challenge on financial question answering and opinion mining.”
How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection (2023, background): “1) medicine 1248 1248 1337 Medical Dialog dataset [6] finance 3933 3933 4503 FiQA dataset [23]”
NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models (2024, unknown): “…Language Inference (Group et al., 2022), SQuAD (Rajpurkar et al., 2016), ArguAna (Wachsmuth et al., 2018), BioASQ (Tsatsaronis et al., 2015), FiQA (Maia et al., 2018), FEVER (Thorne et al., 2018), HoVer (Jiang et al., 2020), SciFact (Wadden et al., 2022), NFCorpus, MIRACL (Zhang et al., 2023) and…”
BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Models (2021, unknown): “30 (QA) Finance FiQA-2018 [44] 7 Binary 14,166 500 648 57,638 2.”
Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, background): “7 FiQA ABSA (Maia et al., 2018) Financial news headlines: 529 samples; financial microblogs: 774 annotated posts JSON nodes with sentiment score ranged from -1 to 1, "target" indicates opinion target, and "aspect" indicates aspect categories according to different level "1": { "sentence": "Royal Mai”
Task-aware Retrieval with Instructions (2022, background): “, 2021), FIQA (Maia et al., 2018), NF Corpus (Boteva et al.”
Dense Text Retrieval Based on Pretrained Language Models: A Survey (2022, background): “Question answering Wikipedia Natural Questions [33] 152,148 152,148 6,515 3,610 2,681,468 Wikipedia SQuAD [55] 78,713 78,713 8,886 10,570 23,215 Wikipedia TriviaQA [56] 78,785 78,785 8,837 11,313 740K Wikipedia HotpotQA [57] 85,000 170,000 5,447 7,405 5,233,329 Web WebQuestions [58] 3,417 3,417 361 ”
Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer (2022, methodology): “We consider 2 QA datasets (BioASQ and FiQA (Tsatsaronis et al., 2015; Maia et al., 2018)) and one IR dataset (MS MARCO (Nguyen et al., 2016)) to evaluate supervised adaptation capability, and 4 other datasets (CQADupStack, TREC-COVID, SCIDOCS, Sci-Fact (Hoogeveen et al., 2015; Voorhees et al.,…”
ColBERTv2: Effective and Efficient Retrieval via Lightweight Late Interaction (2021, unknown): “Looking more closely into the comparison between SPLADEv2 and ColBERTv2, we see that ColBERTv2 has an advantage in six benchmarks and ties SPLADEv2 in two, with the largest improvements attained on NQ, FiQA-2018, and TREC-COVID, all of which feature natural search queries.”
Query Drift Compensation: Enabling Compatibility in Continual Learning of Retrieval Embedding Models (2025, unknown): “FiQA-2018 (Maia et al., 2018) How can a ’saver’ maintain or increase Personally, I invest in mutual funds.”
Beyond Matryoshka: Revisiting Sparse Coding for Adaptive Representation (2025, unknown): “For text retrieval, we select FiQA2018 (Maia et al., 2018), NFCorpus (Boteva et al., 2016) and SciFACT (Wadden et al., 2020).”