Paper: arXiv 2404.01337

Authors: Silvia García-Méndez, Francisco de Arriba-Pérez, Ana Barros-Vila, Francisco J. González-Castaño

Abstract

Finance-related news such as Bloomberg News, CNN Business and Forbes are valuable sources of real data for market screening systems. In news, an expert shares opinions beyond plain technical analyses that include context such as political, sociological and cultural factors. In the same text, the expert often discusses the performance of different assets. Some key statements are mere descriptions of past events while others are predictions. Therefore, understanding the temporality of the key statements in a text is essential to separate context information from valuable predictions. We propose a novel system to detect the temporality of finance-related news at discourse level that combines Natural Language Processing and Machine Learning techniques, and exploits sophisticated features such as syntactic and semantic dependencies. More specifically, we seek to extract the dominant tenses of the main statements, which may be either explicit or implicit. We have tested our system on a labelled dataset of finance-related news annotated by researchers with knowledge in the field. Experimental results reveal a high detection precision compared to an alternative rule-based baseline approach. Ultimately, this research contributes to the state-of-the-art of market screening by identifying predictive knowledge for financial decision making.

Complexity vs Empirical Score

  • Math Complexity: 3.0/10
  • Empirical Rigor: 5.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper uses NLP/ML methods but relies on a manually labeled dataset without statistical metrics or backtesting, placing it in Street Traders due to practical implementation over theoretical math.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Detect temporality in financial news<br>to separate context from predictions"] --> B["Methodology:<br>Combine NLP & Machine Learning<br>with syntactic & semantic features"]
  B --> C["Input Data:<br>Labelled dataset of<br>financial news articles"]
  C --> D["Computational Process:<br>Extract & analyze dominant tenses<br>of main statements at discourse level"]
  D --> E["Outcome:<br>High detection precision vs baseline<br>Identifies predictive knowledge<br>for market screening systems"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a novel system that detects the temporality of finance-related news at discourse level using NLP and machine learning with syntactic and semantic features.

Source sentence
We propose a novel system to detect the temporality of finance-related news at discourse level that combines Natural Language Processing and Machine Learning techniques, and exploits sophisticated features such as syntactic and semantic dependencies.

Assumes

No explicit assumptions found in the text.

Data

  • experimental dataset text-news public
    Source sentence
    Our experimental dataset6 is composed of s =600 news pieces with the following information: identifier, content, ticker (main stock market asset discussed), source and temporality (past and future tags of key opinions in the text).

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The Machine Learning approach was significantly better than the rule‑based baseline, achieving between 8% and 10% improvement in all metrics. performance
    Source sentence
    The Machine Learning approach was significantly better (between 8% and 10% improvement in all metrics).
  • The svc classifier improved over the rule‑based baseline by at least 8% for all metrics. performance
    Source sentence
    the improvement of the svc classifier in Table 9 over the rule‑based baseline was at least 8% for all metrics.
  • Machine Learning algorithms achieved precision and recall above 80% both globally and by class. performance
    Source sentence
    These features were essential for the Machine Learning algorithms to achieve precision and recall above 80%, both globally and by class.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 15 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (55)

  1. FineNews: fine-grained semantic sentiment analysis on financial microblogs and news (2019) — link
  2. Detection of Financial Opportunities in Micro-Blogging Data With a Stacked Classification System (2020) — link
  3. Comparison of individual, ensemble and integrated ensemble machine learning methods to predict China’s SME credit risk in supply chain finance (2017) — link
  4. Forecasting SMEs’ credit risk in supply chain finance with an enhanced hybrid ensemble machine learning approach (2019) — link
  5. Towards Auto-labelling Issue Reports for Pull-Based Software Development using Text Mining Approach (2019) — link
  6. Data Mining (2012) — link
  7. The Nature of Statistical Learning Theory (2000) — link
  8. Important Complexity Reduction of Random Forest in Multi-Classification Problem (2019) — link
and 47 more
  1. Data Mining: Practical Machine Learning Tools and Techniques (2016) — link
  2. Prediction of User Consumption Behavior Data Based on the Combined Model of TF-IDF and Logistic Regression (2021) — link
  3. Extending latent semantic analysis to manage its syntactic blindness (2021) — link
  4. A Summary of the COLIEE 2019 Competition (2020) — link
  5. A comparison of rule-based and machine learning approaches for classifying patient portal messages (2017) — link
  6. Can deep learning predict risky retail investors? A case study in financial risk behavior forecasting (2020) — link
  7. Machine learning for financial transaction classification across companies using character‐level word embeddings of text fields (2021) — link
  8. Enhancing deep learning sentiment analysis with ensemble techniques in social applications (2017) — link
  9. A two-step hybrid unsupervised model with attention mechanism for aspect extraction (2020) — link
  10. Trade the tweet: Social media text mining and sparse matrix factorization for stock market prediction (2016) — link
  11. Pre-processing online financial text for sentiment classification: A Natural Language Processing approach (2014) — link
  12. Coherence relations in a cognitive theory of discourse representation (1993) — link
  13. Processing narrative time shifts. (1996) — link
  14. The Expression of Temporality in Basilang Speech (1987) — link
  15. On temporality in discourse annotation: Theoretical and practical considerations (2017) — link
  16. Using Twitter to Predict the Stock Market (2015) — link
  17. Can Internet Search Queries Help to Predict Stock Market Volatility? (2016) — link
  18. Enhancing Sentiment Analysis of Financial News by Detecting Negation Scopes (2015) — link
  19. Coherence, reference, and the theory of grammar (2002)
  20. Temporal prepositions and temporal generalized quantifiers (2001) — link
  21. Quantifying the Relationship Between Financial News and the Stock Market (2013) — link
  22. Deep Learning for financial sentiment analysis on finance news providers (2016) — link
  23. Building a discourse-tagged corpus in the framework of Rhetorical Structure Theory (2001) — link
  24. Tense Aspect in Verbal Morphology (2003) — link
  25. Do financial blogs serve an infomediary role in capital markets? (2016) — link
  26. Can Facebook Predict Stock Market Activity? (2012) — link
  27. The penn discourse treebank 2.0 (2008)
  28. Time, temporality and consumer behaviour (1998) — link
  29. Natural language based financial forecasting: a survey (2018) — link
  30. Artefacts of Management Academe (2005) — link
  31. Natural Language Processing in Accounting, Auditing and Finance: A Synthesis of the Literature with a Roadmap for Future Research (2016) — link
  32. A Generative Model for category text generation (2018) — link
  33. Social media and its impact on consumers behavior (2014)
  34. Natural Language Processing (almost) from Scratch (2011) — link
  35. Trading strategies to exploit blog and news sentiment (2010)
  36. Unsupervised Neural Networks approach for understanding fraudulent financial reporting (2012) — link
  37. Financial news predicts stock market volatility better than close price (2018) — link
  38. Temporal prepositions and their logic (2005) — link
  39. Sentiment Analysis. Mining Opinions, Sentiments, and Emotions (2015) — link
  40. Stock returns and future tense language in 10-K reports (2016) — link
  41. Dynamic ensemble selection for multi-class imbalanced datasets (2018) — link
  42. Paths in L2 acquisition: The expression of temporality in spatially oriented narration (2012)
  43. Stock market forecasting with financial micro-blog based on sentiment and time series analysis (2017) — link
  44. Deep Learning for stock market prediction from financial news articles (2017) — link
  45. FreeLing 1.3: Syntactic and semantic services in an open-source NLP library (2006)
  46. Multilingual Central Repository version 3.0: upgrading a very large lexical knowledge base (2012)
  47. FreeLing 3.0: Towards Wider Multilinguality (2012)

0 of 55 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.