Paper: arXiv 2404.01338

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

Abstract

Financial news items are unstructured sources of information that can be mined to extract knowledge for market screening applications. Manual extraction of relevant information from the continuous stream of finance-related news is cumbersome and beyond the skills of many investors, who, at most, can follow a few sources and authors. Accordingly, we focus on the analysis of financial news to identify relevant text and, within that text, forecasts and predictions. We propose a novel Natural Language Processing (NLP) system to assist investors in the detection of relevant financial events in unstructured textual sources by considering both relevance and temporality at the discursive level. Firstly, we segment the text to group together closely related text. Secondly, we apply co-reference resolution to discover internal dependencies within segments. Finally, we perform relevant topic modelling with Latent Dirichlet Allocation (LDA) to separate relevant from less relevant text and then analyse the relevant text using a Machine Learning-oriented temporal approach to identify predictions and speculative statements. We created an experimental data set composed of 2,158 financial news items that were manually labelled by NLP researchers to evaluate our solution. The ROUGE-L values for the identification of relevant text and predictions/forecasts were 0.662 and 0.982, respectively. To our knowledge, this is the first work to jointly consider relevance and temporality at the discursive level. It contributes to the transfer of human associative discourse capabilities to expert systems through the combination of multi-paragraph topic segmentation and co-reference resolution to separate author expression patterns, topic modelling with LDA to detect relevant text, and discursive temporality analysis to identify forecasts and predictions within this text.

Complexity vs Empirical Score

  • Math Complexity: 3.5/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Street Traders — empirical and applied, lower math complexity

Why this score: The paper focuses on implementing an NLP pipeline with established models like LDA and machine learning, resulting in moderate empirical rigor due to the detailed experimental setup and evaluation metrics (ROUGE-L, dataset size). Math complexity is low, as it relies on standard probabilistic topic modeling and classification techniques without novel mathematical derivations.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Automatic detection of relevant info,<br>predictions, and forecasts in financial news"] --> B["Data Input<br>2,158 financial news items<br>(manually labeled)"]

  B --> C{"Methodology Pipeline"}

  C --> D["Step 1: Topic Segmentation<br>Group closely related text"]
  D --> E["Step 2: Co-reference Resolution<br>Discover internal text dependencies"]
  E --> F["Step 3: LDA Topic Modelling<br>Detect relevant vs. less relevant text"]
  F --> G["Step 4: Temporal Analysis<br>Identify predictions & forecasts<br>within relevant text"]

  G --> H["Key Outcomes<br>ROUGE-L Score: 0.662 (Relevance)<br>ROUGE-L Score: 0.982 (Predictions)<br>First work to jointly consider relevance & temporality"]

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 NLP system that segments financial news, applies co‑reference resolution, uses LDA topic modelling to detect relevant text, and performs discursive temporality analysis to identify forecasts and predictions.

Source sentence
We propose a novel Natural Language Processing (NLP) system to assist investors in the detection of relevant financial events in unstructured textual sources by considering both relevance and temporality at the discursive level.

Assumes 1

  • There is a direct causal relationship between financial news and asset prices. market
    Source sentence
    Assuming there is a direct causal relationship between financial news and asset prices [17]

Data

  • Experimental data set of financial news items · Popular financial websites (e.g., Benzinga, Motley Fool, Mar text-news unknown
    1st October 2018 to 1st October 2020 · News items mentioning at least one stock in the authors' financial lexica
    Source sentence
    Our experimental data set was composed of 2,158 news pieces (average length of 27.98 sentences and 537.24 words). As previously mentioned, the pieces were automatically extracted with a script from popular and prestigious financial websites between 1st October 2018 and 1st October 2020.

Recognised sources: Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The system achieved ROUGE‑L scores of 0.662 for relevant‑text detection and 0.982 for prediction/forecast detection. performance
    Source sentence
    The rouge-l values for the identification of relevant text and predictions/forecasts were 0.662 and 0.982, respectively.
  • The proposed NLP system outperformed a rule‑based baseline system for extracting relevance and temporality from financial news. performance
    Source sentence
    Our solution outperformed a rule‑based baseline system.
  • Inter‑annotator agreement showed mean Alpha‑reliability of 0.552 and accuracy of 0.861, indicating acceptable annotation quality. statistical
    Source sentence
    The mean values were 0.552 and 0.861, respectively. Previous works have considered an Alpha‑reliability value above 0.41 to be acceptable.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 13 related papers). Coverage is partial, so this is neither replication nor refutation.

Cited by 1 paper in the archive (1 by independent authors)

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 (71)

  1. Pre-trained language models to extract information from radiological reports (2021)
  2. Seeing the wood for the trees: How machine learning can help firms in identifying relevant electronic word-of-mouth in social media (2019) — link
  3. Text analytics approach to extract course improvement suggestions from students’ feedback (2018) — link
  4. Web opinion mining (2008) — link
  5. Deep Learning for Stock Market Prediction Using Event Embedding and Technical Indicators (2018) — link
  6. Event detection in finance using hierarchical clustering algorithms on news and tweets (2021) — link
  7. Economic Event Detection in Company-Specific News Text (2018) — link
  8. Improving text summarization using neuro-fuzzy approach (2017) — link
and 63 more
  1. Information Extraction. A survey (2005)
  2. An overview of graph-based keyword extraction methods and approaches (2015)
  3. LDA Meets Word2Vec (2018) — link
  4. A multi-layer and multi-ensemble stock trader using deep learning and deep reinforcement learning (2021) — link
  5. Incomplete multi-view subspace clustering with adaptive instance-sample mapping and deep feature fusion (2021) — link
  6. An optimal deep learning-based LSTM for stock price prediction using twitter sentiment analysis (2022) — link
  7. MKPM: Multi keyword-pair matching for natural language sentences (2022) — link
  8. A new efficient decision making algorithm based on interval-valued fuzzy soft set (2020) — link
  9. Detection of temporality at discourse level on financial news by combining Natural Language Processing and Machine Learning (2022) — link
  10. A new graphic kernel method of stock price trend prediction based on financial news semantic and structural similarity (2019) — link
  11. Investment strategies applied to the Brazilian stock market: A methodology based on Sentiment Analysis with deep learning (2021) — link
  12. Partially collapsed Gibbs sampling for latent Dirichlet allocation (2019) — link
  13. PAN-LDA: A latent Dirichlet allocation based novel feature extraction model for COVID-19 data using machine learning (2021) — link
  14. An overview of online fake news: Characterization, detection, and discussion (2020) — link
  15. Enhancing Aspect-Based Sentiment Analysis of Arabic Hotels’ reviews using morphological, syntactic and semantic features (2019) — link
  16. Karcı summarization: A simple and effective approach for automatic text summarization using Karcı entropy (2020) — link
  17. Extractive Multi-Document Arabic Text Summarization Using Evolutionary Multi-Objective Optimization With K-Medoid Clustering (2020) — link
  18. Extractive multi-document text summarization using a multi-objective artificial bee colony optimization approach (2018) — link
  19. Improving Coreference Resolution by Learning Entity-Level Distributed Representations (2016) — link
  20. Extractive summarization of clinical trial descriptions (2019) — link
  21. Toward assessing clinical trial publications for reporting transparency (2021) — link
  22. Online Hate Ratings Vary by Extremes (2019) — link
  23. Development and accuracy of an artificial intelligence algorithm for acne grading from smartphone photographs (2019) — link
  24. Assessing the efficacy of a manual‐based intervention for improving the detection of facial pain expression (2019) — link
  25. Content Analysis: An Introduction to its Methodology (2018) — link
  26. Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey (2019) — link
  27. Improvement of Text Segmentation TextTiling Algorithm (2020) — link
  28. Detection of Financial Opportunities in Micro-Blogging Data With a Stacked Classification System (2020) — link
  29. Identifying Banking Transaction Descriptions via Support Vector Machine Short-Text Classification Based on a Specialized Labelled Corpus (2020) — link
  30. A systematic review of fundamental and technical analysis of stock market predictions (2020) — link
  31. NATSUM: Narrative abstractive summarization through cross-document timeline generation (2019) — link
  32. A Generative Model for category text generation (2018) — link
  33. Combined deep learning classifiers for stock market prediction: integrating stock price and news sentiments (2021) — link
  34. Stock market prediction using machine learning classifiers and social media, news (2022) — link
  35. A review of unsupervised feature selection methods (2020) — link
  36. COVID-19 Future Forecasting Using Supervised Machine Learning Models (2020) — link
  37. Whose story is it anyway? Automatic extraction of accounts from news articles (2019) — link
  38. On temporality in discourse annotation: Theoretical and practical considerations (2017) — link
  39. SRL-ESA-TextSum: A text summarization approach based on semantic role labeling and explicit semantic analysis (2019) — link
  40. Natural language based financial forecasting: a survey (2018) — link
  41. The evolution of argumentation mining: From models to social media and emerging tools (2019) — link
  42. Asymmetric dependence between stock market returns and news during COVID-19 financial turmoil (2020) — link
  43. AUSS: An Arabic query-based update-summarization system (2020) — link
  44. Abstractive summarization: An overview of the state of the art (2019) — link
  45. Linguistic summarization of fuzzy social and economic networks: an application on the international trade network (2020) — link
  46. Statistical Features for Extractive Automatic Text Summarization (2020) — link
  47. Template Based Medical Reports Summarization (2018) — link
  48. Framing overtourism: a critical news media analysis (2020) — link
  49. Hyping health effects: a news analysis of the ‘new smoking’ and the role of sitting (2019) — link
  50. Fake News Analysis Modeling Using Quote Retweet (2019) — link
  51. NEWS Analysis Towards Youth Financial Competency Management (2018) — link
  52. Financial news predicts stock market volatility better than close price (2018) — link
  53. Estimating the impact of domain-specific news sentiment on financial assets (2018) — link
  54. Deep Learning Based Part-of-Speech Tagging for Malayalam Twitter Data (Special Issue: Deep Learning Techniques for Natural Language Processing) (2019) — link
  55. Unmasking text plagiarism using syntactic-semantic based natural language processing techniques: Comparisons, analysis and challenges (2018) — link
  56. Applying Natural Language Processing and Hierarchical Machine Learning Approaches to Text Difficulty Classification (2020) — link
  57. Augmenting Qualitative Text Analysis with Natural Language Processing: Methodological Study (2018) — link
  58. Construction site accident analysis using text mining and natural language processing techniques (2019) — link
  59. Speech Technology Progress Based on New Machine Learning Paradigm (2019) — link
  60. A new architecture of Internet of Things and big data ecosystem for secured smart healthcare monitoring and alerting system (2018) — link
  61. An Intelligent Anomaly Detection Scheme for Micro-Services Architectures With Temporal and Spatial Data Analysis (2020) — link
  62. Textual Analysis in Accounting and Finance: A Survey (2016) — link
  63. Predicting sentence-level polarity labels of financial news using abnormal stock returns (2020) — link

0 of 71 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.