{"id":1144,"date":"2023-06-01T11:02:46","date_gmt":"2023-06-01T11:02:46","guid":{"rendered":"https:\/\/wp.lancs.ac.uk\/cfie\/?page_id=1144"},"modified":"2023-10-25T07:17:35","modified_gmt":"2023-10-25T07:17:35","slug":"fincausal2023","status":"publish","type":"page","link":"https:\/\/wp.lancs.ac.uk\/cfie\/fincausal2023\/","title":{"rendered":"FinCausal 2023"},"content":{"rendered":"

FinCausal-2023 Shared Task: “<\/strong>Financial Document Causality Detection”<\/strong><\/span><\/h1>\n

To be held at The 5th Financial Narrative Processing Workshop (FNP 2023)<\/a>, Sorrento, Italy, \u00a015-18 December 2023<\/strong>.<\/p>\n

New<\/strong><\/span><\/span>:<\/strong><\/span> Access the CodaLab by clicking here<\/a><\/p>\n

\n

Important Dates:<\/span><\/span><\/strong><\/p>\n

    \n
  • Call for participation and registration: 3rd June 2023<\/li>\n
  • Registration deadline: 28 June<\/li>\n
  • Training set release: 14 July 2023<\/li>\n
  • Test set release:\u00a05 September 2023<\/span>\u00a011 September 2023<\/li>\n
  • Systems submission deadline:\u00a015 September 2023<\/span>\u00a019 September 2023<\/span> \u00a0–>\u00a0<\/strong>29 September 2023<\/strong><\/li>\n
  • Release of results:\u00a020 September 2023<\/span>\u00a025 September 2023<\/span> –>\u00a0<\/strong>2 October 2023<\/strong><\/li>\n
  • Paper submission deadline: October 30<\/span><\/b>, 2023<\/span> (anywhere in the world).<\/li>\n
  • Notification of acceptance: November 12, 2023<\/li>\n
  • Camera-ready of accepted papers: November 20, 2023<\/li>\n
  • Financial Narrative Processing Workshop: December 2023<\/li>\n<\/ul>\n
    \n

    Introduction<\/span><\/span><\/strong><\/h2>\n

    Financial analysis needs factual data and an explanation of the variability of these data. Data state facts but need more knowledge regarding how these facts materialised. Furthermore, understanding causality is crucial in studying decision-making processes.\u202f<\/span>\u00a0<\/span><\/p>\n

    The Financial Document Causality Detection Task aims to develop an ability to explain, from external sources, why a transformation occurs in the financial landscape as a preamble to generating accurate and meaningful financial narrative summaries. Its goal is to evaluate which events or chain of events can cause a financial object to be modified or an event to occur, regarding a given context.\u202f<\/span>\u00a0<\/span><\/p>\n

    For this task, we propose <\/span>two subtasks<\/span><\/b>, one in <\/span>English<\/span><\/b> and one in <\/span>Spanish<\/span><\/b>. In both of them, participants will be asked to identify, in causal sentences, which elements of the sentence relate to the cause, and which relate to the effect.<\/span>\u00a0<\/span><\/p>\n

    In the <\/span>English<\/span><\/b> subtask, the dataset has been sourced from various 2019 financial news articles provided by Qwam, along with additional SEC data from the Edgar Database. Additionally, we have augmented the dataset from FinCausal 2022, adding 500 new segments.<\/span>\u00a0<\/span><\/p>\n

    In the case of <\/span>Spanish<\/span><\/b>, t<\/span>he task dataset has been extracted from a corpus of Spanish financial annual reports from 2014 to 2018. This is the first year where we introduce a subtask in Spanish.<\/span>\u00a0<\/span><\/p>\n

    However, there will be some <\/span>differences<\/span><\/b> between the two subtasks:<\/span>\u00a0<\/span><\/p>\n

      \n
    1. Regarding the focus of the task<\/span><\/b>. In English it centers on detecting causes and effects when the effects are quantified. Conversely, the Spanish task aims to detect all types of causes and effects, not necessarily limited to quantified effects.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n
        \n
      1. Regarding the text segments<\/span><\/b>. The English segments are made up of up to three sentences, while the Spanish task will involve complete paragraphs.<\/span>\u00a0<\/span><\/li>\n<\/ol>\n

        Despite the differences, it is important to note that the dataset will be presented in the same format for both tasks.<\/span>\u00a0<\/span><\/p>\n

        English Subtask<\/strong><\/h2>\n

        This shared task focuses on determining causality associated with a quantified fact. An event is defined as the arising or emergence of a new object or context regarding a previous situation. So, the task will emphasise the detection of causality associated with the transformation of financial objects embedded in quantified facts. Participants will be provided with a sample of text blocks extracted from financial news, labelled through inter-annotator agreement.<\/span>\u00a0<\/span><\/p>\n

        Cause and Effect Detection<\/b>\u00a0<\/span><\/p>\n

        This task is a relation detection task. The aim is to identify, in a causal sentence or text block, the causal elements and the consequential ones. <\/span>Only one causal element and one effect are expected in each segment.<\/span><\/p>\n\n\n\n\n\n
        Text<\/strong><\/td>\nCause<\/strong><\/td>\nEffect<\/strong><\/td>\n<\/tr>\n
        Boussard Gavaudan Investment Management LLP bought a new position in shares of GENFIT S A\/ADR in the second quarter worth about $199,000. Morgan Stanley increased its stake in shares of GENFIT S A\/ADR by 24.4% in the second quarter.Morgan Stanley now owns 10,700 shares of the company\u2019s stock worth $211,000 after purchasing an additional 2,100 shares during the period<\/td>\nMorgan Stanley increased its stake in shares of GENFIT S A\/ADR by 24.4% in the second quarter<\/td>\nMorgan Stanley now owns 10,700 shares of the company\u2019s stock worth $211,000 after purchasing an additional 2,100 shares during the period.<\/td>\n<\/tr>\n
        Zhao found himself 60 million yuan indebted after losing 9,000 BTC in a single day (February 10, 2014)<\/td>\nlosing 9,000 BTC in a single day (February 10, 2014)<\/td>\nZhao found himself 60 million yuan indebted<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

        Table 1: Cause and Effect Detection Sample for English<\/p>\n

        Spanish Subtask<\/strong><\/h2>\n

        This shared task focuses on determining causality associated with both events or quantified facts. <\/span>For this task, a cause can be the justification for a statement or the reason that explains a result. <\/span>Participants will be provided with a sample of paragraphs extracted from Spanish financial annual reports, labelled through inter-annotator agreement.\u202f<\/span>\u00a0<\/span><\/p>\n

        Cause and Effect Detection<\/b><\/span><\/p>\n

        This task is also a relation detection task. The aim is to identify, in a paragraph, the causal elements and the consequential ones. Only one causal element and one effect are expected in each paragraph.<\/span>\u00a0<\/span><\/p>\n\n\n\n\n\n
        Text<\/strong><\/td>\nCause<\/strong><\/td>\nEffect<\/strong><\/td>\n<\/tr>\n
        El deterioro de activos financieros se increment\u00f3 en un 267,4%, debido a ciertos impactos negativos de la cartera de clientes mayoristas y a la actualizaci\u00f3n del escenario macroecon\u00f3mico.<\/span><\/span><\/td>\nd<\/span>ebido a ciertos impactos negativos de la cartera de clientes mayoristas y a la actualizaci\u00f3n del escenario macroecon\u00f3mico<\/span><\/span><\/td>\nEl deterioro de activos financieros se increment\u00f3 en un 267,4%<\/span><\/span><\/td>\n<\/tr>\n
        El resultado atribuido creci\u00f3 un 10,8% hasta los 1.522 millones de euros<\/span>, <\/span>gracias al buen comportamiento de las comisiones y sobre todo a la significativa reducci\u00f3n de los gastos y a los menores saneamientos y provisiones.<\/span><\/td>\ngracias al buen comportamiento de las comisiones y sobre todo a la significativa reducci\u00f3n de los gastos y a los menores saneamientos y provisiones<\/span><\/span>\u00a0<\/span><\/td>\nEl resultado atribuido creci\u00f3 un 10,8% hasta los 1.522 millones de euros<\/span><\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

        Table 2: Cause and Effect Detection Sample for Spanish<\/p>\n

        For the English and Spanish tasks, participants<\/span> can use any method they see fit (regex, corpus linguistics, entity relationship models, deep learning methods) to <\/span>identify<\/span> the causes and effects.\u202f<\/span><\/span><\/p>\n

        \n

        FinCausal Shared Task Organisers<\/span><\/span><\/strong><\/h2>\n