The aim of the Hackathon is to aid students to apply their academic knowledge within both a commercial and technically secure environment, enabling students to put themselves in the shoes of the professional working in financial and commercial fields.
- News Headline Data (Last 15 Months)
- Decide the companies or industries (FTSE100, Fortune100, DAX40)
- Pull out ESG Controversy Data 2022 - 2023
- Pull out News Headline Data 2022 - 2023
- Enhance the ESG News Filter (Gaussian vs Exact Glossaries)
- Fin-BERT vector representations for News.
- Ensemble Model or Severity Rate (Simple Math) on top of Fin-BERT to perform regression
FTSE100 listed companies link
There is a nice article by Jason Ramchadani talking about how to set proper query to pull out news headline data from Refinitiv EIKON, you can access here.
Using Refinitiv API is quite challenging because there is not enough documentation for that and also there are a lot of customisation. For retrieving the headline news we can use get_news_headline
function with these arguments you can pass on:
-
query: string, optional. Search criteria for the news headlines The query can contain RIC codes, company names, country names and operators (AND, OR, NOT, IN, parentheses and quotes for explicit search…). Tip: Append 'R:' in front of RIC names to improve performance. Default: Top News written in English
-
count: int, optional. The maximum number of headlines to retrieve. Value Range: [1-100]. Default: 10
-
date_from: string. Beginning of date range. String format is:'%Y-%m-%dT%H:%M:%S'. e.g. 2016-01-20T15:04:05.
-
date_to: string. End of date range. String format is: '%Y-%m-%dT%H:%M:%S'. e.g. "2016-01-20T15:04:05".
-
raw_output: boolean. Set this parameter to True to get the data in json format if set to False, the function will return a data frame The default value is False
-
debug: boolean. If this parameter is set to True, the json request and response are printed.
To construct well writen query for the news headline we can head up to news monitor write your criteria on the search bar, copy and paste the result to your jupyter notebook.
Team member: Radhea, Maia, Yogi