Company ethics index

Project goal

is to develop of a toolkit to assess the ethical reputation of Russian companies based on machine analysis of open text data from various sources: news feeds, company websites, social networks, annual and non-financial company reports.

Methods used:

Web-scraping software tools (web-scraping)

This technology allows using software code to automatically convert HTML pages into a more convenient data format for further analysis.
Natural language processing (NLP) methods

NLP methods are used for further processing of text data. These methods have been actively developed and used since the middle of the last century in various fields such as medicine, finance, education and business. However, the potential of using NLP methods to assess the ethics of companies and their leaders has not been fully realised.
Sentiment analysis of texts containing references to the company

This method allows to evaluate the emotional colouring of the lexicon used in the text, as well as to reveal the author's emotional evaluation of the text in relation to the described object.
In-depth learning models

In-depth learning models will be used to analyse consumers' textual feedback on the services of Russian companies.