Natural Language Processing (NLP) Analysis of the Bank of Algeria's Communiques

This research applies text mining and natural language processing (NLP) methods to examine the communicative tactics of the Bank of Algeria through its annual reports from 2002 to 2022. Techniques such as Latent Dirichlet Allocation (LDA) for topic modeling and sentiment analysis with the Loughran-McDonald dictionary were employed to identify main themes and emotional tones. The remarkable changes in the bank's communication indicate the effects of oil price shocks, the global financial crisis, and the COVID-19 pandemic. Initially, this was about economic expansion and monetary policy, but later it became focused on inflation targeting as well as financial stability in times of economic turbulence. Negative sentiment during this period indicates a cautious approach according to sentiment analysis results. Readability tests using the Coleman-Liau readability index also suggest that although the texts are complicated, they remain clear for efficient communication with both the public and employees operating in the marketplace. This is an area where literature lacks because no one has ever studied how Bank of Algeria reports can be analyzed utilizing these advanced techniques in text mining and NLP, as far as we have been able to find so far.