Application of Machine Learning and Government Finance Statistics for macroeconomic signal mining to analyze recessionary trends and score policy effectiveness
2021 IEEE International Conference on Big Data (Big Data)
The budget speech is part of the democratic process that is presented annually to members of the parliament and addressed to Speaker of the house. Budget speech includes details of annual financial statements or financial plans of the government, containing details of revenue and expenditure in the past, along with the estimated spending and projections for the following year. Speech, additionally, consists of new policies and / or reforms announced to address fiscal macroeconomic issues. It takes, importantly, years to witness effectiveness of policies, especially in agriculture and infrastructure sectors. In this research paper, we propose an innovative Machine Learning framework that scores effectiveness of agricultural policies through binning language processing statements with key macroeconomic performance multiclass-multilabel-indicators that are regressed from government finance statistics and macroeconomic time series data. Finally, the paper presents budget speech prototype solution as well as its application for analyzing 2021 Indian budget speech.
Agricultural Economic Trends, Budget Speech, ensemble models, Government Finance Statistics, Macroeconomic Factors, Natural Language Processing
Chandrasekar Vuppalapati, Anitha Ilapakurti, Sandhya Vissapragada, Vanaja Mamaidi, Sharat Kedari, Raja Vuppalapati, Santosh Kedari, and Jaya Vuppalapati. "Application of Machine Learning and Government Finance Statistics for macroeconomic signal mining to analyze recessionary trends and score policy effectiveness" 2021 IEEE International Conference on Big Data (Big Data) (2021): 3274-3283. https://doi.org/10.1109/BigData52589.2021.9672025