The Predictability of Tree-based Machine Learning Algorithms in the Big Data Context

This research work is concerned with the predictability of ensemble and singular tree-based machine learning algorithms during recession prosperity two companies listed in Tehran Stock Exchange context big data. In this regard, main issue that economic managers academic community require predicting models more accuracy reduced execution time; moreover, prediction stock market highly significant. Machine must be able to appropriately predict return sign downturn boom days. Addressing stated challenge will upgrade quality purchases and, subsequently, increase profitability. article, proposed solution relies on utilization The exploits decision tree algorithm, which a traditional algorithm. Furthermore, modern algorithms, random forest gradient boosted tree, has been utilized for prosperity. mentioned cases were implemented by applying tools python programming language PYSPARK library used explicitly data context. current study are shares information Exchange. obtained results reveal applied have performed better than algorithms. Additionally, adding 23 technical features initial subsequent PCA feature reduction method demonstrated best performance among other modes. meantime, it concluded do not possess proper resolution or generalizability, either recession.