Application and comparison of machine learning techniques in business

Data science methodology and, particularly, machine learning techniques, are being widely used in many different fields today. There is the same trend in the private sector, where using machine learning is highly advantageous. The prediction of key variables in business given multiple input parameters is important and may affect the firm's profitability. This research aims to investigate several problems in business, being churn estimation, housing price prediction and sentiment analysis, which require a certain data framework and an algorithmic approach. The applied algorithms are compared for each case. The regarded cases involve an application of classification, regression, and text analysis types of machine learning models, demonstrating their diversity and useful application in the industry.