Towards a Framework for Acquisition and Analysis of Speeches to Identify Suspicious Contents through Machine Learning

Figure 5

Number of samples in each domain of suspicious class.