Twitter Sentiment Tracking for Predicting Marketing Trends - Smart Information Systems: Computational Intelligence for Real-Life Applications - page 66

Fig. 2.10 Flowchart of the web interface
dataset is used, and the number of data in the training set is increased continuously.
Figure 2.11 shows the accuracy and the precision of classification methods.
As it can be observed from the Fig. 2.11 , the precision of the Maximum Entropy
reaches its optimum level at the point where the number of training data indicates
650. It reveals almost no change from 550 training data to 100 training data.
On the other hand, the accuracy level of the Maximum Entropy does not demon-
strate a considerable change. It reaches its peak level at 700 training data point.