A Novel Sentiment Analysis Model of Museum User Experience Evaluation Data Based on Unbalanced Data Analysis Technology

Algorithm 1

A denoising algorithm.

	Input: training data X, the number of neighbors k;
	Output: SV area sample set Ssv, MNSV area sample set SNSV-maj after denoising, FNSV area sample set SNSV-min after denoising;
	Step 1: The training data X is used to construct the SVM classifier;
	Step 2: The data set X is divided into the sample set Ssv of the SV area after passing through the SVM classifier. The sample set XNSV-a in the MNSV area and the sample set XNSV-b in the FNSV area.
	Step 3: Find all the minority class samples a1 in XNSV-a; for each sample in a1, find the k samples closest to it. If all the k samples belong to the majority class, delete the sample, otherwise do not do it Process to get the processed sample set XNSV-maj;
	Step 4: Delete the majority class samples in XNSV-b to get the sample set XNSV-min.