How can I improve the accuracy of my classifier? I tried to remove one of the variables (column 6) but it doesn't help.

i have a classifier and after running it the accuracy decreased to 50% from 94% can someone help me to increase the accuracy for future predicitions? Class A is Class X and so on. I want the classifier to confirm this as well. pLease help me to fix this.

%%Training code created April 2023 by M.C.
% Read in table ta from sheet 'ClassA'.
ta = readtable('CLASSES_trainingset.xlsx', 'Sheet', 'ClassA');
% Take columns 3 to 5 and 7 to 12
ta = ta(:, [3:5, 7:12]);
% Read in table tb from sheet 'ClassB'.
tb = readtable('CLASSES_trainingset.xlsx', 'Sheet', 'ClassB');
% Take columns 3 to 5 and 7 to 12
tb = tb(:, [3:5, 7:12]);
% Combine the tables
tPredictors = [ta; tb];
% Create a vector that has the true classes. Class 1 for the top rows (A) and 2 for the lower rows (B)
trueClasses = [ones(height(ta), 1); 2 * ones(height(tb), 1)];
% Setup a session
classificationLearner(tPredictors, trueClasses);
%% Saving Model
save Trained_Model trainedModel051823

Here is the code for the testset

% Load Trained_Model
load Trained_Model;
% Read in table td from sheet sheet1
td = readtable('Classes_testset1.xlsx', 'Sheet', 'ClassX');
% Take just Columns 1 to the end
td = td(:, [3:5, 7:12]); 
% Read in table tb from sheet 'ClassB'.
tf = readtable('Classes_testset1.xlsx', 'Sheet', 'ClassY');
% Take just Columns 3 to the end
tf = tf(:, [3:5, 7:12]);
% Combine test sets
Ptests = [td; tf];
% Use model to predict classes
labels = trainedModel051823.predictFcn(Ptests);
% Compare predicted classes to actual classes
trueClasses = [ones(height(td),1); 2 * ones(height(tf),1)];
% Create confusion matrix
figure;
cMat = confusionchart(trueClasses,labels);
% Set confusion matrix title
cMat.Title = 'Confusion Matrix for Classes Test Set';
% Set confusion matrix axes labels
cMat.XLabel = 'Predicted Class';
cMat.YLabel = 'True Class';
% Calculate accuracy
Accuracy = sum(trueClasses == labels) / numel(trueClasses) * 100;
% Determine which class the test set belongs to
if isequal(labels, ones(size(labels)))
    disp('This test set belongs to ClassX');
    figure; imshow(Ptests);
elseif isequal(labels, 2 * ones(size(labels)))
    disp('This test set belongs to ClassY');
    figure; imshow(Ptests);
else
    disp('This test set does not belong to either Western or Control diet, try fasted');
end
% Display results
Results = table(trueClasses, labels, Ptests);

Thanks!