Here is a problem about neural networks. I want get solutions as detailed as possible, thank you!!!!

Question 6 (4+4+5+5-18 marks marks) Solve the following practical problem A company wants to develop a music classification by genre using neural networks. Large amount of music songs data set has been collected and the data set contains features from symbolic songs (MP3, in this case), which can be used to classify the recordings by geure. Each example is classified as classic, rock, jazz or folk song. The attributes are duration of song, tempo, root mean square (RMS) amplitude, sampling frequency, sampling rate, dynamic range, tonality and number of digital errors. 1 Describe the neural network model for the classification task, and the possible split of the original data set into the training, validating and the testing data sets. 4 marks 2 How would you normalize the data? [4 marks] 3 If the performance of your model on the training set is good, but the test set performance is significantly worse, what is the most possible problem? Discuss a simple way that may improve your results. 5 marks] Describe and explain two different measurements for the classification performance that you would use. Discuss any possible relationships between the two measure- ments. 5 marks]