Build a Deep Learning Neural Network using Keras to generate Word2Vec vectors for the given corpus. First compute a Skip-Ngram dataset of the corpus. Then generate 1-hot encoded data for the input and output data generated by Ski-Ngram for a window size of 2. Optimize the Neural Network weights using the backpropagation algorithm. The hidden layer of your Neural Network should have 2 neurons. Display the Word2Vec vectors in a 2- dimensional cartesian coordinates plot. Plot of the vectors will show that the words which are close to each other in corpus will be close in the plot. Corpus: I like deep learning. I like NLP. I enjoy flying. Neural networks are one of the core parts of deep learning. For a given neural network, the forward propagation model is given as z(W, W) Wx1 + W2x2, 1 (z) = 1+e f(y,) -y log - (1 y) log(1-), where w, W are weight variables, x1, x2 and y are observations in R. (a) Determine the partial derivative and simplify the result. (b) Compute the partial derivative af z (v.) af -(y,) aw and simplify the result. (10 marks) (10 marks)