In Python.

(a) Recall Manhattan and Euclidean distance calculation by passing by function as discussed in Exercise 4.7. Digital camera images can be formatted in JFIF files with the extension jpg or jpeg. Let's consider photos, each of which can be described over the features such as color in red (0-255), green (0-255) and blue (0-255), the geocode of shooting location in latitude (-90 - 90), longitude (-180 - 180) and altitude in any feet). With those features, we can compute the distances of any pairs. The sample images are computed by the three caller statements (which are in the last three line) below. nPoints = [(115, 115, 115,-25, 110, 10), (125, 15, 115,-25, 70, 20), (115, 55, 115, 25, -110, 30), (15, 115, 115,-25, 70, 40), (115, 55, 125,-25, 110,50), (115, 115, 115, -25, -110, 60), (115, 115, 55, 25, 70, 70), (115, 115, 15, -25, 110, 80). (55, 125, 115, 25, 70, 90), (15, 115, 115,-25, -110, 100)] dist(manhattanDist, nPoints) dist(euclideanDist, nPoints) distinDim Dist, nPoints) Note that the function dist() takes yet another function, nDimDist() as argument. Practice this high order functional programming skill as illustrated in the textbook. The sample output for 5 images, for simplicity, is: ABCDE A = (115, 115, 115,-25, 110, 10) 0.00 108.63 234.31 111.80 72.80 B = (125, 15, 115,-25, 70, 20) 108.63 0.00 191.57 150.00 65.57 C = (115, 55, 115, 25, -110, 30) 234.31 191.57 0.00 220.45 226.72 D = (15, 115, 115, -25, 70, 40) 111.80 150.00 220.45 0.00 124.10 E = (115, 55, 125,-25, 110, 50) 72.80 65.57 226.72 124.10 0.00 Recall HW3. First eliminate the stop words from the word list for each document. (b) Write a Python code to find a pair of the nearest similar documents. Hint: use Jaccard similarity of document pairs. (C) Improve your code developed above. Explain how the code is improved and how much. A sample run for (a): 0.062 ['doc4', 'doc6'] 0.075 ['doc1', 'doc6'] 0.096 ['doc1', 'doc2'] 0.105 ['doc4', 'doc5'] 0.107 ['doc2', 'doc6'] 0.108 ['doc2', 'doc5'] 0.112 ['doc1', 'doc5'] 0.116 ['doc3', 'doc4'] 0.118 ['doc2', 'doc3'] 0.119 ['doc3', 'doc6'] 0.122 ['doc1', 'doc3'] 0.134 ['doc2', 'doc4'] 0.137 ['doc1', 'doc4'] 0.161 ['doc5', 'doc6'] 0.163 ['doc3', 'doc5']