How should the input corpus of gensim LDA look like?
0
I try two different kind of input corpus to put into gensim LDA model My document is: documents = ["Apple is releasing a new product", "Amazon sells many things", "Microsoft announces Nokia acquisition"] texts = [[word for word in document.lower().split() if word not in stop_words] for document in documents] texts1 = for i in texts: for t in i: texts1.append([t]) And use gensim to make it into corpus corpus = [[(0, 1), (1, 1), (2, 1), (3, 1)], [(4, 1), (5, 1), (6, 1), (7, 1)], [(8, 1), (9, 1), (10, 1), (11, 1)]] corpus1 = [[(0, 1)], [(1, 1)], [(2, 1)], [(3, 1)], [(4, 1)], [(5, 1)], [(6, 1)], [(7, 1)], [(8, 1)], [(9, 1)], [(10, 1)], [(11, 1)]] Is there a huge difference if I use this two kind of way to put it into LDA model? When I try th
