A Geo-Social Characterization of Health Impact from Air Pollution in Mexico Valley

Listing 1

Discovering social topics - algorithm.

	Input: tweets_dataset_input
	/ Setting parameters /
	K_clusters = 20
	Stop_words = {“http: //”, “https://”,”@”,”#”…}
	N-Gram_range = {1,4}
	LDA_passes = 10
	Begin
	input = remove_stopwords(filtered_input)
	Corpus = gensim.CreatingN-Grams(input,N-Gram_range)
	Clusters = K-Means (Corpus, K_clusters)
	SocialTopics = LDA_process_gensim(Clusters, K_clusters,LDA_passes)
	Output:SocialTopics_Set
	Call:
	CreatingSocialDataCubes(SocialTopics_Set)