Feel free to submit questions on slack: https://join.slack.com/t/cbdmworkspace/shared_invite/zt-vwpbzhox-CHneZn9uth~iIkB5hEj1qA
This will be a guide to implementing ML algorithms as can be used for the course CBDM at ITU. The models include:
- Clustering (kMeans)
- Classification (kNN)
- Network analysis (iGraph)
- Natural language processing (spaCy)
-
for LDA see https://www.kaggle.com/thebrownviking20/topic-modelling-with-spacy-and-scikit-learn/data. But replace parser with:
def lemmatizer(doc): return [token.lemma_ for token in doc]
def spacy_tokenizer(setence): doc = nlp(setence) tokens = lemmatizer(doc) tokens = [ word for word in tokens if word not in stopwords and word not in punctuations ] tokens = " ".join([i for i in tokens]) return tokens
-
Additionally we look at how to use pandas to clean/preprocess data.