Exercise 1. LDA on Cancer vs. AIDS dataset

0 ×

Extract 4 topics from articles in file Thedata/Pubmed_Articles.csv.
Use the LDA algorithm and describe each topic with 10 keywords.
Draw a word cloud on the discovered 40 keywords, with color by topic and size by LDA weight.
Are the topics uncovering mouse cancer and AIDS related articles?

Nodes

• Component Input2 ×
• Component Output2 ×
• CSV Reader1 ×
• Case Converter1 ×
• Color Manager1 ×
• Excel Reader1 ×
• Number Filter1 ×
• POS Tagger1 ×
• Punctuation Erasure1 ×
• Snowball Stemmer1 ×
• Stop Word Filter1 ×
• Tag Cloud (JavaScript)1 ×
• Tag Filter1 ×
• Topic Extractor (Parallel LDA)1 ×

Extensions

• FeatureKNIME Base nodes
• FeatureKNIME Excel Support
• FeatureKNIME JavaScript Views
• FeatureKNIME Textprocessing

Links

• www.knime.com/knimepress/from-words-to-wisdom

  www.knime.com/knimepress/from-words-to-wisdom