Describe ideas how to improve clustering of WO

[blcham]

Describe possible experiments to improve clustering of MWOs, unless it is already used.

Ideas:

• word2vec approach (representing vectors for concepts based on prefLabel + altLabels)
• similarities in WO texts (word2vec approach for sentences)
• patterns in text-analysis pipeline
• adding information about area
• adding information about aircraft model

[blcham]

If you did some experiments already, put the outputs here.