text - Python, extracting key aspects from consumer reviews -
i have data set of consumer reviews. these reviews extract occurring aspects. process applying includes
- step 1: tokenizing reviews sentences
- step 2: tokenizing sentences words after basic nlp pre-processing. pre-processing removes punctuation , english stop words.
- step 3: pos_tagging , extracting words pos tag of 'nn','nnp','nns','nnps'
- step 4: combining words across reviews find occuring words
- step 5: using top 40 terms aspects
is approach or recommend doing different?
Comments
Post a Comment