Assignment Name - Advance Predictive Modelling Answer Statement – 1. How will you treat text having short cut words (like bcz, u, thr etc…)? Answer: After a text is obtained, we start with text normalization. Text normalization includes: converting all letters to lower or upper case converting numbers into words or removing numbers removing punctuations, accent marks and other diacritics removing white spaces expanding abbreviations removing stop words, sparse terms, and particular words text canonicalization Short cut words can be treated in 2 ways: Expand the short cut words: Stemming can bring the words in root form, though stemming object group needs to be defined for these words. Normalization techniques can be applied to expand these words. Remove the short cut words from text : By Tokenization in python or using “re” regex library or Stop words list can also be updated to remove these words from text.
You've reached the end of your free preview.
Want to read both pages?
- Fall '19