Company_Acquisition_NLP
Statement contains a news about acquisition of a company , by a company and it predicts using Natural Language Processing about [ Acquired_Final ] & [ Acquired_Target ]
Standard Approach
Match unique id from corpus set and training set
Tokenizer splits data into small figments
Use tagger library which informs about nouns, verbs, adjectives etc.
Use parser library defines the dependency of words to each other
Match unique id from corpus set and test set and predict Acq_Final & Tar_Final according to result of above algorithm
Our Approach! Let’s Discuss.
Tokenizer
Stemming
Tagger
Rule Based Approach
Rule Based Approach
Proper Noun:
Buy: Token after buy is target
Sell: Token after sell is target
Sells: Token after sells is target
Sells to: Token after sells to is acquiring
Additional Details
Locate:
Problem_Statement_ML_MnA.zip
in repository for dataset and description of dataset