This program is to analyse and group together positive, negative and neutral tweets during the time of demonetization in India.
- Create the following directories from the root directory of the project -> dataset, dictionary, results
- Downloaded dataset from the following kaggle link
- Upload the downloaded dataset to the dataset directory
- Extract the word weights from nltk word corpus
- Upload the extracted word weights to dictionary directory created in the step 1
- Load the data and perform analysis on the data using Pig Latin
- Set the proper file paths in the Pig Latin Code file for the output
- Run the Pig Latin script using the following command
pig -x mapreduce <path to the script>
- If the steps are followed correctly then output of the code will be generated in results directory