README.txt

1.Data Cleaning.ipynb: The basic steps for data cleaning and data statistics
2.Further Cleaning and lexicon-based method.ipynb: Firstly, tokenization, lemmalization and POS tagging. Secondly, the implementation of three lexicon-based approaches.
3. Initial data analysis.ipynb: Wordcloud and word frequency
4. Data analysis for lexicon-based methods.ipynb:  Statistics of each city, each stage as well as each method.  Complete some data analysis and visualizations.
5.Supervised Approaches.ipynb: The implementation of some supervised classifiers and evaluation