The code comprises of following Python scripts:
1. tot.py:
This is the file which implements a Gibbs sampler
for Events over Time.
The parameters are defined in the method called' InitializeParameters'
2. main.py:
This code calls for the Gibbs Sampler for training dataset.
It can be run as follows:
python main.py <train_file>
e.g: python main.py train.tsv
REMINDER: Please create a folder called 'results' in the
directory where you run the code.
3. visualize.py:
It reads result files dumped in the 'results' folder.
It gives visualization of the results.
python visualize.py
4. analyze_words.py:
It gives top 20 word for each topics.
python analyze_words.py
5. evaluation.py:
It implements time prediction to check the accuracy
of the implemented method.
python evaluation.py <test_file> <first_tp> <last_tp>
python evaluation.py test.tsv 20131229160416 20140602223903
Both test and train files have following tab separated format: <tweet_id> <cleaned_text>