GitHub - mtthss/experiments-in-summarization: Supervised Multidocument Summarization using neural sentence embeddings

Experiment on various multidocument summarization methods

evaluate classifical hand crafted relevance features
learn linear/non-linear models for relevance, training on DUC conference datasets
explore the use of neural sentence embeddings for identifying redundancy
tune the trade-off between relevance and redundancy on development data
compare the use of A* search and Dynamic Programming to find best summary given score functions and trade-off

Name		Name	Last commit message	Last commit date
Latest commit History 64 Commits
.idea		.idea
evals/rouge_output		evals/rouge_output
README.md		README.md
avg_scores.py		avg_scores.py
cnn_language.py		cnn_language.py
data_structures.py		data_structures.py
functions.py		functions.py
postprocessing.py		postprocessing.py
preprocess.py		preprocess.py
rnn_language.py		rnn_language.py
summarizers.py		summarizers.py
test.py		test.py