The Parity App uses the following Natural Language Processing and Machine Learning routines for answering the users queries.

Greedily adds sentences to a summary so long as it decreases the KL Divergence.

It creates a graph amongst all sentences of the messages and chooses the best sentences to create the summary based on the set threshold.

A opensource python library providing various functions including summary generation.

It calculates the score of phrases and words based on the frequency of occurrence, after removing stop words and punctuation

Built on top of LDA it is used to identify global topics (across multiple documents) and local topics (in the same topic documents). Further Reading: http://www.cs.cmu.edu/~pengtaox/poster/uai2013poster.pdf, https://arxiv.org/abs/0801.1063

A opensource python library providing various functions including keyword Extraction

A opensource python library provides keyword tokens

it is a graph based ranking model for text processing that employs the pagerank graph on top of unsupervised representation of a graph.