articlesumm 0.0.2

Custom scientific/research article summarization library based on Statistical features

The summarization of research articles is a complex task compared to general-purpose summaries.This is a result of the distict nature or semantic structure of these scientific articles. The presence of inline citations and summarization modules bias to certain Text features that often work well on less-tecnical text but fail to produce coherence in this area are all underlying factors.

We circumvent these challenges in order to produce more coherent, human-understandable summaries of manuscripts and research text using this libary.

Installation

You can easily install the package using the pip command:

  pip install articlesumm

Usage

The package takes a string as input(specify a path/directory for an article or alternatively pass a string as a variable). The tokenization of the sentences and words can be performed with the first function:

  parse=purge(text)
  type(parse) 

  #tuple

Alternatively, you can tokenize the sentences and words with any other technique and pass the processed text to the summarization model.

Example

  text='''TextRank is a graph-based ranking model for text processing which can be used in order to find the most relevant sentences in text and also to find keywords. The algorithm is explained in detail in the paper at https://web.eecs.umich.edu/~mihalcea/papers/mihalcea.emnlp04.pdf . In order to find the most relevant sentences in text, a graph is constructed where the vertices of the graph represent each sentence in a document and the edges between sentences are based on content overlap, namely by calculating the number of words that 2 sentences have in common.'''

  from ArticleSumm import purge
  from ArticleSumm import summarizer

  parse=purge(text)

  #summary=summarizer(text,parse[0], parse[1], summary_length=3)
  summary=summarizer(text,words=purge(text)[0], sentence_list=purge(text)[1], summary_length=3)
  print(summary)