GrASP: A Library for Extracting and Exploring Human-Interpretable Textual Patterns

TL;DR

GrASP is a Python library that enables extraction and exploration of human-interpretable textual patterns, facilitating data analysis and understanding in various NLP tasks.

Contribution

The paper introduces a novel Python library implementing the GrASP algorithm, allowing flexible pattern extraction with built-in and custom attributes, plus a web interface for data exploration.

Findings

Effective in classification tasks like spam detection and argument mining

Assists in model analysis for machine translation

Helps discover artifacts in datasets such as SNLI and 20Newsgroups

Abstract

Data exploration is an important step of every data science and machine learning project, including those involving textual data. We provide a novel language tool, in the form of a publicly available Python library for extracting patterns from textual data. The library integrates a first public implementation of the existing GrASP algorithm. It allows users to extract patterns using a number of general-purpose built-in linguistic attributes (such as hypernyms, part-of-speech tags, and syntactic dependency tags), as envisaged for the original algorithm, as well as domain-specific custom attributes which can be incorporated into the library by implementing two functions. The library is equipped with a web-based interface empowering human users to conveniently explore data via the extracted patterns, using complementary pattern-centric and example-centric views: the former includes a reading in natural language and statistics of each extracted pattern; the latter shows applications of each extracted pattern to training examples. We demonstrate the usefulness of the library in classification (spam detection and argument mining), model analysis (machine translation), and artifact discovery in datasets (SNLI and 20Newsgroups).