xiajing10/akec

Unsupervised Key-phrase Extraction and Clustering for Classification Scheme in Scientific Publications.

Details coming soon.

---

Introduction

Description of code and data for Unsupervised Key-phrase Extraction and Clustering for Classification Scheme in Scientific Publications.

Requirements

Packages

• tqdm
• AllenNLP
• pke
• pytorch
• spacy
• numpy
• scispacy
• spherecluster

Resources

• scispacy models: "en_core_sci_sm"
• word embedding (save them to ./src/embed_data/:
  • ConceptNet Numberbatch 19.08 (English only)
  • ELMo weights and options

Description

• ./dataset/:

  • ieee_xai.csv: publication dataset collected from IEEE xplore
  • domain_terms.txt: domain glossary terms

• ./src/ :

  • test.ipynb has experiments demo
  • kprank.py has the main utility functions in keyphrase extraction
  • clustering.py has the main utility functions in keyphrase clustering
  • param.py defines the local links to recourses and data files required in code

Reference

• SIFRank: A New Baseline for Unsupervised Keyphrase Extraction Based on Pre-trained Language Model
• TaxoGen: Unsupervised Topic Taxonomy Construction by Adaptive Term Embedding and Clustering