SciLEx

The SciLEx (Science Literature Exploration) project is a basic python scriptbox made for :

• Request and run API crawler related to a research field
• Managing / Parsing / Deduplicating the collected papers
• Consolidate and Enrich a benchmark
• Exploring the citations links and expanding a network of sci. papers

I developed ScilEx scripts in the context of a systematic review conducted during my Phd, and introduced in :

Celian Ringwald. Learning Pattern-Based Extractors from Natural Language and Knowledge Graphs: Applying Large Language Models to Wikipedia and Linked Open Data. AAAI-24 - 38th AAAI Conference on Artificial Intelligence, Feb 2024, Vancouver, France. pp.23411-23412, ⟨10.1609/aaai.v38i21.30406⟩. ⟨hal-04526050⟩

---

SciLEx Framework

1. Crawl of already existing surveys on topic and push it on Zotero
2. Extract models, dataset from PaperWithCode and push it on Zotero
3. Get DOIs and obtain the citation network
4. Distill it with Zotero API / Annotate it on Zotero and distill your annotations

---

🔌 First steps

➕ Install Zotero and Zotero Connector

➕ Create a Zotero API key

➕ Create an account for following APIs:

• SemanticScholar
• Springer
• IEEE
• Elsevier

➕ ‼️ ADD ALL OF YOUR CREDITS IN scilex.config.yml 📋

---

📂 ScriptBox Content:

• 📋 Testing APIs scripts : test and check API services
• 📋 Collect scripts : run a collect > aggregate it and define new collectors
• 📋 Zotero scripts : extract or push papers data in the lib
• 🔧 Paper With Code scripts : extract or push papers data in the lib
• 🔧 Citations scripts : extract or push papers data in the lib
• 🔧 DOI and ORCID scripts : extract or push papers data in the lib
• 🔧Textmining scripts : extract or push papers data in the lib

🤓 How to contribute to SciLEX ?

• By extending the API integrated to SciLex
• By Improving the metainformation integration
• By extending it to analytics and vizualisation tools

Concretely all of theses questions could be leveraged and organize via issues.

API integrated

	SemanticScholar	OpenAlex	Istex	IEEE	HAL	Elsevier	DBLP	Arxiv	Springer
requiere API key ?	optional	NA		X	NA	X			X
Rate limit	100 req/sec	10/seq - 100000/days		10/sec – 200/days				3/seq	8/seq
Year	X	X	X	X	X	X			X
Abstract content	X		X	X
Title content	X		X	X	X				X
Document type	X	X	?	X	X				X
Classification ?	fieldOfStudy	conceptID, Wikidataconcept		IEEE thesaurus, indexterms	acm_classif, HAL classif, keyword, JELclassif...				keywords
title	X		X	X	X	X	X	X	X
abstract	X		X	X	X			X	X
DOi	X	X	X	X		X	X	X	X
citations metrics	X	X		X		X
publication data	X	X	X	X		X	X		X
isOpen	X	X	X	X		X	X		X
journal	X	X	X	X			X	X	X
conference	X	X	X	X			X	X	X
authors	name, author id	name, orcid, inst		X	X	X	X	X	X
publicationType	X	X	X	X	X	X	X		X
referenced_works		X	X
related_works		X
keywords		X	X	X	X			X
related entities		X	X
qualityIndicators			X
enrichments			X	X
fieldOfstudy	X		X						X