AI Support Systems for Academic Research

https://sites.google.com/view/ai4academicswisstext2024/home 

Background

With the surging of AI technologies, academicians have been adapting the ways research works are
produced, transmitted and evaluated. AI support systems are widely used to retrieve information, such as
journal recommenders [1] [2], SemOpenAlex [3], (LLM and KG powered) academic chatbot [4], SciSpace
[5], Consensus [6], Research GPT [7]. In this hands-on workshop, we will go through three presentations
on topics to understand (1) academic networks through the lens of OpenAlex, (2) journal recommendation
techniques, (3) LLM and KG powered academic chatbot. Then the participants will have the chance to
interact with a Telegram academic chatbot we design in-house at ETH Zurich. The academic chatbot
supports retrieving papers, authors and affiliations that are relevant for user queries (e.g., Recommend the
top-5 papers related to the paper “Attention is All You Need” by Ashish Vaswani.).

Organizers

• Dr. Susie Xi Rao, Senior Scientific Staff at ETH Zurich
• Noah Mamié, Research Assistant at ETH Zurich
• Yilei Tu, Research Assistant at ETH Zurich
• Prakhar Bhandari, Research Assistant at ETH Zurich

Invited Speaker

Mahmoud Hemila, Data Scientist at ETH Library

• Topic: “Recommendation System for Journals based on ELMo and Deep Learning”
• Abstract: The work evaluates how adequate recommender systems are for the selection of
  journals that fit to scientific publications. Specifically, several word embedding (word2vec,
  tf-idf, ELMo) and classification (LR, CNN, RNN, MLP) methods were tested and evaluated
  against each other in terms of their recommendation accuracy.
• Source: [1]

Aim

We want to explore together with the participants the frontier of AI support systems for academic research.

Task

• Presentations and Q&A
  • (P1) academic networks through the lens of OpenAlex by Noah and Prakhar,
  • (P2) journal recommendation techniques by Mahmoud,
  • (P3) LLM and KG powered academic chatbot by Susie.
• Hands-on session with academic chatbot

Participant

Accessible to the general audience and specialists.

Support material

We will provide a workshop website (similar to [8] [9]), with which we share the materials and
reports like [10] [11].

Reference

[1] Hemila, Mahmoud, and Heiko Rölke. “Recommendation System for Journals based on ELMo and Deep
Learning.” 2023 10th IEEE Swiss Conference on Data Science (SDS). IEEE, 2023.
[2] https://journal-recommender.sagepub.com/
[3] https://semopenalex.org/resource/semopenalex:UniversalSearch
[4] Rao, Xi Susie, Noah Mamié, Yilei Tu, and Peter Egger. Knowledge-Enhanced Academic Chatbot:
Harnessing Large Language Models and Knowledge Graphs. Under review in ARR, 2023.
[5] https://www.scispace.com
[6] https://consensus.app/search/
[7] https://github.com/mukulpatnaik/researchgpt
[8] https://sites.google.com/ethz.ch/keywordswisstext2022/home
[9] https://sites.google.com/view/greenwashingswisstext/home
[10] Rao, Susie Xi, Piriyakorn Piriyatamwong, Parijat Ghoshal, Sara Nasirian, Emmanuel de Salis, Sandra
Mitrović, Michael Wechner, Vanya Brucker, Peter Egger, and Ce Zhang. “Keyword Extraction in Scientific
Documents.” arXiv preprint arXiv:2207.01888, SwissText2022. https://arxiv.org/pdf/2207.01888.pdf
[11] Lu, Guang, Janna Lipenkova, Susie Xi Rao. SwissText2023.
https://drive.google.com/drive/folders/126J34mGwCgEZ8MKVYdqTSr_MLuHTqzxu