Day 7 – Conducting research
Welcome to day 7 of the 12 Days of AI. Let’s learn about AI research tools.
What is ResearchRabbit?
ResearchRabbit is an AI tool designed for efficient literature mapping based on citations. Its primary function is to streamline your search for relevant references when embarking on an essay, small project, or literature review.
“Here’s how it works: you initiate your search with one or more papers (referred to as seed papers). The tool then identifies and suggests additional papers related to the topic established by your seed papers. With ResearchRabbit, there’s no need to toggle between different search modes and databases, eliminating the tedious and time-consuming citation mining process. The name ‘ResearchRabbit’ was inspired by the desire to avoid these daunting and often overwhelming rabbit holes of research”.
What does ResearchRabbit do with my information?
There’s been a fair amount of comments about privacy policies and how some of these companies keep and use information. We’ve gone back to old posts to add the correct links and moving forward each day will also have a link to the company’s privacy policy so you can learn more.
Explore ResearchRabbit’s privacy policy.
How do I access it?
Create an account with ResearchRabbit. Alternatives are listed below.
Your task:
Follow these instructions:
- Watch this short video on how to use ResearchRabbit (6 Mins).
- Start a New Collection called “Winter Weather”
- Add the following references to your Winter Weather collection:
- ‘Demographics in an alpine reindeer herd: effects of density and winter weather’ by Timo Helle and Ilpo Kojola.
- ‘Activities of sprites and elves in the winter season, Japan’ by Yukihiro Takahashi, Rina Miyasato, Toru Adachi, Kazuhiro Adachi, Masaaki Sera, Akihiro Uchida, Hiroshi Fukunishi
- Add two more articles recommended by ResearchRabbit to your collection
If you have time, create a New Collection of your choosing and add some references to it.
Alternatives to Research Rabbit:
Optional reading:
- ResearchRabbit is out of beta- my review of this new literature mapping tool (Medium)
- Navigating the Digital Frontier: AI-Powered Insights in Dissertation Research (Media and Learning)
Join the conversation:
Leave a reply below or tweet your comment using the hashtag #12DoAI.
- How did you get on with the task – did you find it useful?
- Would this be a useful tool for conducting your own research?
- What did you like about the layout and its ability to find other related articles?
- Can you see any potential limitations?
- Research Rabbit doesn’t explain how its algorithms work – does this matter?
- If a student asked you could they use it to do and assignment for you – what would be your response?
Join the competition:
Enter the competition to win 1 of 3 Tobbie the Self-Guiding AI Robots or 1 of 3 one-year licences for Teachermatic.
Having worked in libraries for a long while now, we have seen lots of similar services/sites over the years (i personally still don’t get those visualisations). They never quite do what you hope they will do, lots of clicking through to articles only to find you don’t have access to them. At present in most library catalogues or academic databases you can set up alerts based on keywords so that when new papers/books appear you get a message. I also found the interface quite cluttered and confusing. This of course may all improve but there is nothing new here for me. It looks like it grabs information from ‘Semantic Scholar’ and builds upon lots of academic aggregators and open access sites.
Let me start recounting the experience. I began with a new collection called ‘Winter weather’, and searched the suggested the items. No result was yielded when I directly copied and pasted the references in step 3. However, when I just kept the article titles, the tool found them in a flash. In addition to ‘Winter weather’, I also created a collection about ‘metadiscourse’ and ‘ChatGPT’ to explore the tool. I tried various functions in the interface, and was amazed by how fast it could come up with similar papers or papers by the same authors. The visualisation function allowed me to see the similar works in terms of not only network but also timeline is also quite handy. The function to add comments is also great. The tool also support connection to Zotero, but I don’t know if it works with RefWorks that UAL has subscription to.
I do share the concern about what the algorithm behind all the magic. How is the tool similar to searches of databases that our library subscribes to. I don’t think that ResearchRabbit could access materials behind paywall; that is to say, it may be very much relying on publicly access materials (especially paper abstracts and book blurbs). This impliesthat the results may be skewed and partial. I also observed that the latest results it could generate are papers in 2022 – probably not the most up to date.
I find ResearchRabbit a very useful and efficient tool for students to explore a topic. However, they may need to explore further based on what’s provided there. It could be used as a (partial) alternative to a management reference tool, too. My main concern is whether it is really ‘more’ efficient and effective than what our library has offered (databases + RefWorks). First of all, the sources from our library searches tend to be credible ones, while it is possible that some papers via ResearchRabbit come from predatory journals. Secondly, UAL library provides many specialist resources of creative disciplines which ResearchRabbit may not have access to. It seems to be me that ResearchRabbit in the current form is best regarded as upscale of random web (Google or others) searches that unfortunately some students very much rely on.
It would seem the two you suggested actually aren’t that linked based on the mind map. I did add my own paper and was pretty okay at finding other relevant works. I wouldn’t use this without running the results by my supervisor though.
Ooh thanks for informing me of this James – I did try them and they worked for me but it does go to show that ResearchRabbit has a limited data base!
I would have found this tool really useful as a way to store and comment on citations when I was doing my research. In my fairly niche research topic it flagged some interesting research that happened at the same time and after mine. Of course it would need to be used alongside the tools from the library but could be particularly useful in areas where many articles are open access and for researchers without easy access to University research management tools.
I found this to be a useful tool in terms of guiding younger students who may be new to academic research. However, the limited time I spent playing around with this tool didn’t seem to yield anything that I couldn’t find on a library database. The visual connection ‘map’ doesn’t indicate how these papers are related and I do think there should be an explanation about the criteria used for ‘suggested authors’, which raises the question about how the algorithms work. I suppose once I got to know the more advanced filters then I could specify in more detail the specific content or approaches within the field I am researching.
I think it has some limited use in research. I uploaded a Zotero collection to see what else it could come up with. Many of the items in my collection were not synchronised by Research Rabbit so I assume it wasn’t able to access them. For the remainder, it came up with its visualisations, which I found a bit hard to navigate but I did like the fact that it came up with some other papers that looked interesting, and Research Rabbit was able to synchronise them back into my Zotero collection. I think I would use this feature to pick up papers that I’d missed in previous research but I wouldn’t use it as a starting point.
I watched the video on how to use ResearchRabbit, then set up an account and did the assigned task and it was very straightforward.
I did a literature review earlier this year about online assessment so I started a collection with one of the key papers and had a look at what came up.
I think this is a tool I would use for research but I would want to follow up the technicalities e.g. the privacy, the algorithm. I think I would be happy for a student to use it provided they shared the relevant collection with me or added it as an appendix.