Existing alternatives

[pesho-ivanov]

What are the existing tools and systems (both free and proprietary) that overlap with CitationGecko's goals? What functionalities do they have that are missing in CitationGecko?

I don't use other tools but suspect existing reference management systems are, for example, recommending new papers based on a user's library.

[aarontaycheehsien]

What are the existing tools and systems (both free and proprietary) that overlap with CitationGecko's goals? What functionalities do they have that are missing in CitationGecko?

I don't use other tools but suspect existing reference management systems are, for example, recommending new papers based on a user's library.

To me the value in citationgecko is more than merely recommendations. The shape of the visualization is where it's real value lies IMHO.

[barney-walker]

There are many tools in the business of helping scientists discover papers.

Loads of tools use keyword input to generate recommendations i.e. Google Scholar, Microsoft Academic, Web of Science, SCOPUS, ScienceOpen, SemanticScholar, Meta. However, I find it can sometimes be difficult to construct a query using keywords that really captures what you are looking for.

Tools that use collections of papers to generate recommendations are rarer but there are a few i.e. ReadCube, Mendeley, PubChase and Sparrho. In my experience the problem with these is that you often need to provide a large number of papers before they start generating useful recommendations. Also they aren't very transparent about WHY a paper has been recommended - it's just some black box machine learning algorithm.

With Gecko you only need a few papers to start discovering more and you can see exactly how the recommended paper is connected to the others and it is always because a human expert in the field has specifically asserted it's relevance via the act of citation.

[aarontaycheehsien]

I've recently discovered another one

https://timwoelfle.github.io/Local-Citation-Network/

Is supposed to be open source on github but the link there from the faq is broken.

It does the time-directed layout where vertical position of the node tells you how old the paper is (kinda like https://www.citnetexplorer.nl/ or current views in Citespace) but overall the interface is not as good as Citation Gecko - which of all the tools I've tried as a interface I can imagine real users to use as opposed to bibliometrics people.

How this tool works is

1. you input a set of dois (even 1) = Citation gecko's seed papers

2. It will check the references of the input seed paper dois (if available), and then pick the top 10 references which have the most "local" citations (ie among the seed dois and references pulled, which are cited the most and not in the original input dois).

3. It will then visualize the citing relationship between these input dois + 10 recommended.

Citationgecko seems to only visualize relationships between seed papers and ignores any inter-reference counts. That's why for Citationgecko you must have at least more than 1 seed/input doi, while this other tools can do recommendations even if you input just one doi.

The main thing I like about this tool is

a) You can upload a set of dois (using MAG and my own key, roughly up to 500+ dois) and it will visualize

b) It allows use of Microsoft academic , including adding of your own key. I think Citationgecko doesnt do MAG anymore?
b)

[no-identd]

The link to the source code from the FAQ seems to work by now:

https://github.com/timwoelfle/Local-Citation-Network

Also, @timwoelfle has since added a shout out to citationgecko to the FAQ:

LocalCitationNetwork/LocalCitationNetwork.github.io@97e0e44

:)