DataCite

Kristian Garza

In the winding world of academic research, the quest for an appropriate repository to store and share scholarly data has been akin to a modern-day odyssey. Traditionally, this journey involved wading through static pages and search indexes, a process both time-consuming and fraught with uncertainty. However, the advent of Language User Interfaces (LUIs) has heralded a paradigm shift, reshaping the landscape of repository discovery and accessibility.

![From left to right: The Evolution of Finding Scholarly Repositories in the Age of LLMs](https://cdn-images-1.medium.com/max/1024/1*WFEnG146R-wCpkUJejrRrw.png)

### The Rise of Language User Interfaces

As with many quests in the realm of academia, various projects were spun to address this problem, each achieving moderate success. These endeavours, while commendable, often lacked the sophistication to fully grasp and cater to the diverse and evolving needs of the scholarly community. It was in this setting that Language User Interfaces (LUIs) emerged as a groundbreaking development.

The emergence of tools like the [Repository Finder](https://doi.org/10.5438/wday-8958) underscores the growing need for sophisticated repository discovery methods. Similarly, the [FAIRsFAIR initiative](https://www.fairsfair.eu/repository-discovery-datacite-commons) has contributed significantly to this landscape by integrating repository information into DataCite Commons, thus improving the discovery of FAIR-enabling repositories.

Nevertheless, researchers faced significant challenges with these methods of finding repositories. These methods, primarily involving static lists and search indexes, had several drawbacks:

1.  Search indexes were limited in understanding the nuanced and context-specific needs of researchers, resulting in less relevant recommendations.
2.  Static lists of repositories offered little to no customization to individual researchers' needs, often leading to a mismatch between the repository features and the researchers' specific requirements.
3.  Navigating through extensive lists or indexes is not only tedious but also time-consuming, potentially diverting valuable time from actual research activities.

In contrast, LUIs address these limitations by offering a more personalized, efficient, and context-aware approach to repository discovery.

[Language User Interfaces (LUIs) are a subset of human-computer interaction where users engage with digital systems through natural language, either textually or vocally](https://www.youtube.com/watch?v=l5mG4z343qg&list=PL1T8fO7ArWleyIqOy37OVXsP4hFXymdOZ). This approach stands in stark contrast to traditional graphical user interfaces (GUIs), where interaction is based on visual elements and direct manipulation. LUIs leverage advanced computational linguistics and Large Language Models, such as GPT-4, LLama and Bard, to interpret, process, and respond to user queries conversationally. This human-centric design paradigm makes them particularly well-suited for complex tasks like navigating the diverse landscape of scholarly repositories, where the subtleties of human language and specific domain knowledge play a critical role.

### Emerging LUI Tools in Scholarly Data Management

[ChatGPT - Goldie the Scholar Retriever](https://chat.openai.com/g/g-nnBbYTm6f-goldie-the-scholar-retriever)

Among these developments is Goldie, the scholar retriever. Goldie, using the combination of OpenAI's GPT-4 Large Language Model and RAG (Retrieval-Augmented Generation), provides a conversational interface for guiding you to find a scholarly repository. This service transcends traditional boundaries by offering a conversational interface where researchers articulate their needs in natural language, and in return, receive tailored recommendations for suitable repositories. While it's a notable advancement, it's one of many tools contributing to the evolution of scholarly data management, helping to offer more personalized repository recommendations. Other examples include the [Dimensions AI assistant](https://www.dimensions.ai/products/all-products/dimensions-ai-assistant/), [Scholar AI](https://scholar-ai.net/), and many others.

<figure>
<img src="https://cdn-images-1.medium.com/max/1024/1*_Pxf8zFfwGIJ2QXfUNe--w.png" alt="UI with suggestions in directly from chatGPT." />
<figcaption>Goldie's suggestion on its landing page.</figcaption>
</figure>

![Goldie conversation for crystallographic data structure repositories.](https://cdn-images-1.medium.com/max/1024/1*B2_mahags6oBvrY-GnFvMw.png)

### Technological Innovations and Enhanced Accessibility

Moreover, the application of OpenAI GPT in the backend of services like Goldie represents a leap in understanding and responding to complex, context-driven queries. Coupled with RAG, it ensures that the responses are not only contextually relevant but also enriched with information retrieved from a vast corpus of scholarly repository knowledge.

While the evolution from static pages and search indexes to sophisticated LUIs like Goldie marks a significant milestone, it's also a journey that continues to evolve. As the creator of Goldie, I am keenly aware that the development of such tools is an iterative process, one that flourishes with community input and collaboration.

Therefore, I extend an invitation to the scholarly community to engage with Goldie. [Your feedback,](https://www.inmedias-res.org/goldie) experiences, and insights are invaluable in shaping Goldie to better meet the diverse and evolving needs of the global research community. Together, we can ensure that the future of scholarly communication is not only more accessible and efficient but also collaboratively enriched.

[Kristian\'s Substack \| Kristian Garza \| Substack](https://kjgarza.substack.com/embed)

![](https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=176c42cd7c4d){width="1" height="1"}


In the winding world of academic research, the quest for an appropriate repository to store and share scholarly data has been akin to a modern-day odyssey. Traditionally, this journey involved wading through static pages and search indexes, a process both time-consuming and fraught with uncertainty.

The Impact of Language User Interfaces on Finding Scholarly Repositories

### Academic Publishing web forms meet your demise: The unstoppable rise of large language models (ChatGPT)

![Prompt to create DOI metadata using ChatGPT.](https://cdn-images-1.medium.com/max/1024/0*VddPt0uQe4DvX791.png)

As we enter the age of artificial intelligence, it's worth considering how large language models will revolutionize how we interact with websites and applications. Web forms have been the dominant method for users to input data and complete tasks online for decades. But as anyone who has struggled to fill out a form on the web can attest, these forms can be clunky and frustrating. Large language models offer a more intuitive and human-like alternative. Imagine being able to speak or type your request to a website or application, rather than having to remember which information goes in which field and navigate through multiple pages of a form. This level of convenience doesn't exist, but chatGPT shows signs of the possibilities.

An area of application in which I'm genuinely interested is **metadata input in academic publications and data**. The replacement of web forms with large language models for metadata input in academic journals has the potential to streamline the submission process significantly. Traditionally web forms can be time-consuming and laborious to complete, especially for academics who may have to input large amounts of information about their research and publication history. With large language models, academics could type prompts about their metadata into a web application, saving significant time and effort. This could also lead to fewer errors in the submission process, as large language models can understand the context and provide more accurate input. Adopting large language models for metadata input in academic publications could lead to a more efficient and user-friendly submission process.

In the past, one potential criticism of this approach is that LLM may only sometimes be able to understand and respond to user requests accurately. While large language models have made significant progress in recent years, they can still struggle with ambiguous or complex language, particularly in domain-specific contexts such as academic publications. This could lead to confusion or frustration for users who may not get the desired results from their natural language queries. However, [OpenAI announcement](https://openai.com/blog/chatgpt/) of the GPT-3.5 series model and its flagship app chatGPT has potentially shattered that criticism. Since the announcement, the web has been floated with examples of prompts generating code, creating copy, and answering questions.

ChatGPT has shown to excel on the tasks above, but how would it fare creating metadata for academic publications? This metadata is often made with large web forms. The content submitted in these web forms is often serialized and put in a payload submitted in APIs. The payloads follow a particular schema. In the case of academic publications metadata, this is often DOI metadata. These metadata payloads follow a schema that lives on the web; therefore is likely that chatGPT knows about it and, when prompted, can generate valid metadata.

I did a small experiment to test that assumption, and the results are pretty interesting. I asked chatGPT to generate the metadata for a dataset with a specific title. The title is actually from a real resource on the web (i.e., [10.6084/m9.figshare.7588511](https://doi.org/10.6084/m9.figshare.7588511)), but the rest of the metadata requested is all fictional. See the example below:

> **User:** Create the metadata, using datacite schema and xml format, of a dataset on the titled "Predicting Outcomes from First Session Interactions in Remote Cognitive Behavioural Therapy for Severe Health Anxiety" under a cc-by license and authored by barry blitt from NYU

> **ChatGPT:** Here is an example of metadata for the dataset you described, using the DataCite schema and XML format:

    <resource xmlns="http://datacite.org/schema/kernel-4" 
              xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 
              xsi:schemaLocation="http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd">
      <identifier identifierType="DOI">10.5072/example-dataset</identifier>
      <creators>
        <creator>
          <creatorName>Barry Blitt</creatorName>
          <affiliation>NYU</affiliation>
        </creator>
      </creators>
      <titles>
        <title>Predicting Outcomes from First Session Interactions in Remote Cognitive Behavioural Therapy for Severe Health Anxiety</title>
      </titles>
      <publisher>NYU</publisher>
      <publicationYear>2022</publicationYear>
      <subjects>
        <subject>Cognitive Behavioural Therapy</subject>
        <subject>Severe Health Anxiety</subject>
        <subject>Remote Therapy</subject>
      </subjects>
      <rightsList>
        <rights rightsURI="http://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution 4.0 International</rights>
      </rightsList>
      <descriptions>
        <description descriptionType="Abstract">
          This dataset contains data collected from first session interactions in remote cognitive behavioural therapy for severe health anxiety. It includes information on patient demographics, therapy session details, and prediction outcomes. The data is intended to be used for research purposes.
        </description>
      </descriptions>
    </resource>

Surprisingly the metadata generated is valid Datacite Metadata, which includes all the metadata supplied. The language model also generated a description and subject/keywords to classify the dataset. These subjects, when compared with the real resource on the web (see [metadata](https://api.datacite.org/application/vnd.datacite.datacite+xml/10.6084/m9.figshare.7588511)), are not precisely the same; nevertheless, the generated subject fall within the knowledge domain of the web resource.

Over the last couple of weeks, I have run similar experiments, all giving similar accuracy. The results are a far cry from traditional web forms, which often require selecting specific inputs to use specific fields. Creating the same output in the usual web form would have required a user to enter data in eleven input elements. Using dialogue via chatGPT presents a superior user experience/convenience and decision support as opposed to filling out a time-consuming form. But the benefits of large language models go beyond convenience. As seen in the example, these models can also understand the context and provide richer responses.

As large language models continue to improve, they will likely replace traditional web forms for many tasks. This change could lead to a more seamless and enjoyable user experience online. So the next time you're confronted with a tedious web form, imagine the potential for a more significant and effortless way to interact with the digital world.

![](https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dbb1bd6c2ee8){width="1" height="1"}


Academic Publishing web forms meet your demise: The unstoppable rise of large language models (ChatGPT)     As we enter the age of artificial intelligence, it’s worth considering how large language models will revolutionize how we interact with websites and applications. Web forms have been the dominant method for users to input data and complete tasks online for decades.

Rogue Scholar Posts

The Impact of Language User Interfaces on Finding Scholarly Repositories

Academic Publishing web forms meet your demise: The unstoppable rise of large language models…