Objective: To help support a literature search strategy, including choosing appropriate research questions and choosing journals and databases Using chatbots to devise a research question ELM can help you start your research by suggesting questions and search strategies. If you know your broad topic but need help refining it into a specific research question, ELM can assist.For example, as an undergraduate beginning a dissertation on AI in agriculture, you might feel overwhelmed by the many possible topics. To narrow it down, you could ask ELM for suggestions.Your prompt could be: "You are an experienced academic researcher. Suggest research areas on how AI can help agriculture, particularly in sustainability and addressing climate change impacts. I am an undergraduate writing a 10,000-word dissertation, so the topics should be suitable for this level."By specifying your academic level and word count, you give ELM useful context. ELM might suggest questions like, "How can AI algorithms improve soil nutrient assessments to reduce fertilizer use and prevent runoff?" or "What are the water savings from AI-controlled smart irrigation in arid regions?"You don't have to use these questions exactly. They are meant to inspire and guide you in shaping your own research question. For instance, if you're interested in soil health, you can request more suggestions on that topic and adapt them based on your knowledge and interests. Using ELM to recommend scholarly journals ELM can be a helpful tool for recommending scholarly journals across various academic fields. Here’s how you can use it effectively:Following the guidelines on the AI Guidance for Staff and Students page, you might create a request such as, "As a research support librarian, list 5 peer-reviewed journals for climate change research, including a brief description of each."ELM might suggest journals like Nature Climate Change, Global Environmental Change, Climate Change, Journal of Climate, and Environmental Research Letters. These are reputable, well-cited publications. To check library access, you can perform a Journals Search using DiscoverEd.It's important to remember that while ELM offers useful suggestions, it might not list the highest-cited journals or those with a broad global perspective, such as journals from the Global South. To explore resources more thoroughly, consider using Scopus to find journals in your subject area, or consult your Library's subject guide. Your Academic Support Librarian is also available to provide advice. Using a chatbot to improve your search strategy If you have a question but are struggling to develop a search strategy for relevant academic literature, ELM can help by suggesting search terms.For example, using the prompt: "I am conducting a literature search on the role of AI in enhancing the accuracy of climate modelling and prediction to improve adaptation and mitigation strategies. I've tried 'climate modelling and AI' and 'climate change and AI' but got irrelevant results. How can I improve my search strategy?"ELM responds with strategies like using advanced techniques such as Boolean operators and phrase searching.After reviewing the options, I decide to refine my search terms and ask ELM for additional suggestions.ELM outlines the process of creating an effective search strategy, including identifying core terms and using Boolean operators. While it doesn't provide a perfect search, it offers a framework to improve my strategy, helping me get better results. Adapting these suggestions to my needs is the best approach rather than using them uncritically.For instance, using the prompt: "Suggest search terms for academic literature on the role of AI in enhancing climate modelling accuracy for better adaptation and mitigation strategies," ELM provides broad phrases like "AI applications in climate science" and "Data-driven climate models," and less specific terms like "Artificial Intelligence" + "sustainability" + "climate." Using these suggestions without refinement could lead to irrelevant results. Therefore, it's better to use ELM's strategies to refine and improve your search. Using ELM to recommend academic databases If you're struggling to develop a search strategy for academic literature, ELM can suggest relevant search terms.Providing context in your queries helps ELM offer better recommendations. For example, when asking ELM to suggest academic databases, mentioning databases you've found useful can guide its suggestions. Additionally, remind the tool to use reliable sources and cite them for easier tracking.For instance:"I am researching for my dissertation on artificial intelligence. I've found ACM Digital Library and Science Direct useful. What other databases would you recommend? Please use reliable sources and cite them."ELM might suggest databases like IEEE Xplore, Web of Science, SpringerLink, JSTOR, and Scopus. It could also mention Google Scholar for locating grey literature and AI-specific repositories like arXiv for pre-peer-reviewed research.This list provides a useful starting point. ELM usually includes brief descriptions and cites the database publishers, allowing you to explore their websites for more information on subject coverage.Remember, access to many academic databases may require subscriptions. Check your Library's Databases A-Z to see if access is available, then use the links provided to search the databases.While AI tools can suggest useful databases, it's important to familiarise yourself with the range of databases in your subject area. Consult the Library's Databases by Subject page or your subject guide for comprehensive information. Attribution: Videos used with permission from the Library Academic Support , The University of Edinburgh, Using Generative AI Tools in Academic Work Licence Type All rights reserved The University of Edinburgh This article was published on 2024-10-08