UJ Library - Honours Research PathFinder: Unit 2: Pose questions and find sources:boost your research with AI tools

See this useful  Generative AI Literacy course designed to equip you with the foundational knowledge and practical skills needed to navigate the AI-driven landscape

Prompt Crafting is the process of designing clear and specific instructions for  Generative AI models to generate desired outputs. It involves framing requests in ways that guide the tool to produce relevant, accurate and useful responses. See this Guide on the basics of prompt crafting, strategies for effective prompts and examples to optimize  Generative AI performance in research.

Always acknowledge the use of AI tools in your work. AI-generated references may be inaccurate—verify them against original sources.

Refer to the UJ Harvard Quick Guide fo instructions on referencing Generative AI software. 

For APA 7th edition citation guidelines on Generative AI, see this guide

Generative AI refers to artificial intelligence models that can generate text, images, code and other content based on patterns in existing data. These tools such as ChatGPT and Bard can assist in brainstorming ideas, summarizing articles and even improving academic writing.  For a deeper understanding, see this guide on the anatomy of AI.

While these tools can be valuable, it is crucial to use them ethically—verifying sources, avoiding plagiarism and ensuring AI-generated content adheres to academic integrity standards. Refer to the CODE for Academic and Research Ethics. 

When used responsibly, GenAI can improve efficiency and critical thinking while maintaining the credibility of your research.

Before incorporating AI tools into your coursework or research, always check your lecturer’s or department’s guidelines.

With regards to research, the UNESCO Quick Start guide proposes possible uses of generative AI at different stages of the research process: from research design (e.g., brainstorming research questions) to data collection (e.g., preparing for interviews), data analysis and writing up. 

Brainstorming and topic selection

• IdeaMap.ai- designed to enhance brainstorming and idea generation 
• Google Scholar – while not AI in the strictest sense, Google's search engine helps identify gaps in research.
• Semantic Scholar -  reveal unexpected research areas relevant to your topic by highlighting connections between papers, even if they do not match your initial keywords.

Identifying research questions

• ChatGPT (Free Tier) – for brainstorming and generating research questions based on input.
• Research Rabbit (Free Version) – allows you to discover related research papers and topics for free.
• Connected Papers (Free) – helps visualize the connections between topics and research papers.

Conducting Literature Reviews

• Elicit - generates literature matrix and a review with real citations
• SCOPUS AI - provides AI-driven insights and summaries from a vast database of academic content. 
• NotebookLM - summarises information and generates insights from uploaded document

Data Analysis

• SPSS -  a valuable tool for quantitative research, helping researchers analyze numerical data and conduct various statistical tests and analyses to explore patterns, relationships and trends. To download SPSS, follow the steps below. 

           >Log into ULINK

           >Click  on UJ Tools

           >Scroll down to Academic/Research Tools/Software

           > Click on the IBM SPSS link and select the relevant operating system.

• ATLAS TI - software for qualitative data analysis. It helps researchers analyze non-numerical data such as interviews and open-ended survey responses. Click HERE to request access. 

Writing and Revision

• Grammarly - helps you eliminate errors in grammar, spelling, style, and tone, ensuring that you find the perfect words to express yourself. It is available for download after authorization. To request a license, follow these steps:

            >Log into ULINK.

            >Click on UJ Tools.

            >Scroll down to Academic/Research Tools/Software.

            >Select Grammarly and click on the Request Access link.

• Bias and Fairness: Generative AI models may perpetuate or amplify biases present in training data, leading to unfair or discriminatory outcomes in research findings.

• Transparency and Accountability: Researchers must ensure that AI-generated content is transparent, with clear documentation on how models are used and the sources of data.

• Data Privacy: Generative AI tools may process sensitive or personal data. Ethical use requires obtaining consent and ensuring data is anonymized or protected.

• Plagiarism and Authorship: AI-generated content should be properly attributed, and researchers must avoid misrepresenting AI-generated work as their own.

• Reliability and Accuracy: Generative AI tools must be validated to ensure that the information they provide is accurate and reliable for research purposes.