Artificial intelligence offers many opportunities to enhance the information searching process in higher education. This guide provides examples of how AI can be used in different stages of information searching and what tools are available.

Artificial intelligence is now involved in almost all information searching. In many services, you can already search for information using natural language questions. This means that you do not need to think about exact keywords or search queries. AI can organise search results, create summaries based on the results, or present the results in a visual format.

When you use AI in information searching, remember the following important points:

• There are different types of AI. Be aware of what data the AI has been trained on.
  • General generative AI applications are trained on open web content. They do not have access to information behind paywalls, such as standards.
  • AI tools in professional services search only the sources available in that service. This means that it’s important to know what information the service includes. For example, Rakennustieto Alvari assistant searches information only from the Rakennustieto service and does not add information from elsewhere on the web. It cannot be used to search for legal information, for example.
  • Special AI applications have been developed for academic information searching. These search for information in scientific publications.
• Do not rely on AI-generated summaries; always check the original source.
  • AI may provide incorrect information or may be missing important details.
  • The reliability and suitability of information always depend on the context. If the context is missing, it is impossible to evaluate the reliability of the information.
• Be careful when entering data into AI. Do not enter any information that you would not be willing to share openly on the web.
• Do not create a reference list automatically.
  • AI may give a source that does not exist or that does not actually discuss the topic mentioned in the text.

Using generative AI chats

General generative AI chat tools (e.g. ChatGPT, MS Copilot or Google Gemini) are suitable for information searches that are difficult with traditional search engines and where the result is easy to verify. For example, you can find out which authorities exist in your field across Europe or where you could find information on your topic.

It is always a good idea to ask AI for sources so that you can check the information.

For actual research information searching, you should use AI applications designed specifically for that purpose. They search for information in scientific databases.

1. Formulating the research question with AI Chats

• AI chat tools, such as ChatGPT or Copilot, can be used to help you refine and structure your topic. They help you think about your topic from different perspectives.
• Enter your topic into the language model and ask it to formulate your search question. Provide background information: what kind of information you are looking for and for what purpose. You can also assign the AI chat a role, from which it will answer your question.
• It is also helpful to ask the chat to ask clarifying questions before answering. This helps you refine your search question from different perspectives.

• Example prompt: “Formulate a research question on ‘the effects of climate change on agriculture in the Nordic countries.’ Ask me clarifying questions before responding.”

2. Brainstorming Keywords with AI Chats

• You can also use language model chats to brainstorm keywords for searching information with search engines.
• Input your topic or research question into the language model and ask it to suggest search terms. You can refine your question by specifying what kind of information you are looking for.
• You can also ask the chat to ask clarifying questions before responding. This helps you refine your information searching question from different perspectives.

• Example prompt: “Suggest keywords for searching research information in Finnish and English on ‘principles of sustainable development in corporate management.’ Provide also specific keywords. Ask me clarifying questions before responding.”

3. AI Applications Designed for Scientific Information Retrieval

• There are AI applications designed for scientific information searching. They search for sources in scientific databases. Using them complements traditional information searching. 
• AI applications designed for scientific information searching can bring a multidisciplinary approach to your information searching. They can also help analysing search results, for example by giving them in a visual form. In many applications, you can also search for information using a question written in natural language. 
• NOTE! Copyright also applies to scientific articles. Do not enter articles or other text written by someone else into an AI application without permission. 
• NOTE! Some tools require creating a separate user account. Before creating an account, read the service’s privacy policy, because most applications do not process personal data in accordance with European data protection regulations. 

Examples of AI applications for scientific information searhing

Web of Science Research Assistant (Requires login with HAMK credentials)

• Web of Science is a multidisciplinary database. The Research Assistant is an additional feature provided for HAMK students and staff.
• The Research Assistant consists of several agents that help you to formulate a research question, search for scientific articles, and find scientific publications.

Google Scholar Labs

• Google Scholar is a free search engine designed for finding scientific information.
• The service includes Labs AI assistant that allows you to search for information using natural language questions.
• The search results appear as a list of documents, each with an AI-generated summary. The summary aims to answer the given question.

Consensus 

• You can search for information by writing a research question in natural language. 
• It produces a summary of the found sources. The summary includes links to the sources. 
• You can limit the search to peer-reviewed sources only. 
• You can try it without creating your own account, with limited free usage. 

Elicit 

• The search works with research questions, keywords, or pdf files. Do not input a research article written by someone else without permission. 
• It summarizes articles and allows you to ask questions about the articles. 
• A mandatory account creation, limited free usage. 

Keenious 

• Search for information using research questions or a pdf file. 
• You can try it without creating an account. Free registration provides additional features. 
• Works within the EU. Keenious does not store entered texts. 

Litmaps  

• The search starts with a single article, which you can find using keywords or a DOI identifier.  
• It creates a visual map based on the article, allowing you to explore the citations between articles.  
• You can try it without creating your own account. Free registration provides additional features. 

Research Rabbit 

• The search works with keywords, author names, DOI addresses, etc.  
• It creates a visual map based on the selected articles, allowing you to explore the citations between the articles. It suggests similar articles.  
• Account creation is mandatory. 

Author: Minna Kivinen