Artificial intelligence applications can be used in scientific information retrieval at various stages. It is particularly suitable for the planning phase of information retrieval and for actual search in occasions that require only a small number of references. AI has also been integrated into many existing information retrieval tools.
At least for now, AI cannot generate extensive searches from diverse sources with comprehensive coverage of references. The answers produced by AI must always be checked by yourself, as is often the case with other uses of AI too.
Prompt and discuss:
Utilize the conversation option of AI applications in your information search. You can ask follow-up questions to the application either on your own or by choosing from ready-made questions. Through conversation, you often get more precise answers and a better overall understanding of the topic. Test various terms in your prompts and try different conversation starters.
Follow the rules:
It is important to follow the guidelines provided for AI use in studies and research. If the use of AI is permitted, it is generally recommended to indicate how it has been used in the work. Check out the AI policy of the UEF in UEF intranet, as well as Ai policy for students at the University of Eastern Finland.
Remember data security! Do not provide personal or confidential information to AI. See the UEF guidelines for more information.
Information retrieval from scientific databases
Traditional library and literature reference databases have been enhanced with AI tools. In UEF Primo and the Web of Science database the tool is called Research Assistant, and in the Scopus database it is called Scopus AI.
Links to these AI tools are in databases' own interfaces. Open Scopus and Web of Science through UEF-Primo, so that access right holds.
These database tools utilize both external language models and their own technologies.
The two tools have similarities:
- Search is done in natural language.
- Search can be performed in different languages.
- The search is first translated into English using AI, so it is advisable to check the translation.
- Follow-up questions are in the same language as the original question.
- The program retrieves a couple of suitable sources from the database's own materials and uses them to create an answer to the question.
- The answer includes references to the sources used.
- The search can be continued with ready-made, often more elaborating questions, or you can ask your own follow-up questions.
The UEF Primo AI Research Assistant can only be used by logging into Primo with UEF credentials (uefians) or with a library card (other users).
The program does not share users' data or input to the third parties. It also doesn't store and use personal data to train the language models, unless a user agrees to this when asked. However, data security is worth remembering in this program as well.
Searching
To start a search, type a question about a topic in the language of your choice. A question prompt works better than single terms without context.
The AI Research Assistant uses the UEF Primo international articles search group to find sources. Although the search group is named international articles, the sources may also include some books and other publications. It is possible to limit your search to articles only, books only or to peer-reviewed materials.
After providing an answer, the program offers ready-made follow-up questions. You cannot continue the conversation with your own questions that refer to previous ones.
Summary
The program summarizes the topic using five results it finds as well as language models. If you are not satisfied with the answer, you can perform the search again.
You can also rate the answer. In this case your search data is delivered further to train the AI model.
Additional Information
In addition to the five sources used in a summary, it is possible to perform a pre-formatted search on the topic, which displays all the results found from the international articles search group. The search query is visible to the user. The query is often quite broad, so it may result a lot of references.
Saving Searches
The program saves the searches made, and you can return to them later. The maximum number of saved searches is 200. You can delete search questions from the search history yourself.
The data entered into the program is not stored in the system, but can be used to enhance the program. Therefore, data security is worth remembering in this program as well.
Searching
The program first formulates a natural language question in English from the user’s prompt and then usually converts the question into a search query, too. The query includes alternative terms.
The search query is visible to the user, so they can check if the key terms are included.
The query is not the only way to retrieve documents from the database, the program uses other techniques, too.
You can also ask Scopus AI for issues other than actual research questions, such as search terms for a specific concept.
Summaries
The answer is based on Scopus materials from 2003 onward. Scopus uses around ten sources, sometimes fewer, sometimes even more. The selection of sources is based on the text of the abstracts, where relevancy, recency, number of citations and especially semantic similarity to the question are considered.
The answer includes actually two different summaries. The other, called expanded summary, gives a more detailed answer, and uses more sources, up to 20.
Additional Information
The program presents a concept map on the topic. The concept map provides detailed information on individual concepts related to the topic. The description of a concept includes references to sources.
The program identifies three key researchers in the field based on their research interest and the number of articles and citations.
The program lists key publications related to the topic, which are selected based on citation counts. These publications can be older than 2003.
In addition, the programme analyses publications related to the topic and analyses already consistent as well as emerging themes of research from them.
Follow-up questions
You can use ready-made questions to explore the topic deeper or you can ask your own follow-up questions.
Saving Searches
If you are logged in to your own Scopus account, the program saves your questions and summaries. This feature can be turned off selecting 'Temporary conversation' on the upper right corner of the screen. You can also delete single conversations from the history.
Read more about program’s operating principles and data security.
Read detailed description about responses produced by the program.
The program does not share data with third parties and does not use it to train the model. However, data security is worth remembering in this program as well.
Searching
The search can be started with a direct question. Another way is to conduct a more guided search with 'Understand a topic'.
The program performs a search using the key terms of the question, alternative terms are included. The number of results can be seen and it is possible to browse the results.
The prompted question can be phrased to look for certain types of publications, such as the latest or the most cited articles.
Another guided search is 'Literature review' where the program asks certain clarifying questions from you and presents different alternatives before the actual search.
'Find a journal' helps you to identify key journals on the topic, not individual articles.
Summaries
The answer is based on the entire Web of Science Core Collection. If you want to emphasize newer sources, it is advisable to mention in the question “… based on recent documents …”. Maximum of eight sources are used to compile the answer. The selection of sources is based on the text of the abstracts, where relevancy, number of citations and semantic similarity to the question are considerated.
In addition to the actual sources used, you can view the entire original search result list. The search query used to retrieve the results is stated at the beginning of the list.
The answer to the question is given in the same language as the original question.
Additional Information
More detailed information is given on the main topic derived from the subject - not precisely related to the original question itself.
The program presents a concept map on the topic. The concept map lists individual concepts related to the main topic. Links of the map provide a ready-made, usually very long list of sources related to each concept.
The program also indicates the top researchers. These most cited researchers are outlined briefly. In addition, it is possible to view the most cited articles of the topic and to examine how much has been published on the topic in the course of time.
Follow-up questions
You can use ready-made questions to explore the topic deeper or you can ask your own follow-up questions.
Saving Searches
Research Assistant shows all searches made during the session, and you can go back to review them. If you have a Web of Science account, you can also save searches for later use.
Watch a demonstration and instructions of the program.
Read more about program’s operating principles and see more examples of search questions.
Searching articles with AI
A new feature has been added to some databases: search methods that enable traditional information retrieval, such as searching for articles and other publications, with the assistance of Ai. In practice, this means that
- the user performs a search using natural language
- the program responds with a list of suitable literature sources
- filters for refining the search results are available
- the program does not provide a summary of the topic in question.
- follow-up conversation is not possible, nor is it possible to modify the “search,” except by submitting a new query
This type of search is suitable for quick information retrieval or for users who are unfamiliar with search techniques such as Boolean operators. The results are not as precise or comprehensive as those obtained through query-based searches.
Natural language search is available under the Basic search tab. Click to activate the ‘Natural language search’ option.
Searches can be performed in various languages.
You can view the query used by the system via the ‘Show refined query’ link.
Search results can be filtered using the filters available at the top, just below the search query.
Activate the search by clicking the ‘Smart search’ button in the top right corner of the page. Later in 2025, this search method will be more prominently integrated into the Web of Science interface.
This search retrieves both publications and researchers.
At the moment searches can only be performed in English.
The search is based on both a system-generated query and semantic similarity. The system does not display the query it uses.
In addition to the combined results, you can choose to display only the search results from either semantic search or Boolean query search.
Search results can be filtered with existing Web of Science filters and analysed with Web of Science citation tools.
Other AI applications for different stages of information retrieval
Primarily, it is recommended to use AI applications acquired by UEF, the provider of which has an agreement on the processing of personal data.
Along with UEF Primo, Scopus and Web of Science AI tools, such a suitable application is Microsoft 365 Copilot Enterprise-version, which needs to be logged in through your UEF Microsoft account. Use the address: https://copilot.cloud.microsoft. Link to Copilot can be seen in UEF Intranet waffle menu and inside Microsoft software (like Teams etc.) You can load M365 app into your mobile phone. UEF Intra provides more instructions about using Copilot.
With the help of AI-generated answers, you can quickly familiarize yourself with a new topic: what it is about, what kind of aspects are involved. AI is also a good tool in defining concepts. Read more about exploring the topic and concepts in the Guide to Information Retrieval.
You can proceed in two ways. In both cases, be prepared for the possibility that AI may “hallucinate,” i.e., produce inaccurate text.
1. Start by asking AI about the topic. The answer will help you get an idea of what the topic includes and where to head your information retrieval. Note that if the topic is completely unfamiliar to you, it is difficult to assess the accuracy of the AI-generated answer. Therefore, verify the information from proper scientific sources.
Examples of how to start conversation:
- “What means microaggression?”
- “What kind of actions can be done to prevent eutrophication of waters?”
2. Familiarize yourself with the topic elsewhere first, for example, by reading research literature, and then ask AI for more specific questions. When you are already more familiar with the topic, you will be better able to assess the accuracy of the answers. You can also get more precise or more relevant answers when you include concepts and terminology that accurately describe your research topic in the prompt.
Examples of how to start conversation:
- “How can teachers promote a positive climate at elementary schools through inclusive classroom activities?”
- “I'm in the process of brainstorming the subject of my thesis. I am familiar with Lev Vygotsky's theory of the zone of proximal development through research literature. In my thesis, I want to compare Vygotsky's theory to some other theory of learning. List five learning theories that I can possibly use as a reference in my thesis.”
Examples of suitable AI applications:
- Copilot (Microsoft) – recommended
- ChatGPT 4o-versions (OpenAI)
- Gemini (Google)
Keywords:
AI applications are already quite good at suggesting keywords for information retrieval, especially in English. Applications are capable of finding word equivalents and synonyms in different languages. You can ask terms from a specific thesaurus, too.
However, they are not very good at formulating keywords in such a way that they work effectively when searching for information in databases specialized in scientific publications. The application might present rather long expressions, for example, ‘dietary effects on gut microbial diversity’ or ‘the importance of sleep for memory function’. These need to be broken down into simple, separate terms, e.g., ‘diet - gut microbes - diversity’ or ‘sleep - memory’.
Keyword lists are rarely comprehensive, at least not on the first try. You can ask the application to provide more terms and refer to the types of words you wish for. You can request plenty of keywords, from which you can choose the best ones. To get a more comprehensive view, you can try multiple AI applications.
Queries:
The best AI applications are also able to formulate ready-made search queries for databases: keywords are combined with operators and phrases are marked properly. However, word truncation is usually lacking and the formulation of individual keywords is unfinished. You can ask for word truncation, but it is likely not to be optimal without detailed prompting.
Always evaluate yourself which keywords are actually useful for your topic. Also, check the logic and functionality of the search query. In the picture you see a correct query model and points to pay attention to.
Examples of how to start conversation:
- “Can you find synonyms for the term sensory defensiveness?”
- “Tell me keywords for information search about the topic how the use of wood affects fine particle emissions.”
- “If there are two main concepts: environmental effects and vehicles, can you give me synonyms and related search terms for these both in a table form, where concepts are in columns.”
- “Use Boolean logic to find publications about eating disorders and young people. Use several synonyms for both concepts.”
Examples of suitable AI applications:
ScopusAI - recommended
Copilot (Microsoft) – recommended
ChatGPT 4o-version onwards (OpenAI)
Gemini (Google)
Perplexity
FintoAI is an application specializing in keywords. It suggests words based on YSO thesaurus in Finnish, Swedish or English. FintoAI is given a text as input, e.g. a summary describing the topic.
When searching for information, it is advisable to first distinguish between applications that do not search for information themselves but base their entire response on a pre-trained large language model. An example of this is the basic version of ChatGPT, first published in 2022, which is neither a database nor a search engine.
Many other applications utilizing language models perform the task using genuine sources:
Applications utilizing generative AI
Choose an application, where genuine information retrieval is done from real sources. The application uses these to generate an answer. Depending on the application, the search can go through the open web sources (e.g. Copilot and ChatGPT 4o) or a specific database containing selected information, such as open access scientific articles (e.g. Elicit and Consensus). The result of the search is a short answer to the question or a summary of the topic, along with the sources used in the answer.
Formulate the question clearly but concisely and include sufficiently precise terms to get a relevant answer. Partition a complex task into smaller parts.
The answer produced by AI is often based on a very small number of documents. There is little information on how the application selects the sources it uses, which undermines the reliability of the answer. Therefore, check the original sources which the AI refers to. Also, do your own search from other sources to get a more complete picture.
Note that AI does not always act the way it is asked of. For example, even if you ask the AI application to base its response only on scientific or peer-reviewed articles, it may not necessarily do so.
In chatting with the AI, you can, for example, ask the application to modify the response in a certain way or request more specific information. Through a dialogue you often get a better answer than by asking just once.
Examples of how to start conversation:
- “What are the key challenges in addressing environmental exposure in the Global South?”
- “What is the role of internal communication in organization change, based by articles in peer-reviewed journals?”
- “I am writing an academic essay about eating disorders in young adults and supporting the healing process. Could you suggest academic articles? Try to find peer-reviewed publications.”
Examples of suitable AI applications:
- Copilot (Microsoft) – recommended
- ChatGPT 4o-versions (OpenAI)
- Elicit
- Consensus
AI is used in search engines in ways other than based on generative language models, too:
Applications structuring and visualizing search results
In these search engines, information retrieval is proceeded in the traditional way, with search queries. The mentioned database contains open access scientific publications, such as articles, making it reliable source of information. The application uses AI to group the search results, making it easier to evaluate the themes related to the topic. The result of the search is a list of references.
Example of a suitable application:
- Open Knowledge Maps
Search by existing article
Some applications use an existing text or document suitable for the topic, such as an article or an abstract, as a starting point. The application analyses the text and searches for similar references. The result of the search is thus a list of references.
Examples of suitable applications:
- Connected Papers
- Elicit
- Keenious
- Research Rabbit
Read more about open access search engines, which are speciliazed to scientific information, from Guide for open publications searching.
Many traditional databases (e.g., Scopus, Web of Science) contain a similar function that suggests further references based on the selected article.
Different referencing styles have their own general rules how to cite AI. E.g. APA style for ChatGPT in-text citation is: (Open AI, 2024) and reference list entry:
Open AI (2024). ChatGPT (May 13 version) [Large language model]. https://chat.openai.com/chat.
Read more:
How to cite ChatGPT? (APA)
How do I cite generative AI in MLA style?
The Chicago Manual Style / Q&A
Nevertheless, always follow the guidelines of your own discipline or publisher regarding the use and reporting of AI and citing to it.
Limitations of AI in information retrieval
Restricted sources:
When using an AI application that genuinely retrieves existing sources, the answer produced by AI is often based on a very small number of documents. There is no exact information on how the application selects the very sources it uses, which may weaken the reliability of the answer. AI applications can reach only open access materials. Some of the most relevant sources may be missing altogether. Therefore, check the original sources to which the AI refers. Has AI been able to create right conclusions? Pay attention to general applicability and up-to-dateness of the references, too.
Also, conduct your own search in actual scientific databases to get a more comprehensive view. Many general AI applications can utilise open access publications only, while documents behind the pay-wall in databases cannot be used.
The quality of results may vary:
AI applications, including those using genuine publications as sources, can produce incorrect or biased responses. The answer is also not always exactly the same, even if you ask the same question.
User accounts and charges:
Many AI applications can be used for free, but their capacity and features are limited. Payments provide more services. Even free versions often require the creation a user account.
Privacy:
From a privacy perspective, there may be problems with the use of AI applications. Applications can store personal information and conversations, which can be used for the tool’s own purposes and possibly transferred further. Therefore it is better to use UEF licence to Microsoft 365 Copilot, which does not share information outside.
Copyright:
AI-generated outputs do not have copyright. Instead, the AI application itself may infringe copyright when using material found online without permission. Many publishers and content providers already prohibit the direct input of their contents (such as articles, etc.) into AI applications, as well as the use of their materials for the development and training of AI applications.
Page last updated: 12.8.2025.