E-learning material

Wikipedia: a starting point for an overview of a topic.

Never cite a Wikipedia article in an academic paper. Check the references below the entry; they will help you explore the topic.

Google: use it to search for news and view companies’ and organisations’ websites.

Google Scholar (GS): as its name suggests, Google Scholar searches academic sources, but it also finds predatory publications. Predatory publications are not peer-reviewed before publication, so you don’t know if the content is correct. For lesser-known authors and publishers, it is worth carrying out a source evaluation. For more information, see the Source evaluation module.

In GS, you can set the full-text content subscribed to by the Corvinus Library to be displayed. To achieve this, follow these steps:

1. GS Settings
2. Library links, search for Corvinus,

3. tick and save.

After deleting cookies, set it up again. Subscribed content will also appear using the VPN.

Library resources contain only reliable academic documents. You can use three searching interfaces, depending on whether you want a specific document or topic. Clicking on the small  icon next to the search box will also gives you this hint.

1. Figure – Search boxes on the webpage of the Library of Corvinus University of Budapest

OPAC – Open Public Access Catalogue

The OPAC mainly contains the library’s physical holdings (printed books and periodicals), but there are also a small number of electronic periodicals. If a printed book has an electronic version, this is also indicated in the OPAC.

A record in the OPAC tells you how many copies of a given title are available, its location (Reading room and shelf), borrowability and availability.  We have books in a closed storage area, they are also borrowable.

The loan period is 2 or 4 weeks, depending on the book, and only those books are borrowable that have a white stripe on the spine, even together with a yellow one. If you only find a yellow, that book can be used only on site. 8 books can be borrowed at a time; use the self-check in/out machines to take them away. Loan renewal is possible 3 times if there are no reservations on the title.

To make reservations, renew your books remotely or request them from the stacks log in to the catalogue. To use these advanced functions, you need your Library card and your Neptun ID.

For information on renewals, reservations, overdue fines, and lost book replacement, please visit this page.

Journals and E-book search platform

In the Journals and Ebooks search, you can find full-text electronic journals and e-books and the bibliographic data of print journals. Use it when looking for a specific online journal, an article, an electronic book, or a chapter. You can also access these remotely using VPN. Always start your search with the title of the journal or the e-book.

Once you have found the journal/e-book you are looking for, under Full Text access you can check its full-text availability (database, period) e.g.:

2. Figure – result list in the e-journal search platform

SuperSearch

 SuperSearch is recommended for topic searches. It simultaneously searches the catalogue, subscribed and free databases containing scholarly literature and several other free, relevant sources. It contains mainly English-language literature. You can refine your search by entering more keywords, using logical operators and other search techniques, and narrowing down to specific fields. After the search, you can save relevant documents.

Databases containing financial, statistical data or legal sources (e.g. Refinitiv, Statista, Jogtár) are not included in SuperSearch; check them separately from the thematic list of databases. For further information, check the Databases section.

Steps for an effective topic search:

• • creating the topic sentence (a sentence containing the most important keywords of the topic)
  • selecting keywords and their synonyms
  • combining keywords with logical operators, using other search techniques
  • filtering to specific fields: our recommendation is the abstract, since if a keyword appears in it, it certainly will be relevant
  • saving the results to My Dashboard or to reference management software. For more details, see the Reference Managers

Video

Topic Search process in SuperSearch

Game
Click on the following game to test your knowledge.

Topic Search order

Search tips for advanced users
There are additional filtering options in All filters (SuperSearch), such as language filtering or suggested subjects.

The difference between a subject and a keyword is that a keyword can be anything, while a subject is a predefined, more precise, narrower term, usually added to the document by experts.

If you get stuck in a search, try the snowball method: find an article among the results that is relevant and look at its references. This way, you may come across older articles, but you can at least get started with your search.

Repositories

The repositories are a collection of publications produced at the University of Corvinus. We have five different repos, collectively called the Corviversum.

One contains Corvinus research papers, another TDKs and MA/BA theses, while the 3^rd one PhD dissertations. In addition, we have two historical repositories, one for university history and the other for the digital collection of Hungarian economic history.

If you are looking for a thesis topic, see previous theses written in the same field to get inspiration, or use their references as a starting point, to avoid writing on a topic that has been written about before. You can also check the repository of lecturers’ publications to see who is researching the field you are interested in. Later on, it can help you choose a supervisor.

Databases

In our databases, you will find authentic, reliable sources, no predatory journals, and you do not need to do source evaluation. Most of them are searchable in one or both search interfaces (Journal/E-book and SuperSearch), look for the symbols next to their names in the list:

Searchable in SuperSearch:

Package content is available in Ebooks and Journals:

If you need statistical, financial or legal data, the thematic breakdown, under the alphabetical list, can help. All our databases can be accessed remotely using VPN.

EduID access also works for some databases, and some require an individual account (e.g. Workspace, Orbis). For content and non-standard access information, see the description of each database.

EconBiz

EconBiz is the online platform of the Leibniz Information Centre of Economics (ZBW). Due to a contract between Corvinus and ZBW, you can access ZBW’s OA collection through their platform. It offers millions of online sources in the field of economics and related sciences. Their collection is available through their platform, from the database list or from SuperSearch’ Supplemental materials.

Their activity goes beyond offering online publications, their main aim is knowledge transfer. For more information, check their homepage.

Game
Click on the following game to test your knowledge.

Where to search for the following?

• it indicates your knowledge about the subject, showing that you have found relevant authors and works on the topic and that you are reflecting on them;
• it supports your ideas and thoughts;
• it makes the literature you have used traceable for others;
• you can avoid ethical misconduct and plagiarism.

Game

Click on the game below to test your knowledge.

References – True of False?

There are two basic ways to cite sources: direct quotation and paraphrasing (quoting the content). In both cases, you must provide the source of the quoted text.

Direct quotations can be used when quoting definitions, apt phrases, legal provisions or interview excerpts. Do not quote more than 2-3 sentences verbatim; the APA standard sets this limit at 40 words.[1]

Example

Some hints for direct quotations:

• If you omit something from the quoted text, mark it with three dots in square brackets, but be careful not to distort the original chain of thought.

“Be prepared to spend [at least] half a day in the […] library.”

• If the quotation does not begin at the beginning of the original sentence, place three dots after the quotation marks and begin the text in lowercase letters.

“…based on the survey, two-thirds of the population is affected.”

• If you modify a definition, figure or table, indicate this by adding “based on” or “using” after the reference.

Figure X: Figure title, (other data) based on own editing

• In the case of oral communication, the name of the interviewed person must also be indicated, unless requested otherwise. Avoid using unverifiable references too often. For more tips, see our guide on referencing interviews.

The formal elements of a verbatim reference are quotation marks in the text and the mandatory information (author’s surname, year, page number).

“An in-text citation is a short acknowledgement you include whenever you quote or take information from a source in academic writing. It points the reader to the source so they can see where you got your information” (Jackson, 2005, p. 16).

Paraphrase: A paraphrase is a concise summary of a longer text in your own words. In the case of a paraphrase, the source data of the work referred to must be given in the same way as in the case of a direct quotation.

The formal elements of a paraphrase are the author’s surname and the year of publication.

Example

Original text:

“What is the role of digital media in contentious politics? On the one hand, digital media plays a central role in informing the public and organizing political movements. On the other hand, it has become a valuable tool for digital repression in authoritarian states. This study concentrates on the patterns of digital media use by pro-government actors in times of nationwide protests in autocracies. It analyzes how pro-government actors establish control over political discourse and information flow online compared to pro-opposition and neutral actors.”

The work cited: Richmond, S. (2018). Broadcasting Messages via Telegram. Political Communication, 41(4), 509–530.

Paraphrase from the original text:

This study examines the role of digital media in contentious politics, focusing on pro-government actors’ use during nationwide protests in autocracies and analyzing how they control political discourse and information flow compared to neutral actors (Richmond, 2018).

We follow the APA standard in the examples.

An in-text citation is only an allusion. When referencing, make it clear where the statement you are supporting begins and ends. There are two types of in-text references: parenthetical and narrative references.

• Parenthetical references: they come at the end of the quotation, indicate the author’s surname, the year and, in the case of a verbatim quotation, the page number in parentheses.

Example

“Heavy social media use can be linked to depression and other mental disorders in adolescents” (Taylor, 2021, p. 43).

• In the case of narrative references, the reference is embedded in the text, and the elements of the reference are separated. The author and the year are included in the text, followed by the page number (in the case of a direct quotation) at the end of the sentence.

Example

Taylor (2021) found that “heavy social media use can be linked to depression and other mental disorders in adolescents” (p. 43).

The reference list is a list of works that you refer to verbatim or as a paraphrase in the main text of your thesis.

Although the reference list is placed at the end of the thesis, reviewers usually read it first. This allows them to see whether you have found the most important works on the topic and how well-informed you are on the subject.

The basic formal features of a reference list are as follows:

• • there is no numbering,
  • it starts on a new page,
  • works are listed alphabetically by the author’s surname; if there is no author or editor, use the title.

See this guide for examples of how to use in-text citations and the reference list in a continuous text.

In this menu item, you can record the data elements of the documents, from which Word can generate in-text references and a list of references in several different styles.

Since each data element must be entered manually, you need to know exactly what type of document you are working with, which category each data element belongs to, and how it should be added. Without this information, the reference will be incorrect.

Limitations: the number of available reference styles is limited, and it might not include the latest version of the style, e.g. in the case of APA, it offers version 6 instead of 7 (January 2026).

Another problem is that it does not handle multi-author works correctly in APA style: according to the rules, if there are more than 2 authors, the first author must be indicated, and the other authors must be referred to as et al., but according to the test, all three authors were listed:

(Smith, Stuart, & John, 2025)  – This is incorrect according to both APA versions 6 and 7.

Correct version: (Smith, et al., 2025)

Reference generators are a transition between manual recording in Word and reference management software. The advantage is that you do not have to record each data element individually, and it usually automatically recognises the type of document you are working with.

Below, we present a few reference generators, highlighting their advantages and disadvantages.

Other well-known citation generators include BibMe, CiteThisForMe, and EasyBib, each with its own features and limitations.

It is essential always to check the generated reference and modify it if necessary. For longer-term, more in-depth research, it is better to use some reference management software.

Compared to the tools presented above, citation management software has more features, of which the following are essential:

• collect, store and organise the sources used
• prepare in-text references (in a form that can be inserted while writing) and the list of references
• handle different citation styles.

Although none of them work with 100% accuracy, their use significantly reduces the time spent on managing references.

1. Figure: The diversity of reference management software

There are countless reference management software programmes available, some of which are free to use and others that require a subscription. Among the best-known free ones are Zotero and Mendeley, while EndNote Desktop is a subscription-based version. The web version of EndNote, which is part of the subscribed Web of Science database, is available free of charge during university studies (on the university network or remotely via VPN).

	EndNote Web	Mendeley	Zotero
Price	Free with WoS database subscription	Free	Free
Access	web	web version, synchronised with desktop version	desktop version, synchronised with web version
Storage capacity	2 GB	2 GB	300 MB
Operating system	Mac, Windows	Mac, Windows, Linux	Mac, Windows, Linux
Service provider	Clarivate	Elsevier	Open source
Web and Word plugin	Yes	Yes	Yes
PDF editor	No	Built-in	Built-in

1. Table: Comparison of basic features of free reference management software

General features:

• Using a browser extension, document data (including full text in the case of subscribed or OA sources) can be saved from databases (e.g. JSTOR, Scopus, etc.), search engine results (e.g. Scholar), and random web pages.
• A plugin helps to insert in-text references (Zotero, EndNote Web – Cite while you write, Mendeley Cite), which typically works in several types of text editors (Word, LibreOffice, Google Docs). The plugin can be used to manage single- and multi-author works, insert page numbers, and enter data from edited volumes, both for narrative and parenthetical references.
• Data can also be entered manually for printed documents or using some other identifiers, such as an ISBN or a DOI for electronic records.
• Notes and tags can be added to saved documents; these, as well as highlights, notes, and annotations made in PDFs, are also stored by the software.
• Saved data can be transferred from one reference manager to another in .ris format, but it is advisable to commit to one reference manager from the outset to avoid data loss.
• Service providers support users with direct assistance and/or training materials.

The most well-known reference manager software

Below is a brief introduction to the most well-known reference manager software, highlighting its significant features and unique functions.

The citation style determines the mandatory elements of the citation and their format. The number of citation styles available in the software varies. Zotero and Mendeley support more than 10,000 styles, while EndNote Web supports approximately 50 styles.

Different styles exist for various fields of science. In the social sciences (including economics), APA is widely used; in law, Oxford; in the humanities, Chicago; and in literature and the arts, MLA. In addition, some journals may require their own style.

At present, there is no mandatory style at Corvinus University; the Study and Examination Regulations recommend APA. During university work, the lecturer/supervisor/department may determine which style to use; if no such regulation is made, we recommend that you use APA. As a basic rule, the same style should be used throughout a work. In the reference management software, styles can be changed freely with a single click, even in the finished work.

There is usually a separate website or manual for each citation style. In the case of APA style, the latest (7th) version is available in the Library, and the previous version is available free of charge on the official APA website. If a rule has been changed, it is indicated next to the entry. This usually occurs with newer types of documents (such as blogs, online media, etc.).

Reference management software can only generate correct references if the data it saves or enters is correct. When documents are saved from various sources (repositories, OA sites, search engines), the data may be inaccurate or incomplete. These can be edited after saving, even if they have already been used in the work.

We have also seen examples when reference management software automatically fills in missing page numbers for electronic articles or misidentifies the document type. For this reason, you should not unquestioningly trust the version generated by the reference manager.

Another problem is that the default language of the reference manager software is English, so Hungarian authors’ names are also rendered in English. Because of that, the comma follows the surname also in Hungarian names, and, in some cases, double-letter first names are truncated (Cs becomes C). Currently, there is no Hungarian APA standard, so this anomaly can only be resolved manually.

Some academic sources provide examples of the reference next to the article, but be careful, they might be incorrect, as in the example below.

Foreword to the thematic issue of the Tourism Bulletin “I. Tokaj-hegyalja Tourism Conference”

The mistake is that in APA style, there is no need for the phrase “Retrieved from” – just the URL:

The correct version is:

Rozgonyiné Drotár, N. (2025). Előszó a Turizmus Bulletin „I. Tokaj-hegyaljai Turisztikai Konferencia” tematikus számhoz. Turizmus Bulletin, 25(3), 4. https://journals.lib.uni-corvinus.hu/index.php/turizmusbulletin/article/view/1930

To achieve better results with the reference manager software, you need to know the type of document you are referencing and be familiar with the referencing style.

The CRAAP test is a widely adopted assessment method in academia developed by an American librarian to evaluate the reliability and credibility of information sources. It provides you with a methodology and a list of questions to assess the quality, relevance, and value of the information that you have found. Applying the test helps researchers critically evaluate a source’s reliability for research purposes by asking questions about the publications they find. The CRAAP test helps to systematically filter out unreliable or inappropriate sources and ensures the robustness of academic work.

If you can add a positive answer to each question, the source is probably appropriate and acceptable to use. If a source fails in more than one indicator of the test, you are advised to search for more suitable option(s).

1. Figure – CRAAP test (own edition)

In rapidly evolving disciplines, it is crucial and essential to acknowledge that older information may become outdated or no longer relevant. When selecting a source, ensure it directly addresses your research question and corresponds with your academic or personal objectives and goals. Evaluating the authority of the source is crucial in determining whether the author possesses the necessary expertise to provide reliable information on the subject. Furthermore, verifying the truthfulness of the content is crucial to prevent dependence on false or misleading information. Finally, understanding the purpose behind the source helps reveal whether it is presented objectively or is intended to influence or manipulate the reader.

Here are some examples and guidelines about using the CRAAP test:

The Australian Education Research Organisation

Meriam Library at California State University

South Central College

GAME!
Click on the game to test your knowledge.

CRAAP test

Both predatory publishers and journals exploit the open-access publishing model by charging authors fees to publish their work without providing legitimate peer review, proper editorial oversight, and professional rigour, which are essential for maintaining research quality. These publishers and journals prioritize profit over academic integrity, often accepting and publishing manuscripts regardless of quality or validity.

Predatory publishing is dangerous because it undermines the credibility of scientific research, disseminates potentially flawed or false information, and can damage the reputation of researchers who unknowingly publish with them. It also wastes researchers’ time, effort, and funding, and can hinder academic progress by flooding the literature with low-quality or unreliable studies.

Typical signs of predatory journals:

• Aggressive and unsolicited emails inviting you to submit papers.
• Promises of very fast publication with little or no peer review.
• Lack of clear information about editorial board members or fake editorial boards.
• Poorly maintained websites with spelling or grammatical errors.
• No transparent information about fees or hidden high publication charges.
• Lack of indexing in reputable databases (like Scopus, Web of Science, DOAJ).
• Fake impact factors or misleading metrics.

Why are they dangerous?

• Your research may be published without proper quality checks, harming your reputation.
• It can affect the credibility and dissemination of your work.
• Predatory journals may not be recognized by academic institutions, limiting career advancement or funding opportunities.
• They contribute to the spread of unreliable or low-quality research.

How to avoid publishing in them:

• Check if the journal is indexed in trusted databases relevant to your field.
• Verify the editorial board members and their affiliations.
• Look for clear peer review policies and publication ethics statements.
• Use established journal databases or directories like DOAJ.
• Consult with mentors, colleagues, or your institution’s library for recommendations.
• Be wary of journals that contact you out of the blue with publication offers.
• Use tools like “Think.Check. Submit.” to assess journals before submitting.

Find more information on predatory journals and publishers on the University Library website.

A list of potential predatory open-access publishers created by a librarian, Jeffrey Beall. The list was intended to identify open-access publishers that did not conduct genuine peer review or a proper publishing process, even though they charged authors an article processing charge (APC) to publish their manuscript.

SCImago Journal and country rank

Directory of Open Access Journals (DOAJ)

ISSN Portal

Think.Check.Submit.

GAME!
Click on the games to test your knowledge.

Wordsearch – source evaluation

Hangman – source evaluation

Would you cite this source?

• The University encourages the careful, ethical, and transparent use of generative AI (GAI) in learning and research.
• Academic honesty, respect for intellectual property, data protection, and confidentiality are fundamental values at Corvinus.
• Even when you use AI tools, you remain fully responsible for everything you submit, publish, or share.

• Any content produced with the help of AI must be checked carefully for accuracy, legality, and compliance with ethical standards. If AI-generated content causes harm or contains errors, responsibility lies with the user, not the tool.
• If AI contributes to your academic work, this must be clearly disclosed, including the specific use case and the name of the tool

The following practices are prohibited:

1.      Claiming AI work as your own

1. • Using AI to write parts of an assignment or thesis (e.g. abstract, literature review, figures) and submitting them as if they were entirely your own work.

2.      AI-assisted plagiarism

Using AI to rewrite or adapt someone else’s work and presenting it as your own.

3.      Presenting AI-generated data as real research results

Treating AI-generated statistics or datasets as valid research findings (except in clearly labelled teaching or demonstration exercises).

4.      Violating confidentiality or data protection

Entering into AI tools: confidential university information, trade secrets, unpublished research data, personal data of students, staff, or others,
unless this is explicitly authorised and institutionally protected.

5.      Infringing intellectual property rights

Using AI in ways that may violate copyright, patent, or trademark rights.

6.      Creating harmful or illegal content

Generating biased, discriminatory, misleading, or illegal outputs, including malicious software or content facilitating cybercrime.

You may use generative AI responsibly and transparently in the following ways:

1.      Editing your own text

Improving grammar, spelling, clarity, or style, as long as the ideas and arguments remain your own.

2.      Supporting research work

Helping with finding and organising literature, preliminary data analysis, reviewing datasets, as long as you critically evaluate and verify all outputs yourself.

3.      Working together with AI

Co-creating content with AI only when this is explicitly allowed, and only if the AI’s role is clearly acknowledged.

4.      Using approved AI tools

Prefer AI tools provided by the University or explicitly approved by your lecturer or programme leader. New or significant tools may require review by the Dean for Artificial Intelligence.

• If AI is used in your assignment, thesis, or project, you must include a written declaration stating whether AI was used, how it was used, which AI system and version were applied.
• Suspected unauthorised AI use may lead to requests for clarification, ethics procedures, application of plagiarism or academic misconduct rules.
• AI-detection tools may be used by teachers, but they are only supportive tools, not definitive proof on their own.

AI can help you learn and research more effectively, but it cannot replace your own thinking, responsibility, or academic honesty. Always be transparent, always review what AI produces, and always consult with your teachers to determine if and to what extent AI use is permitted.

Tool examples: Perplexity, Consensus

Use case: Rapid identification of relevant academic literature, empirical evidence, and cross-disciplinary, evidence-based answers, supported by transparent and verifiable sources that enable users to independently assess validity and reliability.

How it works:
Perplexity Academic combines large language models with live web and academic database browsing, returning synthesised answers with direct citations. Perplexity is built on a freemium AI-powered answer engine (a model in which a service offers basic features for free, while advanced functionality is available only through a paid subscription.) that integrates real-time web search with multiple leading LLMs to deliver clear, concise answers backed by verifiable sources, and each response includes citations to the original materials. Perplexity’s Academic filter prioritises peer-reviewed and scholarly content for research-oriented queries. Unlike traditional link-centric search engines, Perplexity’s output focuses on narrative summaries with embedded source links, designed to streamline both general and academic research tasks.

Consensus takes open-access research and aggregates answers based on the findings of studies. This enables researchers to move beyond simple keyword-based searches by providing fine-grained control over search parameters, including temporal scope, citation thresholds, and journal quality indicators (e.g., Q1–Q4 rankings), thereby improving the relevance of the retrieved results. Consensus searches across a database of over 220 million open-access papers, sourced from Semantic Scholar, OpenAlex, and its own scholarly web crawl, which covers high-impact journals and PubMed content, and grounds each answer in citable literature. It is often positioned as an AI-enhanced alternative to Google Scholar, combining evidence retrieval with AI-based synthesis. However, while every citation refers to a real published paper, the AI’s summaries can occasionally misinterpret a paper’s findings, representing the principal form of hallucination possible in the system.

Best practice:
Use Perplexity for broad, exploratory scanning and Consensus for validating whether a claim is supported by the scientific literature. Always verify cited sources directly. Use AI-powered search as a complement to traditional research methods.

Tool Example: Scite, Sci Space

Use Case: Quickly identify and retrieve relevant academic papers with AI-powered search, summarisation, and citation insights.
How it works: These platforms utilise LLMs and AI classification to efficiently rank and extract key findings; however, they typically require a paid licence to unlock their full potential, as free versions are limited in capability. SciSpace is an AI-enhanced research assistant that goes beyond basic search by providing tools for literature review, interactive PDF chat, AI-assisted writing, summarisation, and citation generation. It operates under a freemium model, in which the free version is limited in both usage volume and feature breadth. Additionally, Scite highlights how papers cite one another by categorising references as supporting, mentioning, or contrasting, thereby enabling researchers to assess the strength and reliability of the underlying evidence and to evaluate the relationships among relevant works for a more nuanced inclusion in their own literature review.
Best Practice: Use AI-powered search as a complement to traditional research methods. Combine it with domain-specific controlled vocabularies, curated databases, and manual screening to ensure comprehensive coverage, verify critical sources, and maintain rigour in literature selection.

Tool Example: Google NotebookLM
Use Case: Create an AI-assisted personal knowledge base that summarises documents, notes, and creates structured insights of these.
How it works: NotebookLM acts as a learning buddy, ingesting the researcher’s own source materials (such as PDFs and web links) and generating structured summaries, explanations, and presentations, including audio overviews, visual overviews, and mind maps, among other formats.
Best Practice: Use as a tool for project notes and multi-document synthesis (text-to-audio/video).

Tool Example: Gemini Gems or ChatGPT’s custom-built GPTs
Use Case: Build tailored AI agents (“Gems”) to handle specialised research tasks (e.g., summarisation, draft templates, prompt libraries).
How it works: Gemini Gems or Custom GPTs as reusable AI assistants that follow predefined instructions, examples, and constraints so they can reliably handle the same type of task again and again. In practice, this means that instead of repeatedly explaining what you want (for example, how to summarise articles, explain academic concepts, or give writing feedback), you configure the AI once, and it then automates those repeated activities consistently, improving efficiency and freeing up time for higher-level research work.
Best Practice: Define explicit task instructions and repeatedly test Gems and GPTs to achieve optimal performance.

Tool Example: Gemini / ChatGPT / Perplexity- Deep Research
Use Case: Perform complex query analysis such as cross-disciplinary context extraction, data comparison, and extensive summarization.
How it works: Unlike regular search engines, which mainly return lists of links based on matching keywords, Gemini, ChatGPT, and Perplexity Deep Researches work by gathering information from several sources, reading and comparing them, and then combining the key points into a single, structured explanation. Instead of simply pointing you to where information can be found, they bring together search, reasoning, and awareness of sources to produce longer, clearer, and more nuanced answers, which makes them especially useful when you want a deeper understanding of a topic rather than quick look-ups.
Best Practice: Use alongside traditional databases (e.g., Web of Science, JSTOR, ScienceDirect) to triangulate findings.

Tool Example: Litmaps, Connected Papers
Use Case: Map and explore connections between papers, authors, and topics; receive updates on emerging research.
How it works: AI algorithms visualise network graphs of literature and track new publications matching your interests, or visualise the citation tree of seminal works in your field.
Best Practice: Use visual overview early in literature review to identify relevant clusters of research.

Tool Example: ChatPDF, AskyourPDF
Use Case: Ask questions directly about the content of research documents to clarify methods, figures, and results.
How it works: ChatPDF.com is an AI-powered tool that lets users upload a PDF or insert a URL and then interact with the source material using natural language to extract answers directly from the document. This approach helps minimise AI hallucination, since it narrows the model’s reference material to the specific uploaded PDF (or linked document) rather than relying on general web knowledge.

A key differentiating feature is that ChatPDF supports URL-based input, while AskYourPDF typically only works with uploaded PDFs. Both tools also offer mobile apps for Android and iOS, allowing for document-based Q&A on smartphones.

Free Plan limitations: ChatPDF’s Free Plan offers only limited usage, usually including constraints such as a restricted number of PDFs per day, page limits per PDF, and limits on the number of questions you can ask per document.

Best Practice: Helpful for interdisciplinary collaborations or when working with complex research materials, such as technical papers that may be dense.

Tool Example: Gamma, Napkin AI
Use Case: Generate publication-ready figures, ppt presentations and interactive dashboards from datasets. Convert textual explanations or research summaries into diagrams, infographics, charts, and mind maps.
How it works: Gamma.app is an AI-powered content creation platform that can generate PPT presentations from written prompts, utilising AI to handle layout and visual formatting, allowing users to focus on content. Gamma’s free plan includes restrictions such as finite AI credits and fewer customisation options.

Napkin.ai generates visuals directly from text, eliminating the need for manual design. It transforms text into diagrams, flowcharts, and other graphic formats that can be exported to PPT, PDF, SVG, and PNG. This means users can quickly produce visual assets for presentations or documentation from natural language input.
Best Practice: Build visuals that tell the story of your findings; test different chart types for clarity.

Napkin screenshot:

Gamma screenshot:

According to the requirements detailed in Annex 3 of the Study and Examination Regulations, the expected length of the thesis refers to the main text, excluding annexes and the table of contents, and is regulated individually by the study programme leaders. All theses must include a reference list and a summary/conclusion section.

In detail:

• external and internal cover page
• table of contents
• thesis => introduction, discussion/main text, summary/conclusion
• list of references and illustrations
• annex (not mandatory).

According to the requirements detailed in annex 3 of the Study and Examination Regulations

• formatting:
  • 12-point serif[1], 1.5 line spacing, justified paragraphs;
  • clear indication of main and sub-chapters (highlights, alignments, structuring);
• numbering
  • page numbering shall be continuous, no need to restart by chapters;
  • main and subchapters have to be numbered, and numbering shall indicate the hierarchy (3.; 3.1; 3.1.1; 3.1.2; 3.2.);
  • illustrations shall be numbered and titled, as well as the type shall be indicated (photo, figure, table, map, etc.);

1. Figure – examples on the indication of chapters and sub-chapters, and on numbering illustrations 

In the previous lessons, you mastered where and how to search scholarly resources for your thesis. In this lesson, you will learn how to extract relevant content from these sources and then how to compile and write a thesis based on the received information and data.

Game!

Click on the games to test your knowledge.

Illustrations in the thesis

Content requirements of the thesis

[1] Times New Roman fits this requirement

The term “note-carding” originates from the pre-Internet era, when authors took notes on paper notebooks and wrote important information and data on cards. Today, more software is available for this purpose, allowing everyone to find their best fit. The goal is to create easy-to-move elements of your notes that make it easy to compile the structure of your thesis.

2. Figure – note-taking by hand or by using software

One card includes all the data and information from one source (article, chapter, book, etc.).

Start writing your notes (note-carding) only once you have appropriately defined your topic, specified its scope in terms of space and time, and stick to these boundaries throughout the thesis. Don’t start the introduction from the very beginning of time, don’t make a global outlook.

The ACCURATE bibliographic data of the source. Never use a source without them, as these data enable you to retrieve the correct source quickly.

Bibliographic data includes minimum the:

• author(s), editor(s) name;
• title (and sub-title if there’s any);
• data of edition (in case of books) (e.g. 2^nd extended edition);
• publishing data (which publisher published when / in which journal it was published);
• page numbers.

These bibliographic data should always be written as they appear in the document.

Your comments and thoughts that arise while reading and processing the literature, encouraging in-depth reflection and critical evaluation of what you have read. These questions and answers will serve as the basis of your own observations, opinions, and ideas, which will be evaluated very positively by the reviewers

Here’s some advice based on experience gained by practice: when taking notes, it’s worth writing the important thoughts verbatim (word by word) because later it’ll be easier to rewrite the original text in your own style than to search for the original based on a paraphrase.

Build the structure of your thesis by using the note-cards. Evaluate and compare the thoughts you noted, extend them with your own opinions and perspectives, and then organise them according to the intended logic – that is, compile the main and sub-chapters, the structure of the thesis. Make sure that the length is consistent with the content.