Search tips

The search begins with a precise definition of the topic: What am I looking for and what am I not looking for?
The first step is to clarify all content-related questions. The need for information often arises from a task or a question. If the topic is only vaguely defined, it can be helpful to formulate questions or sentences that precisely define the information requirements.
It is also important to know what type of information is required: Factual information, i.e. data, or specialised literature, i.e. essays, books and the like? The depth of the desired information must also be clarified: Is a general overview sufficient or are details required? And finally, the question of scope must be decided: Is completeness necessary?

Regardless of the type of information required, all search systems require words that can be used to search. A list of possible search terms should therefore be created before the actual search. Suggestions can come from the question or topic. Firstly, you should think about your own search terms: What terms can I use to search? Encyclopaedias, dictionaries or manuals or a short test search on the Internet can help to clarify the topic and provide further search terms.

Synonyms should be added to the word list. Synonyms are terms with the same or similar meaning. Example: Paprika - peppers - ... Since you do not know which words are used in a text, especially on the Internet, it is necessary to know as many alternatives as possible. You should also take different spellings into account: Donut - doughnut - ...

Depending on the topic, it may be necessary to translate the words in the list into other languages. English is particularly useful, as this language is predominant on the Internet. When searching for a Spanish artist or a park in France, other languages are useful. In addition, some search systems always require English search terms.
The word list can be used again and again as the search progresses. The list should always be kept up to date and new words added as required.

Some search functions can be used in library catalogues as well as in databases and to some extent in internet search services. These include:

•     the difference between keywords and subject headings,
•     the truncation of search terms,
•     the use of Boolean operators (also: logical operators)

Colloquially, there is usually no difference between a keyword and a subject heading. For example, we speak of the subject heading "healthcare reform".

Keyword - A keyword is basically a word that actually appears in the text. If a word comes from the title of a book or essay, it is referred to as a title keyword.

Subject heading - A subject heading describes the content of a text. The word does not necessarily have to appear in the title. It is assigned additionally, for example in the library catalogue for a book. It facilitates the thematic search for the user of a catalogue. The subject headings are often compiled in an index. This makes it possible to find out which words are actually used.

Example - There is a book about the life of Albert Einstein. It is called: "Einstein; the new world view of physics". If you are looking for a biography of Albert Einstein, you will probably use search terms such as Einstein and biography. However, the example above cannot be found using these terms, as the word biography does not appear in the title.
For this reason, the subject heading biography is assigned to the book in a library catalogue. This is how you can find the book.

Truncation makes it easier for you to search with words that have the same root. A wildcard character is appended to the word stem for this purpose. The asterisk (*) is often used as a wildcard character, but other characters may also be required depending on the catalogue or database. Possible other placeholders are ? # $. Please refer to the help texts of the catalogue or database to find out which wildcard character to use.

Input: Fish* finds: Fish, fishing, fisheries, fishermen, ...

In addition to the right truncation described above, in some databases it is also possible to truncate to the left - i.e. at the beginning of the word:
Enter: *fish finds: catfish, shellfish, jellyfish, goldfish, ...

...or to replace individual or several letters within a word with placeholders. Further information can be found in the help texts of the catalogues or databases.
Input: Me?er finds: Meier, Meyer

Boolean operators (also known as logical operators) are used to link several search terms together. They determine whether one or more search words must, may or should not appear in the result.

AND - If you combine two or more words using the AND operator, only results in which all words appear are displayed. The AND operator returns the intersection of the search terms. The input: dog AND cat finds documents in which both the word "dog" and the word "cat" occur.

OR - Alternative terms, synonyms and similar terms are meaningfully linked by OR. In this way, results are displayed that contain at least one of the words. Even if several of the words entered occur, the document appears in the hit list. The OR operator increases the number of hits (union). The input: dog OR cat, for example, finds documents in which either the word "dog" or the word "cat" occurs. However, texts containing the words "dog" and "cat" are also displayed.

NOT - If you want to exclude certain words from the search, you can use the NOT operator. All words before the NOT must appear in the hits, but all words after the NOT must not appear in the texts. The NOT operator reduces the number of hits. The input: dog NOT cat finds texts containing the word "dog" but not the word "cat".

Graphical representation of the results when using Boolean operators here.

A distinction is made between a whole range of publication types. These are roughly divided into so-called independently published and non-independently published literature.

Independently published literature:

•     Monograph ("classic" book)
•     Collected edition (several contributions by different authors in one book)
•     Internet document

Non-independently published literature:

•     Journal article (single article in a journal)
•     Individual contribution in a collective edition

Depending on the type of publication, a suitable search tool must be used.

A book (also: monograph) is independently published literature.

What information do you need to find this book later?

•     Author: Charbel, Ariane
•     Title: A quick and easy way to write your thesis
•     Edition: 5th, updated edition
•     Place of publication: Nuremberg
•     Publisher: BW, Bildung-und-Wissen-Verl.
•     Year of publication: 2006
•     ISBN: 3-8214-7652-4

This data is referred to as bibliographic information or references. They are used to provide the correct source for citations and are compiled in an appropriate form in the bibliography.

Library catalogues are mainly used to search for books.

A single article from a journal is literature that has not been published independently.

What information is needed to find this journal article later?

•     Author: Wiegand, Dorothee
•     Title of the article: Well cited is half written
•     Title of the journal: c't
•     Volume/year:
•     Year: 2006
•     Issue number: 7
•     Pages (from to): 160 - 165

This data is referred to as bibliographic information or references. They are used to provide the correct source for citations and are compiled in a suitable form in the bibliography.

Literature databases are primarily used to search for articles.

In general, an Internet document should be treated like independently published literature. However, there are some special features.

What information is needed to find this internet document later?

•     Author or publisher, originator, person responsible: German Bundestag
•     Title of the page/text: Press releases
•     Web address, URL: https://www.bundestag.de/en/press/press
•     Status (date):
•     Retrieval date:

This data is referred to as bibliographic information or references. They are used to correctly cite sources for quotations and are compiled in a suitable form in the bibliography.

Search engines are primarily used to search for Internet documents.

Nowadays, searching for information on the internet is almost exclusively associated with the Google search engine. The verb "to google" has even found its way into the dictionary.

In addition to searching for books in library catalogues and looking for articles in literature databases, research on the Internet also plays a major role in academic work.
However, anyone conducting academic research on the internet always uses several different search engines. Using Google alone is not enough!
When using texts from the Internet for academic papers, it is important to ensure that the content is reliable, i.e. correct in terms of content. On the other hand, care must be taken with texts from the internet to ensure that the statements made there are also substantiated or documented.
Before discussing the evaluation of websites and differentiating between the various types of search services, we will look at what cannot be found using search engines on the internet.

Not everything that exists on the internet can be found by a search engine. To understand this fact, it is necessary to know how search engines work.

To build search engines, automated computers - called crawlers - surf through the internet. They crawl from one website to the next via links. In this way, they reach a large part of the internet over time. Individual pages are not taken into account.

Each time a website is visited, the crawlers automatically store the content of the page (text, graphics, etc.) and the corresponding URL in a database. This database and its index form the heart of the search engine.

If a user sends a search query to the search engine, the index is searched for suitable websites. The results are sorted by the search engine's automatic ranking and then displayed as a hit list. The original pages can then be accessed via the URLs listed.

Important: A search engine only ever searches in its index - not directly on the internet.

Websites that have not been visited (indexed) by a search engine cannot be found. The pages that cannot be found belong to the invisible web. To remedy this problem, it is always advisable to consult several different search engines.
There are other factors that mean that some websites cannot be found. Some of these include dynamically generated pages, i.e. web pages that are only generated when a visitor enters the page, e.g. the hit list of a search engine. Furthermore, search engine crawlers cannot index content on password-protected pages. These cannot be found either.
Sometimes pages cannot be found because they are too hidden in the hierarchy of a website. In this case, the crawler does not penetrate to the deepest level of the website. Such a hidden page therefore remains undiscoverable by search engines. In most cases, copyrighted material cannot be found on the internet either.

Library catalogues and literature databases make up another area of the invisible web. These can only be searched by directly accessing the individual search systems.
The majority of scientifically relevant literature cannot be found via general search engines.
There are different estimates of the size of the invisible web (in relation to the visible web, i.e. the web that can be found using search engines). They range from the assumption that the invisible web is at least as large as the visible, findable part of the web, to the idea that the invisible area is 500 times larger. The fact that there is no exact figure for this is in the nature of things: what cannot be found cannot be counted.
The following shows how the visible part of the internet can be found.

There are different search services on the internet. Depending on the purpose and information requirements, one or more of the following types should be used:

•     Reference works
•     (General) search engines
•     Meta search engines
•     Specialised search engines
•     Open access search engines

Reference works

Indicator: Certain facts are listed in reference works. Reference works naturally include dictionaries and encyclopaedias, but also telephone directories and address directories.
Purpose: If specific factual information is required, e.g. the life data of a person, or if an unknown term is to be explained, reference works are recommended. Of course, this applies not only to reference works on the internet, but also to printed editions in the library.

Examples:

• Wikipedia - The "free encyclopaedia" is the collaborative work of thousands of internet users. Theoretically, anyone can contribute to the encyclopaedia and add their own content. The content may therefore need to be critically scrutinised. URL: https://www.wikipedia.org/
• LEO - The online dictionary was created by the Technical University of Munich. Translations are available in German, English, French, Spanish, Italian and Chinese. The dictionary is supplemented by a forum on detailed linguistic problems. URL: https://dict.leo.org/

General search engines

The most widely used search services on the internet are (general) search engines. In addition to searching for websites, they often also offer other functions such as image search or searching for music or videos. Important: Search engines only search the automatically generated index; they do not search the internet as such during a search query.

Characteristics: The search is carried out in an automatically generated index. There is no quality control. The hits are ranked according to their relevance.
Purpose: General search engines - when using the advanced search functions - are particularly suitable for highly differentiated search queries, which are not possible in this form in directories, and for information on current topics.

Examples:

• The best known and most used search engine is Google. URL: https://www.google.com/
• A real alternative to Google is the Yahoo! search engine! Suggestions for search terms are made as soon as you enter a search term. As Yahoo! works with its own index, the search results differ from the hits on Google. URL: https://search.yahoo.com/

Meta search engines

A meta search engine is a search engine that simultaneously queries several search engines and provides the user with a joint list of results. A meta search engine is therefore a superordinate ("meta") search engine.

Characteristics: A meta search engine queries several search services simultaneously. The results are displayed in a joint hit list. In many cases, it is possible to see from the hit list in which search service the hit was originally found.
Purpose: Meta search engines make searching easier. They are particularly useful when no or very few hits or only irrelevant results have been found in a single search engine.

Examples:

• The German-language meta search engine MetaGer is operated by the computing centre of the University of Hanover. Here it is possible to specify in advance which search services are to be included in the meta search. Primarily scientifically relevant search services are included. URL: www.metager.de
• A commercial meta search engine is metacrawler. The focus of the search results is on English-language pages. Results financed by advertising are not clearly emphasised. URL: www.metacrawler.com

Specialised search engines

Specialised search engines allow you to search a reasonably well-defined part of the internet. Formal criteria can be decisive here, such as in the case of an image search engine. Or it may be content-related criteria that make up a specialised search engine, e.g. a search engine for physics information.

Characteristics: Only a certain part of the internet is covered by specialised search engines. The limitation can be thematic or formal.
Purpose: For very specialised information requirements or if the search can be narrowed down precisely by topic or form. If very high relevance of the search results is required, specialised search engines are much more suitable than general search engines.

Examples:

• Bielefeld University Library offers a specialised search engine for exclusively academic information on the internet under the name BASE (Bielefeld Academic Search Engine). URL: www.base-search.net/

Open Access search engines

Open Access publications are (scientific) publications published on the internet. Unlike academic texts published in books and journals, open access publications are freely available to everyone; there is no need to purchase expensive, specialised academic publications. As a rule, these texts are published on the servers of universities, libraries or research institutions.