Filtering search results

This module will probably be more theoretical, but because of this, it might also be more interesting. We’ll have a look here at how to display various search results and why this variability can be useful and dangerous at the same time. We’ll have a look at areas where a certain filtering feature can be used and describe how and why it’s used.   

Voice searches and voice assistants.  

When entering a search query, most of us will usually choose a common text search in their web browser window. But, there are also other options – the first, which is interesting primarily for mobile phone owners, is a voice search, which can be used to simply dictate a search query. This option is also interesting for owners of smart watches, as this is their only search option.  

An extremely interesting area for searching for information are voice assistants like Siri, Alexa or Google Assistant (we definitely recommend having a look at what they can do). These are systems that we know primarily from mobile devices, in which the result of a search query doesn’t necessarily have to be a collection of sources that a person then chooses from – it can also be specific information. Although this isn’t the most widespread search method in some countries primarily due to the language barrier, everyone has access to it; however, working effectively with such a system also requires specific training and experience. We’ve already mentioned that information continues to be more quickly and readily available, and voice assistance is one of the ways to ensure this.  

Voice assistants are also interesting in that they usually give you only one search result, so you’ll have to rely on the fact that they’ll choose only the suitable information from all that is available. Their functions and methods of data filtering can be very diverse, as can be seen for example with the Japanese gatebox, which comes in the form of an animated female hologram.  

When two people do the same thing, it’s still not the same thing 

If you enter a query into a search engine in exactly the same way as your friend or colleague, each person may get different results. More advanced methods of ordering results consider not only what key words you’ve typed in, but also what the search engine already knows about you. This can include your previous queries or information taken from your internet browser. For example, if you type the word “cinema” into Google, the first results – thanks to information from the browser – will offer cinemas around the area in which you’re physically located. The same method is used to filter results, which (usually for legal reasons) are available to users in some countries and not others. If you want to remove this filter, you can use a search engine like DuckDuckGo.  


Search results in Rimavská Sobota. Source: Screenshot: KISK. 

The same search in Brno. Source: Screenshot: KISK. 

Personalization of results 

This phenomenon is called the personalization of search results. It is based upon the idea that each user has a certain circle of interests and topics that they deal with, and also content which is not relevant to them. If a search engine is to offer relevant search results, it has to know how to consider whether the given person is a typographer or a chemist and offer him or her suitable content according to this. If the search engine didn’t carry out search result personalization, making sense of search results would be considerably more difficult and complex.  

Information bubbles 

Such personalizations of searches leads to a certain segmentation of society, as various groups of people find differing information and news, because the same search engine gives varying relevance to certain information – this is perhaps one of the reasons why, for example, people who vote for Joe Biden and people who vote for Donald Trump can’t understand one another. They are consuming a completely different blend of information. This leads to something that’s called an information bubble. The more careful and long-term the search is, the more pronounced the bubbles become. Finding your way out of them is extremely difficult, and tools such as DuckDuckGo can at least partially help you do this.  

The totalitarian bubble 

Work with search results in totalitarian states such as China are a specific type of personalization. There, something like the “Great Wall of China” is in place – a massive search results filter that does not allow information in from the outside world that might seem problematic or dangerous to the ruling structure. The filtering of results goes an extra step there, as Baidu, China’s key search engine, is subject to strict party surveillance. Because of these steps, China can be said to have an intranet (i.e. its own closed network) rather than an internet, and a very limited one at that. One of the few possibilities there to access relevant information has been to use a VPN, which has been highly problematized and intensely monitored as of late.  

Local personalization  

The final type of personalization that we’d like to mention is geographical – if a person is looking for something like a café, the search engine should provide information on the closest one or ones that are located in a reasonably defined area before it shows those that are too far away.  

Although result personalization is an extremely useful and functional matter and we’d have trouble imagining today’s internet searches without it, it also creates problems or limits that should be expected and in certain ways reflected upon.   

Recommendations for content filtration 

  1. Think about what the search engine is offering you on the first pages.  
  2. Don’t be afraid to use advanced search functions.  
  3. If you want to avoid personalization, use a service like DuckDuckGo.  
  4. If you’re abroad and content filtering is taking place, work through a VPN.  
  5. Find a way to save search results (from Clipper to GetPocket or by using common bookmarks).  
  6. Don’t rely on one single tool.  
  7. If you’re not satisfied with how your search results are being modified, delete your cookies.  
  8. Try out voice assistants (most of us have one in our mobile phone).  
  9. If you want to filter certain content on your computer yourselves, choose a suitable tool. 
  10. Remember that filtering is never 100% successful.  

You are running an old browser version. We recommend updating your browser to its latest version.

More info