Four years ago, Google launched Hummingbird, one of its most famous algorithms. It differed from Penguin and Panda, because Hummingbird uses semantic searches: A much more effective way of establishing communications and allowing users to find the information they need more easily. Hummingbird makes emphasis on the words written in each query and this allows the whole search to be considered, rather than simply offering certain keywords. The goal of this new method of organizing searches on Google is to display pages that match that particular meaning of the word or phrase written by the user.
You will wonder what semantics is, and here we save you a visit to Wikipedia: It is an area of work and research within linguistics, specialized in the study of the meaning of words and phrases (not symbols.) In this way, semantic searches try to understand the meaning of a certain search, as well as the intentions of users regarding the information that they will have in their hands.
Both semantic searches and semantic HTML are two important trends that have gained momentum since the previous year and are projected as a fundamental part of the SEO of the future.
Because semantics not only studies the meanings of words and phrases but also the contexts in which they are located, some call semantic searches ‘contextual searches’, and the way of optimizing those searches, ‘Contextual SEO.’ In this way, those searches not only assume that keywords are important, but the words or phrases that normally surround them in a paragraph: Its semantic context. Similarly, semantic searches also consider the context of the search topic and the search platform.
This new mode of operations has allowed Google to change its way of perceiving itself: it is no longer a search engine, but, in fact, a response engine. This tremendous improvement of Google was derived, above all, from demanding users and their specific needs. Twenty years ago, we wrote ‘mechanic workshop,’ and Google offered us the results on mechanics workshops in our country or city. But little by little we became more specific: ‘Mechanics workshops in Brooklyn,’ and the results were more useful to users. But the market for goods and services is quite diverse, and the information must be even more specific: ‘Mechanical workshops in Brooklyn for antique cars.’ Then Google will interpret the users’ wishes based on the meaning of their words and bring them to much more accurate information about what they need.
That’s not all. In the case of Google, semantic searches also have a geolocation factor that makes things much easier for users. If the searches are performed from a mobile device, Google will automatically deliver results that relate to that user’s physical location. It will show him the Chinese restaurants of the area he/she is located, for example.
Semantic layout, or semantic HTML-code, is an approach to creating HTML-based websites based on the use of HTML tags in accordance with their semantics (their meanings,) and also assuming a logical and consistent hierarchy of the page. It is opposed to the approach in which the writing of an HTML code is determined by the appearance of the website in question. To design websites written in accordance with the semantics, people normally use cascading style sheets (CSS.) The HTML standard — from the very beginning — included a number of semantic tags, but the semantic layout was very popular after the beginning of HTML5.
The semantic marking of a website content is crucial, both to facilitate the understanding by the users as well as to facilitate the access to search engines. The reason behind it is simple: The order in the structure of a website is as important as the order status of a merchandise warehouse, for instance. In the beginning, when the boxes and shelves of the warehouse are scarce, you will not need much time to find any piece of merchandise. However, as the number of objects grows, the difficulties will increase until it becomes virtually impossible to access any box stored there.
The same thing happens with web pages. When websites are small and few, the lack of an ordered structure may possibly go unnoticed (and the inconveniences produced are also smaller.) However, as the site grows, the appearance of new files, contents, sections, forms, etc., will make the functionality of the website more and more complex. So, another system will be necessary.
An algorithm just can’t distinguish the semantic difference between:
<H1> CSS Goals </ h1>
<H1> Use Visa MasterCard</ h1>.
In fact, the only thing that<h1> says is that it is the most important element of the code line, the owner, but it does not tell the algorithm what is the meaning of such owner. Semantic searches, through more precise lines of code, make the algorithm go further, towards the meanings of things. It is a remarkable advance for artificial intelligence.