Hilltop algorithm
The Hilltop algorithm is a calculation method that evaluates a large number of linked documents and determines those that are most relevant to a search term. This determines whether the website in question belongs to the expert or authority pages. In addition, the relevance for certain keywords is weighted. As soon as a random search query for a specific term is entered into a search engine, the search engine first determines the expert and authority sites that are relevant to the search query.
Unlike the PageRank algorithm, the value of individual documents is not determined, but rather the links between a specific search term and a document.
The Hilltop algorithm was developed at the University of Toronto by the two scientists George Mihaila and Krishna Barat with regard to the increasingly important search on the Internet for the most relevant pages for a search term. The algorithm was an important innovation in the field of internet searches after Google acquired the patent for the Hilltop algorithm in 2003. It cannot be clearly determined to what extent Google uses the Hilltop algorithm for its searchers. However, it is assumed that the algorithm plays an important role in determining search results.
Expert pages
A website achieves expert status if it deals with a specific topic and has many outgoing links from independent websites on this topic. A website is considered independent as soon as it has been created by different authors and organizations.
Authority pages
A website is described as an authority site if at least two, preferably many, independent expert sites refer to the potential authority site. Expert pages must fulfill the following criteria:
- There must be no relationship between the referring expert pages.
- The first three blocks of the IP address must not be identical.
- The generic part of the domain name must be different.
The goal of search engine optimization is to achieve authority or expert status.
Calculation processes of the Hilltop algorithm
The term “hilltop” describes the calculation method of the hilltop algorithm very aptly. In terms of calculation, the pages that provide the greatest overview of a subject area are used first.
In practical terms, this means that the Hilltop algorithm determines the expert pages mentioned above in the first step. The pages that have many links to independent pages on the same topic are identified very simply and usually automatically in the background. In addition, the expert pages can be easily updated, as this is usually a very small number.
The search for the authority pages takes place afterwards and is normally much simpler, as only those pages to which at least two independent expert pages refer are accepted as authority pages.
Prevention of spam through the Hilltop algorithm
Since search engines cannot check content for its real relevance, but can only recognize it on the basis of character strings, other methods must be chosen for this. The Hilltop algorithm allows superficial content to be excluded from search engines, as such low-quality pages are generally not linked to by many expert sites.
Spam pages or other pages with low-quality content fall through the cracks of the Hilltop algorithm. Normally there is now no further possibility of being classified as an authority page. The relevance is automatically assessed as low, so to speak.
Differences between the Hilltop algorithm and the PageRank algorithm
There are important differences between the Hilltop and PageRank algorithms, particularly with regard to the calculation method. When it comes to general search queries with a high frequency for relatively general search terms, the Hilltop algorithm is considered to be much more accurate. In addition, the search results are better validated by being based on independent expert sites. These can also be falsified less easily.
When it comes to very specific search queries, the Hilltop algorithm reaches its limits. This is due to the fact that there are comparatively few sites on the Internet that deal with such specific topics or search queries. The PageRank algorithm, on the other hand, determines the link popularity of a document or website in particular. According to the newer calculation methods, even entire blocks, i.e. an entire domain, can be defined.
The PageRank algorithm is based on the “random surfer model”, which is true in many areas, but PageRank only reflects the interest of other website operators. This means that the individual interest of users is not taken into account. This effect is relativized by many and regular adjustments, but the use of some additional algorithms remains. These additional algorithms include, for example, “TrustRank” or “Hub and Authorities”.
In contrast, the Hilltop algorithm, with its concept based on independent experts, offers an evaluation that is particularly focused on the user and their benefits. This cannot be undermined as easily as with the PageRank algorithm.