Metasearch Systems: Examples, Principle of Operation
The Internet, a couple of decades ago, onlyemerging in the most developed countries, is currently developing at a rapid pace, with nothing comparable. Every day the amount of information in it grows exponentially, representing dynamically variable data. Naturally, such a mass of information needs somehow to be controlled. It was for this purpose that search engines were introduced in due time. But at the moment to cope with all the amount of information that every minute comes to the Web, not a single search engine. Especially if you take into account the difference of search algorithms. After all, by asking the query in different search engines, even equivalent to each other by databases, the user can get completely different results. Therefore, it was necessary to improve existing systems. Using the aggregation of results, the most developed PCs implemented additional resources in their software - meta-search systems.
What does metasearch mean?
In the Internet environment a meta-list is usually calledA machine that can process the user's query and the results of the most common search engines. In other words, the interface of the program is no different from that of a conventional search engine. But when a specific request is given to it, it does not use its resources, but redirects the request to the leading search engines, and as a result, the user sees a combined list of results from the databases of different search engines. Moreover, for the convenience of using this type of search, all duplicate links are completely eliminated, which contributes to the improved results of data output.
The most important advantage that can beboasting a meta-search system, is the ability for a short time without too much effort to obtain the results of various search engines without duplicating the links. Using just one program to structure information, you can get a wider range of results from different sources, without having to use a lot of leading listings.
Many users often face this type ofa problem that some rare information, documents or programs can be seen in one search engine and completely absent in the other. Thus, a person begins to use those where he, in fact, found rare information. And again, faced with a similar problem. Already another query the selected search engine does not find, but it is displayed in the old one, which previously did not meet expectations. It is this shortcoming that makes it possible to eliminate metasearch systems in real life, giving users the most balanced results.
All the disadvantages of metasearch follow from hisadvantages, being their logical continuation. There is no own index database in meta-search as such, therefore it is impossible to add URLs of own sites to it. The second, rather important minus is a rather meager list of syntactic possibilities, that is, it is quite difficult to form an expanded information search.
In general, up to twenty differentdata search systems, and, naturally, their advanced search options are different. Of course, at the moment there are already metasearch systems developed on the principle of a new school. They take into account all the differences between the MS itself, and, accordingly, the expanded search has more possibilities. But so far there are few such systems, so the question remains unresolved.
Before we consider functionalthe possibilities of these systems, it is worth noting that there are differences in this type of search. The classical type, discussed above and based on the fact that the MS handles the indexes of other search engines, and then passes the information to the user, is considered the most common and claimed. But in view of the improvements and changes in such resources, there are also other varieties of metasearch systems.
One of the most common at the momentis the case in which the information about the results of the request is immersed in frames, and all the information is within the same meta page. It is worth noting that each frame in this case contains the original page of the search engine, which was chosen by the target. In some cases, a couple of pages are opened, depending on the number of PSs selected by the user.
Also the variant "All in one search" is popular. That is, the user has a search form open, with which he can access an infinite number of different search systems. But when making a request, he can only apply to one chosen one. In other words, one search is one search engine. This is not as convenient as, for example, in the classical version.
The most common search engines are meta
Among systems of this kind the most popularThe classic platform was the Vivisimo metasearch system. She recruited up to twelve million unique visitors in one month. But it is worth noting that popularity does not mean automatic confirmation of the quality and comfort of using the resource.
At the moment there are manyclassical meta-finders. All of them have a long history of creation, promotion and existence. Practically each of such systems has the same principle of action, therefore it is not worthwhile to consider them separately.
Therefore, we present classical meta-search systems, the list of which looks as follows:
Also worth paying attention to IxQuick, MetaEureka, ZapMeta, WebCrawler and WindSeek.
The main difference is the interface. Therefore, choosing the best among them is quite difficult, because each user finds a resource depending on their own preferences. Only after trying personally to work with each of these systems, you can understand which one meets all needs and meets expectations.
At the moment, meta-search systems are being createdquite another type, from the so-called "new school". Their main difference from the classical ones is the use of clustering. Technically, its implementation takes place in different forms. Usually it looks like an automatic selection of keywords and phrases with information added in the cloud, indirectly related to the user's request. The screen displays additional keys, by which you can refine the search, make it more narrowly focused. In other words, meta-search bibliographic systems offer options that make the query more focused, narrowing the scope of information retrieval and helping the user to find the data he needs.
In addition, the new generation systems allowmore clearly work out the results. In other words, they use the so-called search strategy, that is, if necessary, the user can specify its parameters independently.
From the new generation to meta-search systems are: Clusty, Kartoo, Mooter, Izito, WebClust and iBoogie.
All that the search donors give out issubject of ranking and is analyzed by meta-search. This happens immediately, as soon as the system receives the results, the entire process is automatic. The resource ranks all descriptions of the found data. Further, a repeated analysis is carried out, the main criterion of which is the place in all components of the search, and all similar data found by the searcher in a specific and other PS are taken into account.
In addition to high-grade search engines, there are alsothey are often called "all in one." In their properties, they are very similar to the PS, but they are not. We have already discussed many meta-search systems, examples of pages created on their similarity: iTools, AllSearches, AdClick.ru and Searchalot. Despite their inferiority as a PC, many users still prefer to use these resources for searching. In fact, this is not surprising, since these pages have very similar functionality and implement the same features that are available, for example, in iBoogie.
At the moment there are several hundredresources, which are meta-search systems. The Internet is developing at an active pace every year. With regard to individual choice, it often depends on subjective factors, for example, a successful range of colors or the presence of comfort using the interface. It is worth noting that some meta-finders were initially unsuccessfully created, their functionality is not able to improve the search, and in some cases even aggravate the process itself. In this connection, it is best for the user to select a suitable resource for himself.
Naturally, Western search engines in many respectsahead of domestic, and many use them as a tool for searching. But it's impossible not to mention the quality and usability of Rambler and Yandex. These meta-detectives perfectly cope with their work and help to find optimal results even for complex and rare queries. Also I want to pay attention to Nigma.rf. It was the very first to appear on the open spaces of the Runet as an intellectual system for metasearch. Also at the moment, the system for searching social networks called yoname.com is gaining popularity.
In any case, the search for information withMeta-search systems are much more efficient than the classical use of different MSs. After all, they display much more results and cover a large database of information on the network. And given the speed with which the amount of content on the Internet grows fast, such systems will be very useful even for normal work on the Web.