ACM Comput. Selberg E. The MetaCrawler architecture for resource aggregation on the web. IEEE Expert, 12 1 —14, A methodology to retrieve text documents from multiple databases. IEEE Trans. Knowledge and Data Eng. Yuwono B. Server ranking for distributed text resource systems on the internet. Zhao H. Fully automatic wrapper generation for search engines. World Wide Web Conf.
Weiyi Meng 1 1. So, what are these users looking for? The fundamental benefit of using a metasearch engine is rather obvious: you get much more information than you would using a standard search engine.
Anyone who uses the Internet to learn wants the broadest, most complete picture possible. With each different search engine comes a different index, a different crawler , a different algorithm, and so on. Metasearch engines eliminate this risk. It was a pioneer in the field however, it was not very reliable since it was restricted to simple searches.
MetaCrawler was launched in by a student and a professor at the University of Washington. It was a narrower and more accurate improvement of the SearchSavvy. It had its own fair share of limitations and over the years, numerous metasearch engines have been launched to correct the previous mistakes. Metasearch engines work by sending requests made by a user in the search slot of a metasearch engine to numerous other search engines. The servers of the metasearch engine wait for answers from the requested search engines before they can show the final results.
The results are presented based on certain guidelines depending on how they are adjusted. Duplicate results are filtered to prevent a URL from appearing twice in the search result list for a query. The image below shows the process visually:. Metasearch engines display their results in two ways, including single and multiple lists. Most meta search engines present results in a single merged list from which duplicate entries have been removed. Others do not combine multiple engine results but instead display them in separate lists as they were from each engine.
This may however result in multiple entries. When it comes to search engines, Google leads the way in terms of the number of users and popularity. Some people wonder whether Google is also a Metasearch engine. So is google a metasearch engine? And the answer is no, Google is not a metasearch engine but a search engine. The difference between a search engine and a meta search engine is that a search engine like Google uses robots to gain enough information to create a record or database of the visited sites.
The data and corresponding algorithm is then applied to form a basic index. Metasearch engines on the other hand, build their index from the results of other indices rather than from the web. They get the result from a collective grouping of various search engines. The algorithm displays the data to the user according to their preferences. As pointed earlier, numerous metasearch engines in the market today. Here are metasearch engine examples.
It is a German metasearch engine that processes searches anonymously. Its key feature is the web associator that presents semantically identical items to the search query and code search, which shows the open-source code. For example:. More prominent results get more stars. Its outstanding feature is the different setting options including powerful refinement, advanced search and limitations to European servers.
Metacrawler offers a professional search and aggregates German and international sources. It was originally developed in at the University of Washington. Yabado is a German metasearch engine that processes searches anonymously in about ten different sources.
This American metasearch engine offers several functions to users. The Intellifind gives search recommendations, Preferences sets search preference, and Favorite Fetches function that displays searches from other users. For instance:. It uses the same interface as Dogpile. Surfwax is one of the early metasearch engines, and its key feature is its specific search indexing and local site search usability.
Vroosh metasearch engine can be used by anyone however, it does not contain web or image. It offers a country-based search for more relevant results. The image below shows what the homepage of Vroosh looks like:. This is one of the biggest search engines since it takes information from other metasearch engines like Mamma, Ithaki etc.
It offers a heap of information that most other metasearch engines do not provide. Unabot is a consolidation of all metasearch engines. Dispatcher: The dispatcher has the work of query generation. Display: The display uses the queries to write back the results on the screen. It uses methods such as page ranks, parsing techniques, cluster formation, and stitching to give the desired result. Personalization: The personalization in other words is being user-specific.
This involves comparing the results with each other. Operations of the Metasearch Engine A metasearch engine does not create a database of itself rather it creates a federal database that is actually an integration of the databases of various other Search Engines. The 2 main ways of operations involved are : The architecture of Ranking: Various search engines have their own ranking algorithms.
A metasearch engine develops its own algorithm where it eliminates duplicate results and calculates a fresh ranking of the sites. This is because it understands that the websites which are highly ranked on major sites are more relevant and would thereby provide better results. Fusion: Fusion is used to create better and more efficient results. Fusion is divided into Collection Fusion and Data Fusion.
The collection Fusion deals with search engines that contain unrelated data. The data sources are then ranked based on their content and the likelihood of providing relevant data. This is then recorded in a list. The Data Fusion deals with the search engines that have indexes for common data sets. The initial ranks of the data are compared with the original ranks. A process of Normalization is applied using techniques such as the CombSum algorithm.
It writes back results from the individual search engines of Google and Yahoo. It combines the search results of text, images, new, etc. Sputtr is comparably one of the best meta search engines that combines the results from various popular search engines like Google, Yahoo, Bing, Ask.
0コメント