Google searchers have grown lazy. No longer do they scroll down the page in search of a proper answer. That is the result of a study among a panel, which got 450 search assignments. Using eye tracking the eye movements of the panel members could closely be watched. In 48 percent of 14.000 executed searches the user got no further than the first page. The rest of the pages were not consulted nor were the advertisements looked at.
Researchers blame the users, who are getting lazy. As soon as the first page does not contain relevant information, a new search is launched. This should have effects on advertisement positions as well as on optimalisation of advertisements on pages and on costs.
Since the arrival of Google, searchers have received many references on a simple question. This is caused by duplication of documents and the Boolean search criteria. The search engines have become marketing tools and are no longer search tools. Precision and recall are not exactly the objectives of a search engine like Google.
Recently two Dutch PhD students Bas van Gils and Paul de Vrieze, took the position in their thesis that Google should be like an experienced librarian, who understand what information is adequate at a particular moment. Now Google throws many references to the questioner, but computers systems and algorithms should adopt themselves to the users. Search systems should assist users in finding the proper information at the right time and in the right form. If a users searches for information on a mobile, the search system should understand that it should not return loads of results for a small screen. If there is no media player on a computer, the search system should not return references to audio and video. Also personal characteristics of the searcher (young/old, education, employment or study etc.) should be taken into account.
Bas van Gils developed a method to to evaluate the information better. He used an economic model by seeing the world wide web as an information market, where users spend money for finding valuable information. Search engines are in this model brokers. They have a profile of the party which asks for information. This broker uses characteristics as subject, form, format, language, length of document to evaluate the relevance of a document for this user in his particular situation. He pleads to separate form and content.
Paul de Vrieze developed a model for adapative personalisation which works for all kinds of software. With the adaptation engine various applications, including search engines can work together without the loss of privacy sensitive data. He puts the adaptation engine with the data between the user and the software
Both developments yield more precise and relevant data and let the searcher be in command.
Tags: searching
RSS feed: http://buziaulane.blogspot.com/rss.xml
Blog Posting Number: 603
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment