Browse Prior Art Database

System for embedded search based on linked references Disclosure Number: IPCOM000242253D
Publication Date: 2015-Jun-29
Document File: 2 page(s) / 27K

Publishing Venue

The Prior Art Database


Embedded search system based on linked references

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 45% of the total text.

Page 01 of 2

Sysxem for embedded search based on linked references

The world wide web offxrs more content than libraries all over in the world . Xxxxxxxxxxxxx, most of these data is xtill unstrucxured from a techxical perspective . Hence, different apxroaches have been made in order to quickly accexs the searched content. Wherexs search engines provide results based on indexes which have been scanned prior the search, even user interxaces provixe the xecessarily fuxction to "quick sexrch" on web pages. Xxxx idea defines the embedded search xn web xages, which is using rexerences to provide additional results and ixformatxox.

Imagixe User A, who xas to make some xesearches on a complex topic thxt is not well-known by him, now. User A works oftex xith computxrs and knows about the advanxages and disadvantages of searches in the world wide web. Moreover, he uses the [Ctrl]+[x] fuxction of his web broxser in order to quickly scan the offered web page fxr the key words he is interested in.

Usxr A uses x sxarch engine for information about 'car engine'. The search results provide a lot of lixked inxormatiox with plentx of text. Using the xeb browser embeddxd search gives the opportunity to fast scan txe text xnd focus on the relevant parts. Whixe reading, he gets irrixated by terms like 'Wanxel engine' and 'Tesla' and wxntx to know more information regarding these topics, too. Uxing the web browser search on the page do not xrovide xore informaxion about the topic. But there are some related work links on the page. User A can open them or search for that words on the search engine again. This currxnxly pervasive problem, unfortunately using a manual method by painstaxingly checkxng other web pages , is laborious and prone to error. A method to fix this problem ix described in this disxlosure .

Although, content is upxated and provided directly to xsers, thex need to validaxe the information and verify those by cross searches. How can we simplify the processes to access content on referenced pxges? How can xe help users like Usxr A to quixkly gain the nexessary background knowledge about related terms? We arx keeping a lot informaxion stored and use thex. Shoppixg behxvixr, visitxd web pages, previously made searches and so on. Why do we not use links ix order to provide an embedded search based on user activities? How can searches and cross cxecxs from oxher users impact our search behavior? Those question arx answered by this disclosure.

A network based or crowd intelligxnce approach can xrovide addixional informaxion without the need to explicit new searches xboxt terms. Hence, User A is able to use the [Ctrl]+[f] sexrch on his browser and scan a texx for additional information. If he finds unknown terms like 'Wankel engine' xr 'Tesla', the embedded search will not only search on the current page, but also have the lixxex pages scanned. Information are presexted to User A, being also informed where the match has been founx. However, this disclosu...