November 24, 2006

Extended Indented Google Results Bug?

Philipp Lenssen spotted listings with more than one indented results for a search on get fuzzy. The first two results were from comics.com and the two other indented results were from Yahoo News. All these results were under the first listing, so we had three indented results showing under the top listing. Typically, there is only one indented result and no more. I cannot replicate this, it seems like a weird bug that may have been fixed. Philipp has a screen capture of it in action.

Posted by Barry Schwartz at 8:42 AM | Permalink

November 3, 2006

See And Find With Quintura

Quintura is the latest in the line of visual search engines such as Kartoo, Mooter and WebBrain. Quintura basically takes your search term, runs a search and then translates the results into a tag cloud effect on the screen. Users can then simply look at the results (powered by Yahoo) listed under the tag cloud and click on the link as per normal, or they can explore words displayed in the semantic map to focus the query more closely.

The results section of the page is not exciting - title, URL, keywords in context and sometimes the size of the page, but that's about it. I'd like to have seen more information, and keywords in context are no more than a word either side, which doesn't provide any guidance at all. This section of the page is also squeezed into the bottom half of the screen, giving it a cramped feeling. This is probably the weakest element of the search engine, and clearly the one that the developers spent least time on because they wanted to get onto the interesting and fun section of the semantic display.

This is where Quintura does become more interesting and quite fun to play with. After the search runs the search terms appear on the screen and are surrounded with other hopefully appropriate terms. My search on 'search engine watch' for example returned keyword suggestions such as 'blog', 'forum', 'search engines' and so on. The closer to the search terms, the larger the keyword suggestions (both in terms of font size and bold), the more relevant they are deemed. Holding the mouse over a term - note that you don't need to click - will display a new set of results in the bottom window and will also show another keyword cloud overlaying the original, which does get a little confusing at times, and it's quite hard to work out exactly what you're searching on.

However, it's an interesting approach to search, and users who enjoy different approaches to the display of search results will enjoy using it.

It does obviously have more flexibility though, as it's a reasonably well rounded search engine. Keywords can be dropped from the semantic display by clicking on the appropriate icon, and any associated keywords are also dropped at the same time. Excluding 'baseball' from the search on my name also excluded 'statistics' for example. Words can be added to a search by simply clicking into a blank area in the screen and typing them into the search box that appears; words can be excluded in a similar manner as well with the usual minus sign in front. It's possible to save searches as a favorite or it can be emailed to a friend or colleague.

There are additional things that I'd like to see with Quintura; a more indepth 'help' guide, RSS feeds, greater search functionality by type - at the moment it's limited to web or images, and news, blogs and so on would be a nice addition. Equally however it's in beta mode, so it would be unfair to be overly critical.

As previously mentioned, people who enjoy visual search results pages will get a kick out of this one, while for everyone else it's a bit of an oddity they'll ignore. While the semantic element is clearly the emphasis I personally found the cramped results section too irritating to want to use this engine for any length of time.

Posted by Phil Bradley at 5:39 AM | Permalink

October 4, 2006

Clusty Labs & Clusty Cloud Creator

Given that Clusty has just had a birthday they're celebrating by introducing the Clusty Labs. If you're familiar with Google Labs you'll know exactly where they're going with this. There's not a great deal there are the moment however; a one stop site about Benjamin Franklin (Clusty Ben) Shakespeare Searched and something new - the Clusty Cloud Creator (I wonder if they could have managed any more alliteration?)

Clouds are of course nothing new, but what this offering allows users to do is to simply put in their own search term(s) and the CCC will produce a cloud on the spot. Clusty is also providing the code as well, so it's easy for web authors or bloggers to add it to their pages or posts. Since the clouds are created real time they're always up to date, and of course searchers can always simply click on a link to run a fresh search directly from the search engine.

The functionality is nice and straightforward, and clouds can be created literally within seconds, with the resultant code being cut and pasted as required. I can see lots of ways in which this could be used, and indeed Clusty has come up with a few examples themselves. Information professionals could create clouds for subjects that their patrons were interested in and put them onto a specifically designed site, teachers could use them with children to assist with projects, or bloggers could include a cloud to help illustrate the subject matter and interest of whatever they were writing about.

The functionality is also increased because users are not limited to creating clouds for simple searches (although of course they can), but can increase them in complexity using more advanced search syntax. For example, I created a successful cloud for a search on 'Tony Blair' from .gov.uk sites that excluded the term 'elections'. I can see that it would be very interesting to create clouds for similar subjects, and just to twist them slightly with the addition or subtraction of specific terms to see the differences.

Certainly worth exploring.

Posted by Phil Bradley at 9:33 AM | Permalink

May 19, 2006

Building Better Search Interfaces: A Workshop About Exploratory Searches

When someone uses a search engine to try to locate information about a subject which they know very little about, a simple search interface isn't very helpful. These types of endeavors have been referred to by information scientists as exploratory search activities, where the target of the search is mostly undefined.

Gary Price, at Resource Shelf, pointed earlier today to a paper from a University of Maryland workshop held last year, which brought together researchers from the "fields of Information Seeking (IS), Information Retrieval (IR), Human-Computer Interaction (HCI) and Information Visualization (IV)" to take a closer look at how to make it easier for searchers to perform exploratory searches.

The workshop, Exploratory Search Interfaces: Categorization, Clustering and Beyond, was part of The Human Computer Interaction Laboratory?s 22nd Annual Symposium, hosted at the School last June. The main issue explored at the workshop is nicely stated in the report:

Although the Semantic Web is making new richer and more flexible kinds of Web-based querying possible a burning question remains ?Are current search technologies enough to meet all user demands??. The answer of course is yes and no. Search engines, bibliographic databases and digital libraries provide adequate support for users whose information needs are well-defined. However, they do not work well in situations where users lack the knowledge or contextual awareness to formulate queries or navigate complex information spaces. For example, what if you want to find something from a domain where you have a general interest but not specific knowledge? How would you find classical music you might enjoy if you do not know what Beethoven or Berlioz sounds like? What a Sonata is? The difference between Baroque or Romantic? What do you type into Google?

Some experienced searchers have developed strategies to help them with these exploratory searches, but part of the impetus behind the workshop was to consider interfaces that may make such searches easier and more useful. One conclusion of the workshop participants was that interfaces that could aid searchers "are a vital (and missing) part of most current mainstream search technology."

They offer a number of suggestions in their report on the workshop for future study on the subject and the development of user interfaces that could be created involving:

• The context of searches,
• The support of different searching strategies,
• Testing methodologies for those interfaces, and;
• The creation of workspaces for searchers - from simple note taking to collaborative authoring environments.

More information about the workshop is available on the workshop homepage, which includes some excellent slide presentations from participants, like the one by Abdur Chowdhury of AOL on How good is Web Search and Does Clustering Help?. (Some interesting numbers on one of his slides showing a breakdown of queries by category at AOL in 2004.)

The 23rd Annual Symposium of the University of Maryland's Human Computer Interaction Laboratory will be held on June 1-2, with the Symposium itself on June 1, and workshops and tutorials on June 2. While today isn't the last day that you can register, they note that they need registration and payment by today to "guarantee reservation of lunch and materials." This year's event doesn't have a workshop on a search related topic, but there are some presentations during the first day that do involve search.

Posted by Bill Slawski at 4:26 PM | Permalink

May 18, 2006

New Search Patents: May 18, 2006 - Yahoo's Active Abstracts are Calling

Some new patents and patent applications from the last week cover a wide range of search related topics. On Tuesday, I wrote about some possible enhancements to Ask's Binoculars. Yahoo seems to be looking at providing more information on search results pages with smarter snippets, or what they call "active abstracts" that can lead to such things as refined queries or maps or can even allow you to place a phone call. Microsoft comes up with an intermediary query interface which could be used with more than one search engine. IBM and Hitachi both present different ways of looking at clustered results after a search. The dominant Korean search engine describes methods for selling keyword-based advertising. S.L.I. Systems presents a human powered search, which uses community to tell it when to update pages, and which keywords to associate with those pages.

Yahoo

The first patent application I have listed for Yahoo is under their name in the US Patent and Trademark assignment database. The second isn't there yet, but it shares two authors, and is on a very related topic. They both cover very similar ground.

When a snippet, or abstract, is returned with a search result, it may be helpful to include a link in the snippet which might lead to something the searcher would want to see based upon the initial query. The search engine would try to include relevant "terms of interest" in the abstract.

These "terms of interest" could be single words, or a string of words. Some examples:

• Yahoo defined keywords and categories,
• Names - proper names, business names, organization names, place names, etc.,
• Uncommon words,
• Product names,
• Trademarks,
• Service marks,
• Titles - music titles, book titles, titles of television shows, etc.,
• Street addresses,
• Telephone numbers, and;
• Others.

These are all words or phrases likely to be used in a secondary search for information. These are called "active" terms in that they may be tied to some different actions, such as being links leading to other pages, or to Yahoo's mapping server, or a dictionary lookup, or a "voice over IP" call for a phone number, and so on.

The second patent application builds upon these active abstracts by providing additional ways that they can be used and activated. This can include translation, showing links to the most popular pages of a site, leading to reviews of products or services offered on the page, or even some results based upon information that the search engine might know about the searcher.

Search system presenting active abstracts including linked terms Inventors: Chad Carson, and Douglas Michael Cook US Patent Application 20060101012 Published May 11, 2006 Filed: June 10, 2005

Abstract

Upon receiving a search query, a search to identify at least one resource relevant to the search query is performed. At least one excerpt is extracted from the at least one resource, and a term of interest is identified in the excerpt. A link to a referral document is associated with the term of interest. Upon selection of the link, the referral document is displayed. Alternatively, the link is associated with a second search query, and the second search query is automatically performed upon selection of the link. A network telephone call can be automatically placed when a telephone number is the term of interest.

Active abstracts Inventors: Chad Carson, Douglas M. Cook, and Kalpana Ravinarayanan US Patent Application 20060101003 Published May 11, 2006 Filed: June 28, 2005

Abstract

Techniques are provided for generating search results that include, instead of or in addition to the traditional abstract links, other links to content that the user of a search engine may consider helpful. These other links are referred to herein as "non-traditional abstract links". In general, the search engine generates non-traditional abstract links that attempt to anticipate the "next move" by the search engine user. By selecting a non-traditional abstract link from the abstract of a matching resource, the user is able to navigate from the search result listing directly to locations other than the top of a matching resource.

Microsoft

Microsoft's patent application focuses on providing more efficient searches, by first putting a query into a structured form before looking to a document index at a search engine or cache files logs on that search engine or the user's browser. This process can involve the use of more than one search engine.

Application programming interface for text mining and search Inventors: Eric Brill and Robert J. Ragno US Patent Application 20060101037 Published May 11, 2006 Filed July 1, 2005

Abstract

Systems and methods are described that allow programmatic access to search engine results and query logs in a structured form. The search results can be retrieved from the search engine in an intermediary form that contains the information that is in the HTML pages provided to web browsers (potentially with additional information). This intermediary form can then be broken down on the client machine, using local resources, to assemble the structured objects. The library also provides for caching of the search results. This can be provided both on the local machine and on a remote database. When the results for a query exist in the caches, they can be retrieved from such location instead of querying the search engine. Documents and/or web pages can also be cached. The library can also be directed to operate only from the cache, effectively exposing a local data set instead of the remote search engine.

IBM

Method for organizing a plurality of documents and apparatus for displaying a plurality of documents Inventors: Zhong Su, Li Zhang, Yue Pan, Li Bai, Li Ping Yang US Patent Application 20060101102 Published May 11, 2006 Filed November 7, 2005

Abstract

The present invention relates to a method for organizing a plurality of documents and an apparatus for displaying a plurality of documents. Said plurality of documents are clustered, and the resulted clusters of different levels are displayed as virtual directories, thus helping the user to navigate to the target document quickly. The navigation may be performed with the aid of topics and abstracts. Furthermore, the user's operations may be reduced through controlling the displayed contents to be within the size of the screen.

Hitachi

With some clustering search engines, the categories created are displayed in a list of folders or a vitual directory that you can only drill deeper into one at a time. This patent shows the results of a search, in clusters, arranged on a grid so that users can see their relevance to the term or terms that were searched for. The images attached to the patent are pretty illustrative of how it would work.

Document information display system and method, and document search method Inventors: Osamu Imaichi, Tetsuo Nishikawa, Toru Hisamitsu, Makoto Iwayama, Masakazu Fujio US Patent 7,047,255 Granted May 16, 2006 Filed February 27, 2003

Abstract

The present invention visualizes the contents of a plurality of documents without a lack of the listing property. Two document units are extracted from a document database and relevance degrees between individual elements of a group of the document units are calculated. The results are displayed on a two-dimensional coordinate plane depending on the relevance degree.

NHN Corporation

The Korean search space tends to be strongly dominated by Korean search engines, as can be seen from an article in the Korea Times back in January, which discussed Why is Google Struggling in Korea. It points to the NHN Corporation's Naver as the dominant search engine. This patent application looks at some steps that may be taken by NHN Corporation involving advertising based upon keywords.

Method for generating a search result list on a web search engine Inventors: Joon Hong, Woo Sung US Patent Application 20060100988 Published May 11, 2006 Filed: February 27, 2004

Abstract

A Computer-readable recording medium in which a program for implementing a method according to any one of claims 1 to 12 is recorded. The present invention relates to a method for selling a search word to a network information provider and providing a search result list to a searcher based on the sales information if the searcher inputs the search word in an Internet search engine. The present invention includes maintaining a database for storing a plurality of search listings therein, each of the search listings including a search word and a network position associated with a network information provider, receiving the search request from the searcher, identifying a search listing having a search word corresponding to the search request, providing the identified search listing to the searcher as a search result list, if the searcher clicks a search listing of the search result list and accesses a network information provider's web page associated with the clicked search listing, receiving predetermined information on the use of the web page by the searcher, and charging the network information provider based on the received predetermined information on the use of the web page by the searcher. According to the present invention, there is provided an Internet search service method and apparatus for providing a variety of billing means to a network information provider by properly reflecting the profit made by the network information provider.

S.L.I. Systems, Inc

The intent behind this patent application is to create a unique search engine and "ameliorate the aforementioned disadvantages of conventional search engines by harnessing the cerebral power of the human operator." Pages selected in search results by users are the pages that get updated in search results. The more frequently a page is selected, the more frequently the information in the database about it gets updated. Users can also supply their own keywords to describe the content of a site, which can then be shared with others.

S.L.I. Systems is the founder and a co-owner of Eurekster, which Danny wrote about a couple of years ago in Eurekster Launches Personalized Social Search. This document appears to describe many of the features of Eurekster's swickis.

Search engine Inventors: Grant James Ryan, Shaun William Ryan, Craig Matthew Ryan, Wayne Alistar Munro, and Del Robinson US Patent Application 20060100956 Published May 11, 2006 Filed: December 21, 2005

Abstract

The present invention provides for a method of updating an internet search engine database with the results of a user's selection of specific web page listings from the general web page listing provided to the user as a result of his initial keyword search entry. By updating the database with the selections of many different users, the database can be updated to prioritize those web listings that have been selected the most with respect to a given keyword, and thereby presenting first the most popular web page listings in a subsequent search using the same keyword search entry.

Update

The CEO of SLI Systems, Shaun Ryan, sent me a friendly note correcting a couple of my statements about the relationship between SLI Systems and Eurekster. While they are a founder, they own a piece of Eurekster, and not the whole company. Thanks, Shaun.

My usual reminder about patents: Some of the processes and technology described in patents are created in house, and some are developed with the assistance of contractors and partners. A percentage are never developed in a tangible manner, but may serve as a way to attempt to exclude others from using the technology, or even to possibly mislead competitors into exploring an area that they might not have an interest in (sometimes skepticism is good.)

There are times when a Google or Yahoo acquires a company to gain access to the intellectual property of that company, or the intellectual prowess and expertise of that company's employees. And sometimes patents are just purchased.

Want to comment or discuss? Visit our Search Technology & Relevancy area of the Search Engine Watch Forums.

Posted by Bill Slawski at 5:53 PM | Permalink

May 11, 2006

NetworkWorld Talks With Raul Valdes-Perez, CEO of Vivisimo & Clusty

Gary Price points to a NetworkWorld.com interview with Vivisimo CEO Raul Valdes-Perez, of Clusty Search. Clusty uses clustering technology to provide results. The NetworkWorld author was convinced, during his interview with Mr. Perez, that Google/Yahoo/MSN provide "incomplete results." Is clustering the future of search? Give Clusty a try for yourself. Also, for past articles at Search Engine Watch on Clusty click here.

Posted by Barry Schwartz at 8:49 AM | Permalink

March 17, 2006

Clusty Adds Search Feature To Help Cluster Results For End User

Clusty, a search engine that is focused on clustering results, has added a new feature according to ResourceShelf that makes viewing clustered results easier. Gary explains that if you search on site:mil iraq reports at Clusty, you will notice the clustered results on the left side. Currently highlighted in red is "Air Base, Iraq", and if you scroll your eyes to the right bottom, middle portion of the page, you will notice a new link named "Search for more results like these." Clicking on that link brings you to a search query which basically added the "Air Base, Iraq" string to your previous query, better refining your search.

Want to understand this query better, click on "details" on the top line that reads; "Top 150 results of at least 465 retrieved for the query mil iraq reports Air Base, Iraq (Details)". You will see how comprehensive Clusty can be.

Posted by Barry Schwartz at 11:07 AM | Permalink

January 13, 2006

Google Testing New Refinement/Clustering Links -- Sign Of Google Base In Web Search?

Maybe it's new -- maybe not, but another Google query refinement test is out in the wild, as covered in our SEW Forums thread Google Refinement Keywords. There are links to screenshots showing that after a search, a "Quick results" area appears at the top of the page with links that seem to narrow refine results into clusters/topics. Something similar was seen last month.

The topics and implementation resemble what happens after a search at Google Base , such as with apple, which brings up this at the top of the page:

Refine your search:  products   housing   homes for sale   apartments   wanted ads   services  

It makes me wonder if the refinement is going along with some Google Base material flowing into the main web index as part of a test.

Posted by Danny Sullivan at 9:53 AM | Permalink

November 30, 2005

Clusty Launches Japanese Version of Its Metasearch and Dynamic Clustering Service

Word from Pittsburgh today that Vivisimo (the company that powers Clusty, the metasearch/dynamic clustering engine) have just released a version of for Japan. Clusty Japan is available at Clusty.jp

The new engine will deliver custom crawls of Japanese news sites and Japanese Wikipedia as well as meta-searches of Web crawlers and then clusters the results for easier navigation.

I've posted the complete news release about the launch here.

Posted by Gary Price at 12:09 PM | Permalink

June 23, 2005

More on Grokker's Visual Search Results

When Grokker released its free online I Grok search tool last month, Gary blogged a brief post about it. Essentially, I Grok clusters Yahoo search results into categories, and represents these visually on a page.

That's cool, but I Grok actually does a lot more than that, and I've found the service to be useful in a number of different ways. Want to know more? Click on to read today's SearchDay article, Visualizing Yahoo Search Results.

Posted by Chris Sherman at 9:24 AM | Permalink

May 2, 2005

Clusty and Indeed.com Partner to Offer New Service

The new and impressive job listings metasearch database, Indeed.com and Clusty, the metasearch engine offering dynamic clustering of results from Vivisimo that debuted last September, have gotten together to offer Jobs.clusty.com. It allows the searcher to dynamically cluster job search results from Indeed.com several different ways.

The Jobs.clusty.com homepage presents all Indeed.com listings clustered by state. You can also keyword search the database and then cluster results by topic, company, cities (location), or source (the underlying database or site where Indeed.com gets the listing from). To change the cluster, simply use the pull-down menu option in the left column of a results page.

Jobs.clusty.com might not only be useful service for people looking for employment but also to business and competitive intelligence researchers.

About six weeks ago, Clusty launched another specialty search tool, Gov.clusty.com, that clusters results from U.S. Government "sources" including First.gov, an MSN search limited to .gov sites, and others sites and databases.

Posted by Gary Price at 10:29 AM | Permalink

April 6, 2005

Google Testing New Related Searches & Clustering System?

Over at Search Engine Lowdown, Andy reports that a member of his team has possibly discovered a new feature that Google might be testing. Then again, it could be adware.

So what is Google possibly testing? As Beal explains (screenshots provided), an icon at the bottom of web results pages labeled "try alternate searches" allows the searcher to open a new page with searches for related terms. For example, a search for "desktop computer" offers a link to a page of "Try alternate searches" for the queries:

+ computer support + computer buying guide + free desktop wallpaper + desktop computer prices

The first three results for each search are provided along with a link to get more results.

I ran a few Google searches but was unable to duplicate the results. Not a surprise however since Google is likely testing with a small group of users. However, as Andy notes at the end of his post, this could all just be adware. Stay tuned?

Postscript: Google has confirmed that they are testing the alternate results feature.

Posted by Gary Price at 12:23 PM | Permalink

January 27, 2005

FyberSearch Now Offers Results in RSS

Nathan Enns, the developer of FyberSearch has let us know about few a new services and advanced features. + Results are now available in RSS format. + Domain Clustering: no more than one result per domain name is displayed unless the user specifies. + The keyword density option has been improved. It allows you to view pages that contain your search terms at least "X" number of times.

More on the Fybersearch blog.

Posted by Gary Price at 12:23 PM | Permalink

January 11, 2005

Web Search Clustering from Microsoft (and other Clustering Tools)

Yesterday, we blogged about a discovery that allows you to receive MSN Search Beta results via RSS. It will be interesting to see what Bill G. and company does with this feature in the future.

Today, even more MS search news. An excellent post on the SQL Full Text Search blog (thanks John!) alerts us to a few clustering tools, including a demo from Microsoft Research Asia.

+ Microsoft Research Asia (MSRA) Cluster Demo and Toolbar You can either cluster results from MSN, MSN Beta, or MSN Newsbot using a web interface. Browsing inside of a clusters seems to work better with IE compared to when I tried using Firefox. A toolbar from MS Research Asia that offers direct access to the MSRA clustering technology is also available. It's also very easy to toggle the clustering on and off. Techies might find a paper by the MSR Asia team worth a look: Learning To Cluster Web Search Results. The research was presented at a conference last summer.

+ Dowser An app (Windows, Mac, Linux, Unix) that not only clusters results but also locally caches every page viewed in your browser and offers change tracking. I'm looking forward to testing this one.

+ Top Gist An app (download required, Windows only) that allows you to cluster results from several engines including Google.

OK, we've all got plenty to take a look at. I'm looking forward to testing these tools out and reporting back.

Final Comments As web databases grow larger and larger and search habits stay pretty much the same, I believe clustering will become an even more important tool to help the searcher find what they're looking for since we all know that good results can exist after the first 10 (what most people look at). Clustering can also be an excellent knowledge discovery tool that can help you quickly "see" things that would be difficult, if not impossible to quickly piece together looking at one result at a time. In 2004, both Chris and I had positive things to say about Vivisimo's release of Clusty. At the Web 2.0 conference Google demonstrated some of the clustering work their doing.

I was also happy to see that the SQL Full Text Search post includes a link to Vivisimo's excellent paper: Needed: A More Selective Ignorance. It does an excellent job (IMHO) to succinctly explain the importance of clustering.

Ok, that's it. Go forth a cluster!

Posted by Gary Price at 11:58 AM | Permalink

See More Posts From:

This Week | This Month

 

• Send us feedback
|
• Technical questions or bug reports?
|
• Newsletter problems?
|
• Receive news via SEW RSS feeds

• Incisive Interactive Marketing LLC. 2009 All rights reserved.
| Legal Notices, Licensing, Reprints & Permissions |
• Privacy Policy
|
• Affiliates

var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www."); document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E")); var pageTracker = _gat._getTracker("UA-564586-7"); pageTracker._setDomainName(".searchenginewatch.com"); pageTracker._trackPageview(); window.collarity_appid = "incmedia"; //> //>

• Send us feedback
|
• Technical questions or bug reports?
|
• Newsletter problems?
|
• Receive news via SEW RSS feeds

• Incisive Interactive Marketing LLC. 2009 All rights reserved.
| Legal Notices, Licensing, Reprints & Permissions |
• Privacy Policy |
• Affiliates