 |
Please type your question or key words below  |
|
|
User Verdicts
| Robert C. May, Business Pundit, USA |
Lexxe Beats Google
I've spent the last few weeks trying different search engines. But instead of doing just one search, I try a new engine for an entire week. I only use Google if I can't find what I'm looking for. This week has been Lexxe, and I would have to say it is far from perfect, but still beats Google by a mile. The results are much more relevant, and clustering helps you quickly narrow your search. Very nice. [Jan. 12th 2006] |
| http://www.businesspundit.com/50226711/lexxe_beats_google.php |
| Beverley Head, Sydney Morning Herald, Australia |
Giant Leap for the Search Unknown
To gauge how different search engines work, try the classic primary schooler's question, "Why is the rainbow curved?" in Lexxe, Yahoo!, Google, Baidu, LookSmart and Live.com.
One question - many answers. For my 10-year-old, Lexxe's early version natural language search was far and away the most useful and contextually appropriate, followed by Baidu, Google and Yahoo!.
[Sept. 11th, 2007] |
| http://www.smh.com.au/news/technology/giant-leap-for-the-search-unknown/2007/09/10/1189276633425.html |
| Don Dodge, Microsoft Emerging Business Team, Don Dodge on The Next Big Thing, USA |
The Top 100 Web Search Engines No One Ever Heard Of
Entrepreneurs are trying everything they can think of to beat Google, Yahoo, and Microsoft Live. Here are some of the approaches and the companies that are trying... Technology advancements include; Hakia , Powerset , Lexxe, Mindset, Surfwax, Swoogle, which search based on meaning or intent.
[Feb. 2007] |
| http://dondodge.typepad.com/the_next_big_thing/2007/02/index.html |
| Orhan Karsligil, Ph.D. MIT, Research Diary, USA |
Intelligent Search Engines
For the longest time I was looking for a search engine that would make my life easier for Research related searches. Google is great but it does not always return focused results especially if the site I am looking for is at the fringes of the web and does not have too many connecting links. I started using Lexxe recently and for now I am pretty happy with the results. I can focus my search based on clusters, a functionality I have seen before in search engines like Teoma and Norhtern Light years ago. Lexxe can also process natural language. That might be useful for some people, especially if the results are also formulated in natural language rather than set of links. MIT AI lab had the Start project that did it years ago. It is good to see commercial products coming out from similar research.
[Nov. 28th, 2006] |
| http://karsligil.homeip.net:8000/~okarsligil/wordpress/?p=146 |
| Richard MacManus, SearchReadWriteWeb.Com, New Zealand |
"Search 2.0 vs Traditional Search"
Lexxe
does what TSE's already do, but more efficiently. Lexxe is designed to extract short
answers on-the-fly, instead of finding the page on which the answer might be located. It
emphasizes the processing of language rather than symbols - using the level of words and
the meanings associated with them.
Key Feature: Linguistic Search
How is it useful compared to TSE (Traditional Search Engine)?: Although they claim to be "50% more accurate and
relevant than any other search engine, including google", I'm not convinced. However,
they do have mechanisms in place to determine fairly accurate answers for short questions,
compared to Google. For example: Who was Louis-Nicholas Vauquelin? Compare
Google's answer to
Lexxe's. [Jul. 21st, 2006] |
| http://www.readwriteweb.com/archives/search_20_vs_tr.php |
| Mark Buzinkay, http://www.buzinkay.net/blog.html, Austria |
Lexxe - die 'natürliche' Suchmaschine (Lexxe - the "natural" search engine)
Lexxe is easy to describe - enter a question and receive an answer. By concept, Lexxe is a search engine to understand natural language. Let's try this. We enter 'Who is Gerhard Berger?'. The SERP (Search Engine Result Page) lists various topic-oriented cluster ('Formula 1 driver' e.g.) on the left side as well as a bunch of relevant links. If you click on of the cluster (and confirm again your choice) you receive search results only from that cluster.
Let's face Lexxe with results from Google. The keyphrase 'Gerhard Berger' produces a lot of hits (around 3 Millions), but almost all of the first 10 are related to the Formula 1 driver Gerhard Berger. Why should we use Lexxe when Google is highly relevant too? The first thing I see is the possibility to cluster. That's important, if I conclude a query with an ambiguous term. The second argument for Lexxe: you receive immediately the most important (and very relevant) content excerpt for your query in the SERP.
The concept of Lexxe will be more visible with a query like 'Who was Nietzsche'. As expected, Google produces a very long list of relevant links, but there is no topical order. In contrast Lexxe which is clustering:
• German Philosopher of the Late 19th
• Superman (Übermensch)
• Minister
• Mother
• Reader
• Clergyman
• Author
• Etc.
I'm interested in his relationship to his mother and get more relevant links to his biography, to his medical history, to his mother etc. That's the true strength of Lexxe - topical drilling. [May. 23rd 2006]Special thanks to author Mark Buzinkay, who also translated his comments on Lexxe from German to English. |
| http://www.buzinkay.net/2006/05/lexxe-die-natrliche-suchmaschine.html |
| Sharon Betts, TechTwaddle, USA |
| Need a factoid? |
|
Lexxe is a new type of search engine — a web 2.0 engine. Just enter your question and wait for the clustered results. Although in alpha, this is a great new tool for students and laymen alike.
Lexxe is designed to extract short answers quickly, instead of finding the page on which the answer might be located. It emphasizes the processing of language rather than symbols - using the level of words and the meanings associated with them. I am a crossword puzzle buff and Lexxe is my last resort to solve those testy New York Times Sunday editions. [Aug. 22, 2006] |
| http://sbetts.edublogs.org/2006/08/ |
| Craig Burton, Craig Burton Weblog, USA |
| Natural Language Search Engine |
|
I stumbled on this site today. I entered "When did NetWare first appear?" and it came back with a whole list of links and powerpoint presentations saying NetWare first appeared in 1983.
The first versions of NetWare came out in 1981, but hey, who's keeping track? So far I like this engine.
If you type the same sentence in Google, it comes up with links to Novell and info on the current version of NetWare. [Aug. 21, 2006] |
| http://www.craigburton.com/2006/08/21#a9659 |
| Mohit Kishore, http://mohitkishore.blogspot.com/, India |
|
It looks like Google has some competition from the new kid on the block - Lexxe, which claims to be powered by an advanced natural language technology. The USP is that Lexxe can understand natural language queries that begin with 'what' 'who' where' etc. Lexxe also organizes results by clusters on the left of the screen. In addition Lexxe functions as a normal key word search engine too. I have been using Lexxe for a while now, and it actually seems smarter than google.
To test its natural language processing abilities, I tried "Why is it cold in winter?" and it actually came up with something about the "northern hemisphere being pointed away from the sun"... not a perfect answer, but certainly more than I expected for an inane question like that!
The more one thinks about Google's whole page rank concept (quality of a site determined by the number of incoming links and outgoing links), the more one is convinced that it is reaching the end of its leash. Since everyone has figured out how google works, webmasters know all the loopholes in the book to ensure good page ranks. Maybe its better for search engine companies to not reveal how they actually work. [Apr. 27, 2006] |
| http://mohitkishore.blogspot.com/2006/04/lexxe.html |
| Tara Calishain, ResearchBuzz.org, USA |
Lexxe -- Natural Language Search Engine With Clusters
Hat tip to PTJ for pointing me to Lexxe, a natural language search engine that also offers results in clusters. Lexxe is in alpha and available at http://www.lexxe.com/.The front page has a query box and also some sample, Olympics-related questions that you can ask. I didn't choose any of the available ones, instead asking What is Skeleton?. Despite the fact that the word "skeleton" makes it a bit of trick question, Lexxe gave as its first result the Wikipedia entry for Skeleton-the-sport. There was also a dictionarylike "Answer" at the time that gave more traditional definitions of "Skeleton".Clusters show up on the left. The clusters for this particular question include "Human Bones," "Olympic Sport", and "Anatomy". Unfortunately some of the clusters also looked fairly useless: ".Search", "Results", and "World's". A couple of them were cute ("Key" and "Closet".) Results include title, snippet, URL, and page size; click on a cluster and the search is rerun with the cluster words included. For example, Run What is Skeleton? then click on the Human Bones cluster, and the search will be rerun as What is Skeleton Human Bones. There is some goofiness going on with what words are included in the search; What is Skeleton? Human Bones Human Skeletal had more results than What is Skeleton? Human Bones. You can "drill down" through clusters, though Lexxe seems to get slower and slower as you do so. I got all the way to What is Skeleton? Human Bones Human Skeletal Anatomy Bone Yahoo and decided things were getting too weird.I noticed that some of the Answers at the top of the page were not very useful. For one search result my answer was "skeletal system the following links will allow you to access real photographs of the human skeletal system" with no link. Just a sentence fragment. thought the results of Lexxe were pretty good. You have to be careful choosing your clusters, though, they vary a lot in quality. Further, drilling down more than one or two clusters deep doesn't seem to offer a lot of good. [Feb. 23rd, 2006]Reproduced with permission of ResearchBuzz ( http://www.researchbuzz.com ). |
| http://www.researchbuzz.org/2006/02/lexxe_natural_language_search.shtml |
| Uri Hurvitz, Jerusalem, Israel |
|
Hello to the development team at Lexxe, I have been interested lately in a subject that seemed to be quite searchable in the Web - "the value of mentoring for students of music". I was disappointed in all the searches I was able to make by using the usual engines. Then, tonight, I happened to find a reference to your site in eHub. Not expecting too much, I posted my topic as a question and got a hundred relevant references. This result is really outstanding by any criteria. I am looking forward to similar surprises from using your search engine, as well as to learn about the technology that you have developed (I'll look in the site and beyond). And I'll spread the word. What you did is a very basic and important tool for people like me.Thank you for the wonderful surprise.With best wishes for your success,Uri Hurvitz
Jerusalem, Israel
[Feb. 2nd, 2006] |
| Via email to Lexxe |
| Arion Londraville, Use Bombs, USA |
The Wonderful Wizard
Tonight Brett and I had some fun playing with Lexxe; a new search engine. It seems like a small scale operation and runs rather slowly.
A main concept to this tool is that it will answer questions you type in, preferably based in fact. Lets see how it did with the questions we threw at it.
Q: How many feet are there between earth and the moon?
A: figure the amount of feet between earth and the moon and you have the kind of statistic that gives statistics a bad name
Fair enough
Q: How much do eye lasers cost?
A: $40000
Remember that around next Christmas folks. *hint - hint*
Q: how much do hookers cost?
A: $1000
Considerably less.
Q: How do I kill Bush?
A: Shooting Him on the Street or Blowing Him Up with a Car Bomb
Ah, the first accurate answer of the night.
Q: How did this all come about?
A: Divine Revelation As They Maintain
Seems this search engine believes in Intelligent Design. Flying Spaghetti Monster anyone?
Q: What is the URL for Baskin Robbins?
A: Free Ice Cream
Sounds good to me.
Q: What is the website for Baskin Robbins?
A: baskin robbins ice cream flavors cakes milk shakes sorbets sherbets and frozen beverages are available from 4,
I feel disillusioned.
Q: When will I die?
A: 1/13/02 10:52am
(Other answers included “01-03-2003 2:54″ and “Tuesday Maybe.”)
That’s a bit foreboding …and approximate.
Q: Who will I wed?
A: Heather Mills in Ireland
Q: Why not Heather Mills in the USA?
A: Sir Paul Mccartney
Yes, he always has had plans to throw a wrench between the two of us.
Q: What can you tell me about Heather Mills in Ireland?
A: heather mills heather mills http://www
Well, we drove it into convolutions. Care to see if you can do the same? Feel free to post your results in the comments below.
[Jan. 6th 2006]
|
| http://www.use-bombs.com/archives/2006/01/use-bombs/the-wonderful-wizard/ |
|
|
