http://news.google.com/intl/en_us/about_google_news.html
Google News is a computer-generated news site that aggregates headlines from more than 4,500 English-language news sources worldwide, groups similar stories together and displays them according to each reader's personalized interests.
Traditionally, news readers first pick a publication and then look for headlines that interest them. We do things a little differently, with the goal of offering our readers more personalized options and a wider variety of perspectives from which to choose. On Google News we offer links to several articles on every story, so you can first decide what subject interests you and then select which publishers’ accounts of each story you’d like to read. Click on the headline that interests you and you'll go directly to the site which published that story.
Our articles are selected and ranked by computers that evaluate, among other things, how often and on what sites a story appears online. As a result, stories are sorted without regard to political viewpoint or ideology and you can choose from a wide variety of perspectives on any given story. We'll continue to improve Google News by adding sources, fine-tuning our technology and providing Google News to readers in even more regions.
Features
Personalized news: No one can read all the news that’s published every day, so why not set up your page to show you the stories that best represent your interests? Learn more.
Alerts: You can sign up to receive weekly, daily or as-it-happens email alerts on any topic that interests you. Learn more.
News for your mobile phone: If you have a mobile data plan, you can access a special version of Google News right from your mobile phone. Learn more.
Feeds: You can receive updates of various sections of Google News or Google News search results by subscribing to RSS or Atom feeds in your favorite feed reader. Learn more.
News archive search: Search and explore information from historical archives dating back over 200 years. Learn more.
More Information
Look for more information, find answers to your questions, and send us suggestions in our Help Center pages.
If you're a news publisher, look for answers to your questions and useful tools in our section Help for Publishers.
Thursday, April 10, 2008
Wednesday, April 9, 2008
AMAZON RECOMMENDER WEEK 11
This week I tested Amazon simply by typing key words such as "disk" and try to see how smart Amazon was in finding things I wanted that related to photography memory card. Out of 546 amazon returned about 8 items which were photography related san disk memory card. This tells me Amazon does not know me too well when it comes to simply having it guess at key words.
I tried also typing "hood" and out of 48 items it recommended one photography camera hood. Also not good.
I then tried "filter" it out of 48 it returned 4 items. Not bad but then not very good.
Overall it seems as if Amazon knows I'm interested in photography but can't read my mind.
I tried also typing "hood" and out of 48 items it recommended one photography camera hood. Also not good.
I then tried "filter" it out of 48 it returned 4 items. Not bad but then not very good.
Overall it seems as if Amazon knows I'm interested in photography but can't read my mind.
Wednesday, April 2, 2008
MEDIAUNBOUND
http://www.mediaunbound.com/index.ph
MediaUnbound a company that provides software application to help provide music recommendation through use of different techniques. Company clients includes Napster, Rhapsody, and others. Unlike other recommendation Mediabound utilizes music analyst or human along with technology to make recommendations. Recommendations in Rhapsody or URGE is made more on the human side where computer can't analyze such as mood and tastes. This site mainly explore the company and it's technology and product.
Upong stumbling upong the company I had some questions about the company so emailed a contact at MediaUnbound. Turns out the CEO responded. MediaUnbound is a combination of content and collaborative filtering recommender system. It analyzes personal preference to make accurate and interesting recommendations.
One example of MediaUnbound's work.
http://www.videovalve.tv/
Utilizes elements from URGE system.
MediaUnbound a company that provides software application to help provide music recommendation through use of different techniques. Company clients includes Napster, Rhapsody, and others. Unlike other recommendation Mediabound utilizes music analyst or human along with technology to make recommendations. Recommendations in Rhapsody or URGE is made more on the human side where computer can't analyze such as mood and tastes. This site mainly explore the company and it's technology and product.
Upong stumbling upong the company I had some questions about the company so emailed a contact at MediaUnbound. Turns out the CEO responded. MediaUnbound is a combination of content and collaborative filtering recommender system. It analyzes personal preference to make accurate and interesting recommendations.
One example of MediaUnbound's work.
http://www.videovalve.tv/
Utilizes elements from URGE system.
AMAZON RECOMMENDER WEEK 10
For the past two weeks instead of searching for photography related items I tried searching soemthing new to see how good Amazon can give me recommendation. I find GPS systems very interesting so decided to test Amazon. My search consisted of car GPS units by TOM TOM, GARMIN and other companies. I was particularly interested in TOM TOM because I had one. I wanted to see what Amazon had to offer. I noticed Amazon was trying to sell me more than just GPS. The recommendation that came out of the search inluded books such as Washing Post Dining Guide, Innoncent Man by Tom Selleck or Magnum PI. All the books were on the last recommendation page. I figure Amazon wanted to present something that was of interests to me and keep my intrest while trying to sneek some books in. I've never bought a book from Amazon so maybe it was trying to sell me or get me to start buying books.
During my one week searching for GPS related equipment I noticed Amazon too picked up on what I wanted to see. When I simply type street map Amazon presented me with car unit GPS devices along with GPS maps. With only one week Amazon picked up on my taste as it did with photography. In my mind I knew I wanted to see some GPS units when I type in street map. I wanted to see how smart Amazon was. I have to say simply by typing closely related words Amazon was able to present me at least with a page or two of items that I wanted to see. Again, I think it has to do with keeping my intrestes alive while trying to sneak in something new.
During my one week searching for GPS related equipment I noticed Amazon too picked up on what I wanted to see. When I simply type street map Amazon presented me with car unit GPS devices along with GPS maps. With only one week Amazon picked up on my taste as it did with photography. In my mind I knew I wanted to see some GPS units when I type in street map. I wanted to see how smart Amazon was. I have to say simply by typing closely related words Amazon was able to present me at least with a page or two of items that I wanted to see. Again, I think it has to do with keeping my intrestes alive while trying to sneak in something new.
Thursday, March 27, 2008
MY AMAZON RECOMMENDER SYSTEM
In the past weeks I've searched for items other than just photography.
I've realized that my recent searches have simply been photography related and so Amazon was unable to give me much new recommendation on other items that would be of interests to me.
I realized that Amazon.com has a good handle when it comes to photography.
I think it has figure me out fairly well.
By simply typing NIKON I get interesting photography items that is of interests to me.
I realized too since I've been searching for expensive fast lens the lens recommendation that I'm getting are not the average lens.
I'm getting professional lens that are expensive and fast.
I've been testing the system on words words relating to photography but may not be obvious.
For example, I would search for REMOTE and Amazon knowing I've search so much on photography it tosses a Camera Remote along with standard Television remotes.
Out of the 48 items Amazon typically recommends one is photography related.
Again, I test the sysem on common word such as BAG. Amazon was still sneak in 2 photography carrying bags along with purse and other carrying on bags. Another search was simply typing in SOFT. Amazon again was able to throw in a soft nikon camera bag.
Instead of searching for specific photography related words I'm just tossing out words that may have very little correlation to photography. Though most recommendations are not what I'm after at least one or two is items I want to look at and possibly buy or rate it.
I've realized that my recent searches have simply been photography related and so Amazon was unable to give me much new recommendation on other items that would be of interests to me.
I realized that Amazon.com has a good handle when it comes to photography.
I think it has figure me out fairly well.
By simply typing NIKON I get interesting photography items that is of interests to me.
I realized too since I've been searching for expensive fast lens the lens recommendation that I'm getting are not the average lens.
I'm getting professional lens that are expensive and fast.
I've been testing the system on words words relating to photography but may not be obvious.
For example, I would search for REMOTE and Amazon knowing I've search so much on photography it tosses a Camera Remote along with standard Television remotes.
Out of the 48 items Amazon typically recommends one is photography related.
Again, I test the sysem on common word such as BAG. Amazon was still sneak in 2 photography carrying bags along with purse and other carrying on bags. Another search was simply typing in SOFT. Amazon again was able to throw in a soft nikon camera bag.
Instead of searching for specific photography related words I'm just tossing out words that may have very little correlation to photography. Though most recommendations are not what I'm after at least one or two is items I want to look at and possibly buy or rate it.
GETTING TO KNOW YOU:LEARNING NEW USER PREFERENCES IN RECOMMENDER SYSTEM
http://www.iuiconf.org/02pdf/2002-001-0019.pdf
Paper talks about different strategies to learn about new users.
Experiment utilizes MovieLens dataset both online and offline.
Paper talks about different strategies to learn about new users.
Experiment utilizes MovieLens dataset both online and offline.
Wednesday, March 19, 2008
TRIBLER.ORG ( RECOMMENDER RESOURCE)
http://www.tribler.org/
Tribler.com com is a software that people can download free of charge.
It's a P2P social application to help people find contents such as videos, audios, pictures and etc.
Main screen shot of application.
http://www.tribler.org/screenshots/8
Files overview screenshot.
http://www.tribler.org/screenshots/4
Persons overview screenshot.
http://www.tribler.org/screenshots/6
Tribler.com com is a software that people can download free of charge.
It's a P2P social application to help people find contents such as videos, audios, pictures and etc.
Main screen shot of application.
http://www.tribler.org/screenshots/8
Files overview screenshot.
http://www.tribler.org/screenshots/4
Persons overview screenshot.
http://www.tribler.org/screenshots/6
Subscribe to:
Posts (Atom)