NewsSeer is both a straightforward news search engine and an adaptive tool that's constantly learning your interests to deliver personalized news tailored to your own needs.
Recently, I had the chance to chat with Dr. Steve Lawrence, NewsSeer's creator. Lawrence works at NEC Research and is highly regarded (with very good reason) in information retrieval circles for his research, writing, and the creation of some very useful and important tools, including Inquirus and ResearchIndex.
NewsSeer is one of Lawrence's recent projects. Very little documentation is currently available for the service. Here's a brief overview to get you started.
NewsSeer has been around for several months. Recently, Lawrence made some major cosmetic changes to the page layout including adding logos to help identify sources.
NewsSeer currently crawls about 30 well-known news sources, listed on the lower left side of the home page. You can expect this number to increase in the near future.
News sources are recrawled for new content every 5 minutes. New material is added to the database immediately.
The crawl is not deep. This means that if an article is not linked directly from one of the pages NewsSeer is crawling, it will not be discovered and added to the database. According to Lawrence, news stories are archived for about 30 days. However, I did find some older material.
Every entry includes a time stamp indicating when the story was first discovered and placed in the database. Examples: 15M is 15 minutes, 3h is 3 hours, 5d means the article was crawled for the first time 5 days ago.
Searching NewsSeer's database is very straightforward. The search form is located at the top of the home page and will automatically return results organized by relevance and date. In the future, Lawrence plans further enhancements, including the ability to view keyword search terms in context.
NewsSeer employs automatic phrase detection so using quotation marks is not required. Boolean searching is not available at this time.
NewsSeer: Learning From You, Recommending to You
In addition to its search capabilities, NewsSeer will attempt to learn your interests by using several criteria from the material you select to view. These criteria include article selection, the text of the article, how long you looked at a story, the source of the material, the age of the story, and so forth. This process is automatic. However, you can also choose to assist NewsSeer by rating your interest in the story.
Let's look at the layout of the NewsSeer home page.
The left side of the page is where you'll find current news organized by relevancy and by time. You'll also find options to rank news sources (you can always change them), change the font and point size of the page, view stories in a separate browser window, and change your source rankings.
To eliminate a story from the list, simply select the x to the left of the story title. After viewing a story (you may need to reload the NewsSeer page), you'll see four boxes to rate your interest in the story, ranging from no interest to high interest. After making your selection, your preferences are sent to the NewsSeer computer.
The only personal information stored on your local computer is a cookie enabling your personal page to automatically load. All material appearing in the left column is generated from your NewsSeer profile.
The right side of the page organizes content via many different criteria. You'll see recent stories from sources that you've given a high interest rating, the most popular stories on NewsSeer, stories related to the most recently viewed article, and direct links to the four most recent stories that the crawler has added to the database.
A box at the bottom of the left column lets you sign up to have "relevant" news stories emailed directly to you. This additional service is completely optional and is also free.
If you use several computers and want to use your profile on all of them, simply go to the bottom of the left column and select, “Access profile on another computer or browser” link. You'll be provided with a specific url to access your page from any web-connected computer. You can also send this link to yourself or others via email.
Remember, NewsSeer is a relatively new, experimental project. To get the most utility from the service, use NewsSeer enough so that it can observe your news viewing habits, allowing it to learn about you and develop a profile based on your own personal interests.
Steve Lawrence and his colleagues at NEC Research have a well-deserved reputation for turning out interesting and useful tools. NewsSeer is a project worth keeping an eye on even if you don't use it on a daily basis.
An experimental news search engine from Dr. Steve Lawrence.
Newsblaster: An Automatic Weblogger
SearchDay, Mar. 6, 2002
Newsblaster is a great new tool for news junkies, and also points the way toward some seriously cool automated web harvesting technologies that will be a boon to searchers.
Google News Search Leaps Ahead
SearchDay, Oct. 2, 2002
Google has dramatically enhanced its news search service, serving up a portal of real-time news drawn from more than 4,000 sources worldwide.
Gary Price (Gary Price Library Internet Research Consulting) is the author of the essential weblog for searchers, The Virtual Acquisition Shelf & News Desk.
NOTE: Article links often change. In case of a bad link, use the publication's search facility, which most have, and search for the headline.
This Year's Premier Digital Marketing Event is #CZLSF
ClickZ Live San Francisco (Aug 11-14) will bring together the industry's leading online marketing practitioners to deliver 4 days of educational sessions and training workshops. From Data-Driven Marketing to Social, Mobile, Display, Search and Email, the comprehensive agenda will help you maximize your marketing efforts and ROI. Register today!