Refining Twitter: how to filter out (or search for) tweets by specific keywords — using Tweetdeck

Using Tweetdeck, you can hide tweets if they contain words you specify — and, conversely, set up filters like a search, to show only tweets showing specific keywords. There are two main ways of doing this and, on the day of the iPad2 goes on sale in the UK, I’m using ‘iPad’ as the keyword to filter out or (Apple fans, please note) search for.

Filter out anything you don’t want to see from Twitter

One way is to set a filter to affect everything in Tweetdeck; this applies to all columns and accounts. In the settings, look for the Global Filter menu — and type in the relevant word(s). You can also filter out tweets by people and source. Farewell those unwanted updates from Foursquare or Paper.li, perhaps.

To filter out tweets from all columns/accounts, use the Global Filter

To filter out tweets from all columns/accounts, use the Global Filter

The other, more selective way is to apply a filter to a chosen column — which you can also use as a ‘positive’ filter to show only tweets as specified.

Filter columns for specific words in Twitter

Look for the row of icons at the foot of the column you wish to filter or search, and click on the filter icon (an arrow curving down to a line). Using the default settings that then appear, you can type in a word or other text to exclude. To remove a filter, click the ‘x’ to the right.

Use the column filter to hide tweets

Use the filter to hide tweets containing specific words

Use column filters to find relevant tweets

Finally, the small drop-down menus in a column filter also allow you to search for tweets containing specific words or other text — simply change the minus sign to a plus. This ‘positive filter’ can be a useful shortcut, eg to hunt down a tweet you glimpsed and need to find again, or quickly to show particular tweets or only those with links (filter for ‘http’).

Use a column filter to show only specific tweets

Use a column filter to show only specific tweets

You can also filter by name, source or time of tweets instead of text. The column filter provides additional flexibility when used with a search column, eg to remove (old-style) retweets from a search on a particular hashtag (filter out ‘RT @’).

Providing the information you didn’t know you wanted — Google CEO Eric Schmidt on newspapers, monetisation and the semantic web

Snippets from a Wall Street Journal interview with Schmidt:

Says Mr. Schmidt, a generation of powerful handheld devices is just around the corner that will be adept at surprising you with information that you didn’t know you wanted to know. “The thing that makes newspapers so fundamentally fascinating—that serendipity—can be calculated now. We can actually produce it electronically,” Mr. Schmidt says.[…]

On one thing, however, Google is willing to bet: “The only way the problem [of insufficient revenue for news gathering] is going to be solved is by increasing monetization, and the only way I know of to increase monetization is through targeted ads. That’s our business.”[…]

“As you go from the search box [to the next phase of Google], you really want to go from syntax to semantics, from what you typed to what you meant. And that’s basically the role of [Artificial Intelligence]. I think we will be the world leader in that for a long time.”

Read more here [link]

Cervical cancer vaccine, online news, Google and SEO

SEO expert Malcolm Coles kicked off an interesting experiment yesterday, to shift the emphasis in Google’s search results away from “negative and inaccurate information” (eg some news stories) linking a girl’s death to the cervical cancer vaccine and towards NHS pages about the vaccine.

More by Malcolm here about the tendency of some news stories to suggest (or make) a connection between the death and the vaccine.

He has been encouraging bloggers and others to publish web links, with relevant linked text, to influence Google’s search results, such as cervical cancer jab, cervical cancer vaccine, and cervical cancer vaccine Q&A.

So far, the NHS seems to have bought ‘sponsored links’ against some search key words, but I don’t see any of the NHS sites in the first page of Google’s search results for “cervical cancer jab”, which continues to be dominated by news stories.

Two delicious tools: improved search, and an online portfolio

First, del.izzy, which addresses one limitation of the standard delicious search, enabling you to search all of the content of the pages you bookmarked. But they claim they need your password for this.

Second, a clever way of setting up an online portfolio on delicious. Michele Martin outlines how it works, using the optional tag description field to head the page with an introduction, and then tagging anything you wish to show up there.

A neat idea: not the most beautiful, but it works, and is easy to update. It has two other benefits, says Michele Martin:

  • The del.icio.us feature that shows how many other people saved the item acts as a kind of “recommendation” system. […]
  • If people sign up for the RSS feed to this tag, they can automatically be notified when I add new items to my portfolio.

And then of course there’s the RSS feed to do other things with, if you want to take it one step further and embed that somewhere, have it post automatically to a blog… etc

New image search tutorial from Intute and TASI

Worth a look. I find the interface clunky but it's a useful resource to which to point people who might not know their way around this area (from the Intute blog):

"Internet for Image Searching is a new, free online tutorial to help staff and students in universities and colleges to find digital images for their learning and teaching:

http://www.vts.intute.ac.uk/tutorial/imagesearching/

[…]
The emphasis of the tutorial is on finding copyright cleared images which are available free; facilitating quick, hassle-free access to a vast range of online photographs and other visual resources."

Read more here [link]

Echo-bloggers and blog search: does every linked or cited article count?

One article or many? Scope for some social network analysis on Google’s new blog search, suggests edublogger Stephen Downes, using an example from online journalism to make his point:

[But] here’s where the network analysis comes in – if the WSJ releases an opinion piece, and it is dutifully cited by the same 79 blogs that cite all such pieces of that political bent, should that really count as ’79 results’? Or is it just one opinion – the WSJ’s – repeated by echo-bloggers 79 times?

And then what does that mean for ranking, linking, position in other search engines’ results etc…?

In response to a comment, Downes explains what he envisages:

Google groups blogs by topic; take the groups so grouped and see how they link to each other. Compare linkages between the same blogs over different topics.

Any takers?

Seven blog news trackers compared | Webware : Cool Web apps for everyone – CNET

CNET's overview of blog-tracking services for news:

In many ways, Wednesday's release of an updated front page to Google Blog Search has put blog news tracking into the limelight. Google didn't get there first though. Sites like Techmeme, Blogrunner, and Technorati have been tracking the hottest blog posts for quite some time. Now's a good point to take a look at what makes these sites (and others) individual and different from Google's new tool.

Read more here [link]