Google Reader, a flexible RSS reader beloved by its users, recently came to an end. Everyone from casual consumers to business people made use of it. For business users who were pushing Google Reader’s limits, Newsdesk may be a great solution.
While Google Reader allowed you to view chosen RSS feeds, Newsdesk gives access to millions of articles and posts across News sources, private content licenses, blogs, forums, video, and other social media sites. Being search-driven, Newsdesk gives you the option of following individual sources, searching for keywords across all sources, or exploring content using our categories.
Newsdesk takes information discovery to the next level and puts power directly into the hands of corporate communications and information professionals. Beyond discovery, Newsdesk is designed to help deliver targeted business news across the organization. over intranet, email and handhelds.
Leave a Comment July 2, 2013
With the recent revelations of widespread surveillance by the NSA in the United States, the concept of metadata has been in the news lately. One of the controversial aspects is their collection of metadata about phone users and how revealing that can be without the content of the calls themselves. The answer, it turns out, is a lot.
The what and why of metadata
Literally “data about data,” Metadata is the descriptive information about content. They let you search for and pinpoint precise content that matters for market and competitive intelligence. It also allows broad analysis of the data as a whole.
The trouble with large amounts of data is well known. How do you draw conclusions from large amounts of data? How do you narrow it down to only what’s relevant to the task at hand?
Metadata is revealing
Metadata can give more valuable information than aggregated content by illuminating trends within that content. For a broad topic such as Microsoft, trend discovery without metadata is difficult. There’s just too much content to read. Metadata makes the process much more manageable.
For example, by doing a simple search for “Microsoft” in Newsdesk, I see that Don Mattrick and Xbox are mentioned often. This reflects the recent controversy over Microsoft’s Xbox One. We can break down what percentage of stories about Microsoft mention Xbox to get a different perspective:
We can also show that nearly all news coverage of Don Mattrick, President of Microsoft’s Interactive Entertainment Business, also mentions Xbox.
How it works
At Moreover, we are experts in metadata. We collect information about the publishing sites themselves, including topic, language, and location. For individual articles, we collect data about the author, time of publication, headline. Many metadata can be added using the article text, such as companies, people, and topics mentioned.
Each of the millions of postings that Moreover monitors gets over 50 pieces of metadata attached to them. This facilitates the generation of flexible and detailed reports of the trends across the data. In searches, robust metadata helps you zero in on exactly the news and social media you need.
The power of metadata
While the specifics differ on the metadata the NSA collects on cell phones and Moreover collects on internet articles, the underlying concepts are the same. This “data about the data” is key to gaining insight about lots of content.
Utilizing metadata effectively puts you at an advantage in analyzing, searching, and managing data. Want to learn more about how you can leverage valuable metadata for your company? Let’s talk.
Leave a Comment June 26, 2013
Newsdesk Newsletters can now show the companies and people mentioned in the articles. This enhancement was released on Tuesday and is available to clients with access to Newsletters.
Click to Enlarge
By giving newsletter recipients more information at a glance, you increase the value of each article. Readers can get a better understanding of what they’re about to read. This can be invaluable for Newsletters containing many articles.
To enable Companies and/or People to Newsletters, open the Newsletter in Newsdesk and click “Edit Template.” Each can be enabled separately by ticking its box and clicking “Save.”
Click to Enlarge
We are planning to add this option to the main search results view in the future and welcome your feedback on this new feature.
Leave a Comment June 14, 2013
When creating a new search, it is important first to understand what kind of data you are trying to find. Newsdesk provides many specialized ways of getting at data, but not all will be appropriate for every type of search. Not knowing what you want ahead of time can lead to unnecessary frustration with the process.
When getting noise (i.e., irrelevant results) in the search results just isn’t an option, you’ll want to use these strategies to create a tight search. This is useful when the search results are being fed directly into email communications or are integrated into an intranet or website. The top priority is relevance and “clean” results. It’s a trade off, however, as filtering the irrelevant results will block some good articles as well.
Getting tight results
The general philosophy of creating tight results is to be increasingly restrictive in adding search terms. This means using exact phrases, Boolean ANDs and multiple filters to winnow out anything that is not a guaranteed match.
Exact phrases are multiple words separated by spaces, contained within double quotation marks. For example, here is a simple search using 2 exact phrases:
“Moreover Technologies” AND “media monitoring”
Any news article or blog post that contains both of these phrases will almost certainly be about Moreover Technologies. Using “Moreover Technologies Inc.” would be even more precise, but we’d likely lose even more relevant results due to it being less common. This is an example of the trade off mentioned above.
Headline and START
Another way of ensuring that all search results are going to be relevant to your aim is for the keywords to appear at the beginning of an article. In Newsdesk, this is accomplished by using the Emphasis tab under the Advanced search options:
To specify that keywords must appear in the headline of an article, add them to the Headline Include field. To specify position in the article text, use the Position filter, which tells Newsdesk that the keywords in the main search box must appear within the first X number of words.
Additionally, you can use the START search parameter, detailed here.
On the other side of the spectrum from Tight Searches are Broad Searches. These are most appropriate for internal communications, research, or in any case where the opportunity cost of missing a relevant result is greater than receiving a bad match.
Creating Broad Searches
The idea is to run a query that is specific enough to be relevant, but loose enough to catch all the articles you want.
In Newsdesk, a great place to start is the Topics filter. Grouped into broad channels of like topics, there are many pre-built categories available to you. If the topic has to do with agriculture, one useful topic may be “Agriculture News.” This can be used to create a broad search and keywords or other filters may be applied to narrow it down in a particular direction.
The below search takes at least one word from each group. This way, it’s flexible, but also will contain at least 3 specific keywords. The aim of the following query is to find information about the price of agricultural commodities:
(corn wheat alfalfa barley) AND (commodit* ETF bale* bushel*) AND (price prices value)
There is no wrong way to create a search, but it saves time and frustration to think about which strategies may be best before beginning. Tight Searches and Broad Searches are at 2 ends of the spectrum. It’s likely that your searches will fall somewhere in the middle.
What other search strategies do you use?
Leave a Comment June 7, 2013
Continuing our look at advanced searching in Newsdesk, this week we’ll examine the START operator.
START allows you to specify that keywords should appear in the first X number of words in any articles returned in your search results. Simple searches look for keywords that appear anywhere in an article. The drawback to this is that a passing mention at the end of an article will still match, but it’s less likely to be relevant to your search results.
Keywords used in the introductory paragraphs are likely to be the subject of the post. START gives you the option of distinguishing between items that are simple mentions and those that are the focus of an article.
Here is a simple search, designed to find articles about Google (i.e., where Google is mentioned within the first 10 words of the article) and that mention smartphones anywhere in the article.
START/10:(Google) AND (smartphone OR “smart phone”)
We can do some simple Boolean logic with START as well. This query will find articles where Tumblr and Yahoo! are both mentioned in the first 10 words.
START/10:(Tumblr AND Yahoo)
This next search is looking for items about the major search engines.
START/10:(Google OR Bing OR Yahoo)
This search will return articles mentioning the search engines, but does not include Tumblr in the first 15 words. The idea here is that Tumblr should not be the focus of an article, though it may be mentioned.
START/15:(Google OR Bing OR Yahoo NOT Tumblr)
Another option for position-based keyword searching is to use the Position filter under the Emphasis tab in the Advanced search options:
Position doesn’t provide as much granularity as START, but it’s quick and easy, only requiring the tick of a radio button.
Test it out and feel free to contact us if you have any questions.
Leave a Comment May 30, 2013
Yesterday we unveiled a Newsdesk enhancement that is the culmination of months of planning, development efforts, and collaboration with clients: Article Tagging.
- Adding sentiment scores to articles and posts
- Applying your own editorial classification to content
- Setting up tag-based searches
- Organizing editorial workflows
Article Tagging allows you to add a layer of qualitative analysis to the data, giving professionals in your organization deeper insight into the articles they receive.
Manual or Automated
Tagging specific articles one at a time after manually reviewing them ensures that each one is relevant. You can also add tags to articles automatically by setting tags at the feed level. This can be used to complement a manual tagging effort, where a search result is sufficiently reliable that you always wish to add selected article tags to the articles going into that feed.
Click to Enlarge
Search Using Tags
You can drill into search results by combining keywords and article tags in the same search. Combine searches by using two searches for the same automated tags. Search an entire tag hierarchy, a specific tag, or in any combination.
Click to Enlarge
Clients that purchase the optional Article Tagging feature will have another tool for managing content in Newsdesk. Tagging articles manually or automatically based on search strings lets you create searches that can feed to your intranet, website, or Newsdesk emails.
Contact us to get started today.
Leave a Comment May 22, 2013
To help you get up-to-the-minute information faster and easier using Newsdesk, we’ve put together a few quick tips for you:
Do you frequently make the same changes to your searches? If so, setting up a search template can save you time. Then, when you search, it’ll use your template already customized to your search preferences. Here’s how:
- From the “My Content” drop-down menu, select “Edit search template”
- Add keywords, set media types, languages, etc. to customize the template for the types of searches you usually perform
- Click “Save”
Want to add your own articles to your search? You can manually insert articles and company information such as a link to PDFs and announcements to send out to your website or intranet using an RSS feed via the Export tab or in a Newsletter from the Email tab. It’s easy:
- Open the drop-down menu for the search you would like to add an article to
- Click “Add Article”
- Enter the headline, body, source name, etc.
- Refresh your search and see your newly added article. It will automatically appear in any alerts or exported feeds containing that search.
Need to do further analysis? You can download data into Excel from Newsdesk. Here’s how:
- Open the drop-down menu for the search you would like to download the information from
- Click “Analyze Feed”
- Click “View Analysis”
- From the “Download” drop-down menu, select “Feed Statistics” or “Article Data”
Leave a Comment May 17, 2013
A topic that has been growing in popularity is “crypto-currencies” with the most common of these being Bitcoin. Bitcoins are a digital currency that are bought on an exchange and then used to anonymously buy and sell goods online. If you haven’t heard of them, check out this video.
Bitcoin’s pricing has proven to be very volatile. For example, on April 10th the exchange rate of one Bitcoin dropped from $230 to $165. With a 28% plunge in just one day, wouldn’t it be useful to see if there are any indicators of whether Bitcoin will rise or fall over the next days? To get some insight, we pulled data from our media monitoring product, Newsdesk, into Excel and matched it up against historical data for Bitcoin’s exchange rates.
Unlike gold or a publically traded stock, Bitcoin’s pricing is not tied to any physical assets so it is vulnerable to fluctuations not normally seen in stocks.
Digging deeper with Newsdesk, we wanted to see how much the media’s hype about Bitcoins seemed to impact its price. As the graph below shows, trading volume seemed to lag after media mentions. This makes sense since a lot of the people using Bitcoins aren’t necessarily professional traders, so there is a disconnect between the information available and the market’s pricing; often a full day lag from media mentions to a spike in volume.
As shown here, simply using media mentions to determine the increase or decrease in Bitcoin’s exchange rate is an imperfect model.
As we move forward in an increasingly digital world, it will be interesting to see what role Bitcoins continue to play; whether they will be considered an investment or simply someplace to flee when traditional currencies have problems. Until then, if you’re a Newsdesk user and want to follow Bitcoins, or to delve deeper into your company’s online mentions, here is a simple way to download an analytics feed:
- Click the Search Tab
- Click New Search
- Select News only
- Search term: bitcoin*(* is a wildcard that searches for bitcoin, bitcoins, etc)
- Default search settings: English only, 60 days, all languages
- Click Analytics Tab
- Line graph, 60 Days by Day
- Drag Bitcoin feed over, Click “View Analytics”
- Download -> Feed Statistics (Excel)
For the examples in this post, we copied the Bitcoin trading data from here and put it together with the exported Newsdesk coverage in Excel. Excel allows us to easily dig through both sets of data to identify any hidden trends we wouldn’t see otherwise.
Can you think of any topics you’d like us to mine for in Newsdesk?
Leave a Comment May 14, 2013
Recently, we brought the NewsRight brand into the Moreover Technologies fold, reinforcing our commitment to providing seamless access to the high quality content our customers rely on.
Moreover Technologies has been closely aligned with NewsRight to meet the demand for trusted content from news publishers. With NewsRight joining the Moreover family and in conjunction with partner BurrellesLuce, there is now a single, consolidated go-to source of licensed content available.
Moving forward with one catalog will make online content licensing, tracking and distribution far easier for our clients. We are pleased by the vote of confidence from the 29 publishers/owners in the NewsRight catalog.
Our value and service grows with these changes:
The NewsRight name and logo will be rolled into Moreover Technologies
The Moreover / BurrellesLuce Metabase Premium service, featuring full access to premium licensed content, will be renamed NewsRight
We look forward to continuing to serve our clients in the years to come with a focus on compliance with publishers. Please contact Client Services if you have any questions.
Leave a Comment May 8, 2013
What is it?
Moreover recently introduced a new search operator that functions similar to proximity searches (“keyword1 keyword2″~5). You can specify 2 groups of keywords that must appear near each other. You can also set how many words apart they should be.
(CAT caterpillar) NEAR/10 (tractor construction)
This looks for the words “CAT” or “caterpillar” within 10 words of “tractor” or “construction.”
Why use it?
NEAR is another tool in a user’s toolbox for creating a focused and relevant search. If the combo of Boolean logic and Newsdesk’s filters aren’t providing enough granular control over search results, NEAR may be the answer.
NEAR fills the gap between exact phrases, which may be too strict, and simply using the AND operator, which may be too broad. It allows more flexibility.
(gold silver copper palladium platinum) NEAR/5 (ETF* “exchange traded fund” stock*)
Using NEAR here saves you the time of writing out “gold ETF,” “silver ETF,” and so on. It also catches cases where an article phrases them in a different order, e.g. “ETFs for gold and silver.”
The NEAR operator can be used within a longer search. Here I’m looking for stories mentioning Les Wexner in conjunction with specific electric car companies.
(“Les Wexner” “Leslie Wexner” “limited brands”) AND ((fisker coda) AND ((car* vehicle*) NEAR/10 (electric green “plug-in” plugin))) NOT (“wexner center” osu “ohio state”)
Go try it out in your own Newsdesk account.
Leave a Comment May 2, 2013