New techniques in content management

06-Aug-08

 

Access all areas: Top tools for content management

content management
Companies can get so overwhelmed with surface information, it becomes harder to single out what is going to prove most valuable. Louise Druce takes a look at some of the information-access tools emerging on the market that are helping businesses drill deeper.

 

 

If you want to be truly competitive in business you need to have the best and most up-to-date information at your fingertips to get inside your market and the customer psyche, which is where content management can become your goldmine.

Fact extraction

While a search engine can return thousands of pages of results based on the key words you input, it does not understand the human thought process of relevance or connotations behind the text. For example, if you search for 'plumber' in Google, you then have the option of searching for a career in plumbing, finding a professional in your local area, or contacting the institution governing the trade, among other things.

There are ways to be more specific in your search but it can be very time consuming (and frustrating!).  Fact extraction does what it says on the tin by allowing you to pre-define a particular 'fact', such as an event, activity or relationship, together with a set of documents where it might be found. Retail sites such as Amazon.com and Tesco.com use it to recommend other products a customer might like to buy based on the nature of their original purchase, for example.

You can also use it to track a person by specific interest, rather than trawling through every individual document where they might appear. Or you can track information on competitors through external documents such as product announcements and press releases, as well as internal sources such as analyst reports, to be pooled into one electronic report.

Web harvesting

The next step along sounds like a virtual day down the farm but web harvesting (also known as web scraping) collects data for analysis by 'crawling' through URLs and cherry-picking the ones that match pre-defined content criteria. For example, if you take the old and cumbersome approach, search engines such as Google and Yahoo will index everything related to the search term you specify. You would then need to read through pages and pages of results, weed out the information you need and then manually cut and paste everything into a Word document for analysis.

A web harvesting search engine will only index the URLs it is directed to, such as chat rooms, news feeds or blogs. Extraction tools then automate the reading, copying and pasting needed to get to the analysis stage. The reduced list size means a faster searchable application that provides more selective and higher quality results because the URLs are already filtered according to the topic of interest.

The competitive advantage web harvesting has is that it can be used to collect information on your direct rivals, such as pricing information, product descriptions, financial data and future direction. You can even base the search on a particular person. The information can also be continually searched so you receive all the latest updates.

Social bookmarking

In much the same way as you'd save a website you frequently revisit to your favourites list, social bookmarking is a more public way of sharing useful sites, blogs, articles, documents etc. Digg, del.icio.us and StumbleUpon are among a growing number of websites that allow you to create your own account and invite selected people to access the links you've been looking at. The advantage is you can share and swap links with people who have similar interests and create a virtual library of information that could be used for things like product development.

However, a lot of people have cottoned on to the value of using these sites for commercial purposes and self-promotion. Competitors might also try to take a sneaky peek at the sorts of things your customers are interested in or what projects you might be working on, so exercise some caution. You can read some personal insights into how social bookmarking works here: http://www.ukbusinessforums.co.uk/forums/showthread.php?t=66859

Social tagging

Once you have bookmarked content, social tagging allows you to index it using searchable words or phrases that make it easier to retrieve the information at a later date. This has an advantage over traditional folders as many words can be used to tag the content in keeping with the company brand or jargon – for example, 'Tom Jones', 'competitor', 'marketing', 'product-launch', 'priority follow-up' – rather than an ambiguous catch-all such as 'marketing'.

You can also view bookmarks associated with a chosen tag and include how many users have bookmarked them. More sophisticated sites create clusters of tags based on their relationships.

Tagging can help improve sales and marketing alignment and also means the company retains a history of knowledge, even when a person leaves the company. However, because tagging is a human process, there is the danger of duplication, miss-spellings, or misunderstandings about what the content is if people try to use slang or words that only mean something to them.

Also, websites which are constantly bookmarked and tagged using popular search terms can be found more easily and replicated by spammers. Adjusting security settings can help. Some social booking marking sites have also addressed this by adding visual passwords or CAPTCHAs.

 

Details

Author:
Kate Phelon
Publisher:
KnowledgeBoard
Date:
06-Aug-08
Sections:
Home , News

This article has been read 994 times.

Member comments (3)

Share your views with other users: add your own comments to this item.

Alakh Asthana
Alakh Asthana, 26-Aug-08 @ 10:56AM
Text Mining

SaS is another interesting and reliable application for text mining.

Anonymous
Anonymous, 25-Aug-08 @ 05:28AM
Top tools for content management

This issue seems to be growing all the time as even indexing web pages is now a huge issue. The grey literature for various sectors is almost impossible to keep track of and when you do capture large volumes of textual data what do you do with it all? Reading is all very well but once you have more than several hundred documents this can become unwieldy without further tools and analytical structures.The whole issue of sense-making and analysis needs to be part of this KM discussion. Perhaps the field of content analysis has something to offer here including some of the software available out there that goes into the computational linguistics field. I suspect that the data mining sub-field of text mining could contribute eg. products like Leximancer, LIWC and T-Lab that come out of social science research perspectives. These sorts of tools can help speed up analysis processes and maintain some consistency and objectivity in analysing electronic textual data.

Alakh Asthana
Alakh Asthana, 21-Aug-08 @ 06:49AM
Intranet Content Extraction
The first thing that comes to my mind with the word "Content" is our corporate Intranet.

Most of the time Intranet web parts / pages become nothing but information dump. There is a virtual team that is stationed in some corner of the cubicle and keeps posting "organizational revelant" content for people to read. The only way to measure how good or bad the article may be a survey tool at the end of the article / posting. Something like "Rate This Article".

Within these articles and the readers lie SIGs, and identifying these SIGs to cater to, meets the objective of penentrative Knowledge Management.

Best way what we have identfied is to have discussion boards at the end of the article -> identify people with similar thoughts and interest -> gather these readers and facilitate CoPs within them.

The more people centric the content management is, the higher value it would generate for any organzation.

Regards,
Alakh Asthana
eClerx Services Ltd