|
|
|
|
Resource Library
| Tours & Walkthroughs |
 |
Walk through Grokker’s
features.
View it |
|
See how Grokker
works in the enterprise.
View it |
|
See
how Grokker delivers premium content and adds value to
it.
View it |
| White
Papers & Data Sheets |
 |
Learn how to enhance and manage
search information and resources across the enterprise.
View as: PDF |
|
Grokker Features Overview – Learn about Grokker’s key features.
View as: PDF |
|
Learn how Grokker can extend the capabilities and maximizes the value of your Google Search Appliance or FAST deployment.
GSA Datasheet: PDF
FAST Datasheet: PDF |
|
White Paper
on Grokker's clustering engine.
View as: PDF |
| Client Case Studies |
|
Learn how Sun Microsystems® uses
Grokker to facilitate cutting-edge research.
View as: PDF |
|
Learn how Stanford
University leverages the power of Grokker to deliver best in class information resources to the entire campus community.
View as: PDF |
|
Frequently Asked Questions
How does Grokker make data more useful?
If you tried to page through 25 pages of listed results, you would probably
not find much useful information at all. Every feature in the Grokker application
is designed to add to the usefulness of large data sets.
- Federated search gives users the ability to simultaneously query disparate content sources.
- Clustering gives users an idea of the categories and relationships among
data sources.
- Visualization allows users to see many results in a single view.
- Filtering helps users pinpoint specific data.
- The working list makes search results persistent.
- Collaborative tools such as email, export, posting, etc. allow users to
share new knowledge.
 Top
How much training does Grokker require?
The training required for GrokkerESM is minimal. The UI emulates popular web
search engines so that users can get something out of the application the first
time they use it. Once users get their first set of results, they can then
explore additional features and function.
Groxis produces a standard tour that gives users a step-by-step overview
of the application, as well as a help section. Both can be accessed from the
application at any time.
Top
What are Grokker’s reporting capabilities?
Grokker comes with a standard suite of daily, weekly and monthly reports that
allow the customer administrator to track access, session time, application
access (individual sessions), total number of queries, access by data source,
number of results sets sent via emailed and exported results sets, query term,
session times, and query source (portal, search box location, application start
page, etc.). We can also create customized reports to client specifications.
Top
Does Grokker work with proxy servers?
Yes. We have more than five years experience working with a wide variety
of proxy scenarios, so we feel as though we’ve got most of them covered.
Top
How does Grokker cluster (categorize) results?
Grokker’s clustering engine combines common phrase discovery (keywords,
for example) with latent semantic indexing (the intent of the phrase) to extract
key concepts from returned results. The algorithm compares both the common
elements of returned data and the strengths of the relationships among the
data from each source and then graphs them according to the rules of the algorithm.
Please contact our sales department for a more detailed white paper on Grokker’s
clustering engine.
Top
Do results appear in more than one category? Why?
Yes. Grokker places documents and web site into as many categories as are
appropriate (with certain limits) to the content contained in the document
or site. This is especially useful in the process of discovery, because
clearly end users can interpret a broad subject in any number of different
ways. By having multiple paths to that same document, it increases the
chance for discovery.
Top
Can Grokker cluster very different types of data at once?
Grokker analyzes a wide range of metadata depending upon the source of the
result. For instance, Grokker analyzes snippets and titles for most web results,
and abstracts for many premium databases. Grokker can look to any portion of
the document, including the full text, as it creates categories and drops documents
into them.
Top
How does Grokker rank documents?
The goal of clustering and visualization is to add to the utility of large
data sets so that users don’t have to rely upon ranking. This is especially
important when federating data sources because ranking methodology is different
for each source. That said, the list display in the right pane is always ordered
by relative rank.
Top
Why does Grokker take longer than a regular search engine?
Grokker will take a few seconds longer than a search engine because Grokker
collects all the data before beginning the analysis and clustering process.
The amount of time it takes for can also vary depending upon the number of
sources the user selects and the number of desired results.
Generally, results from across several sources are retrieved, organized, and
presented within about 20 seconds. While this is a long time compared with
a standard search engine, it’s actually much faster than accessing several
data sources separately and then merging the results. If a particular data
source does not return results within a given window of time, Grokker will
present only the results from those sources that delivered.
Top
What premium content does Grokker access?
Grokker can access content from any electronic content provider and already
has connectors and partnerships in place for many of the industry’s leaders.
In cases where we don’t have a connector already, we work with our customers
to forge strong partnerships with their content providers.
Top
Can Grokker access and federate an enterprise’s proprietary
databases?
Yes. We approach customized connectors in a standard fashion – analysis,
design, implementation, testing, and deployment. The first step is always a thorough
analysis so that we understand what kind of data is in your database or content
store, what kind of meta data is associated with it, and how you would like
to use it. Our informatics engineers work together with our clients to ensure
that custom connectors, which take about three weeks to create, test and deploy,
will maximize the value of the data.
Top
Does Grokker integrate with any kind of search engine?
Grokker relies on an open API interface so it can integrate with almost any
search engine or enterprise search appliance.
Top
How does credentialing work for subscription or private data sources?
With both internal data and premium content, we pass through credentials according
to the source’s requirements. The credentials are passed to the source
along with the query. The sources pass back only partial results (metadata,
abstracts, etc.) into Grokker. Links back to full text documents are credentialed
further according to the source’s requirements.
Top
Is Grokker fully hosted or can it be deployed inside of an enterprise’s
firewall?
Grokker can be deployed as a fully-hosted application under a subscription
model. With this model, the deployment time and initial expenditure are minimal,
so it’s ideal for mid-sized enterprises as well as for start-ups that
need to remain agile.
Grokker can also be deployed as a thin client infrastructure solution within
the enterprise. It runs in any backend environment, including Linux. While
it requires more infrastructure support and longer leads for updates and modifications,
this scenario works well for enterprises that are very security or control
conscious.
Top
How long does it take to deploy Grokker?
With the fully hosted solution, Grokker can be up and running with three days.
Internal deployments take about a week to ten days.
Top
How much does Grokker cost?
Please contact our sales team for more specifics about pricing.
Top |
Glossary of Terms
| Atom |
A type of web feed that provides web content or summaries
of web content together with links to the full versions of the content,
and other metadata. Grokker allows users to create Atom feeds of results
or the contents of the working list. |
| Clustering |
Adding value to search results by identifying common characteristics
among the results and then grouping them according to those common characteristics. |
| Connector |
A computer program that interacts with another program. In Grokker, most
information sources are plugins, meaning that the program that connects
to the source is different from the program that does most of the work.
In this way, Grokker can support any configuration of information sources.
Sometimes referred to as a plugin. |
| Content |
Information and documented experiences created by individuals, enterprises
and technologies to benefit audiences in venues that they value. |
| Content Publisher |
A company that aggregates and publishes premium content from a wide variety
of sources. EBSCO is a key content publishing partner to Groxis. |
| De.licio.us |
A popular website that allows users to post, share and tag (write about)
their favorite websites. Grokker allows users to post bookmarks to del.icio.us
or to nearly any other bookmark hosting site. |
| Enterprise Search Applications |
Many companies provide different ways for enterprises to index and search
their databases, internal content, and web content. Some examples are Grokker
partners such as FAST Search & Transfer and the Google Search Appliance.
Grokker supports connectors to most enterprise search applications. |
| Federated Search |
Access to multiple and disparate data sources with a single query string
located within a single interface. |
| Grok |
A Martian word found in Robert A. Heinlein's 1961 sci-fi classic "A
Stranger in a Strange Land," meaning literally ‘to drink’ and
metaphorically ‘to be one with.’ To grok something is to understand
something so well that it is fully absorbed into oneself. It is to look
at every problem, opportunity, action, and point of view from any and all
perspectives. This is the spirit behind Grokker. |
| Information Assets |
The body of knowledge available to the enterprise that advances the enterprise's
ability to innovate. |
| Information Visualization |
The use of interactive, visual representations of abstract data to aid
understanding and discovery. |
| Java |
Java is a programming language that is well suited to web-based applications.
Grokker's Map View is a java applet. You can use all of Grokker's other
features without Java. If you need to download or upgrade your browser's
java capabilities visit the Sun java site.
|
| Metadata |
Data that describes other data. With search results, the descriptions
are usually metadata. |
| Plugin |
A computer program that interacts with another program. In Grokker, most
information sources are plugins, meaning that the program that connects
to the source is different from the program that does most of the work.
In this way, Grokker can support any configuration of information sources.
Sometimes referred to as a connector. |
| RSS |
RSS is an abbreviation for Really Simple Syndication. RSS is a type of
web feed that provides web content or summaries of web content together
with links to the full versions of the content, and other metadata. Grokker
allows users to create RSS feeds of results or the contents of the working
list. |
| Software as a Service |
A model of software delivery where a company hosts its customers' application,
diminishing the customer's need to deploy and maintain that software. |
| Thin Client |
Software in which much of the processing is done on a host server rather
than on the user's computer. |
|
|
|
| |
|
|
|
|
Resources & FAQs
