| |
| |
|
| Finding
the Unknown : Concept Recognition and Pattern Analysis |
|
| |
iQuest Analytics,
Inc. has taken the analysis of human behavior and applied those
principles to data. We leverage deep
knowledge of human behavior and the analysis of unstructured data,
applying our methods not just to people but to their ideas as well.
We surface these patterns by looking at
people, their words, their ideas and their relationships to understand
behavior and infer meaning.
iQuest Analytics software utilizes the iQuest
methodology. This unique methodology involves the clustering of themes and
concepts to show the most important threads and to surface patterns and
associations across the information network. This methodology also marries
social and grammatical network analysis, token extraction, text analysis
technologies and object mapping to derive meaning from unstructured data.
Our methodology allows for the creation of concept clusters, a radical and
practical approach to understanding complex ideas as expressed in unstructured data.
iQuest can rapidly analyze anything with words or
symbols – articles, web pages, reports, memos, e-mails, telephone call logs,
transcripts, message boards, blogs, survey responses, RSS feeds and more –
to provide intelligent insights in context.
iQuest discovers predictive patterns by looking at people, their words,
their ideas and their relationships to infer behavior. iQuest creates
value by giving users the ability to draw on those inferences in developing
successful responses to a challenging and difficult digital world.
The Social Life of Words
Social Network Analysis (SNA) is growing in importance. Most of the time,
however, SNA is used in its conventional form where people are the nodes or
agents in the network graph and relationships (established, for example, by
communication) are the links. There are many SNA analytics that can be applied
to these graphs such as degree of centrality, degree of betweeness, etc. These
analyses form a very powerful set of tools for observing the nature and time
evolution of social networks.
We use SNA in a radically different way. Instead of making people the central agents
in the structure, iQuest makes the documents and terms the central agents in the network
structure. SNA can just as easily be applied to these structures, giving clear pictures
of the “social life of words.”
When SNA is applied to document and term data, features such as degree of centrality
and degree of betweeness reveal very important information about key documents in
the set. For example, documents that have a high degree of betweeness are frequently
central to the flow of ideas within the document set. These key indicators provide the
user with critical insight into which documents, blogs, message boards, reports,
news feeds, etc. to scrutinize more carefully or which documents represent good
starting points for further analysis.
|
|
|
|
|
|
| |
|
|
|
|
|
