|
There are two main types of streaming analytical software products, the
most prevalent are the deductive streaming group which operate by user-define business rules and are used to monitor massive
and multiple streams of data – reacting in real-time to consumer events as they take place. Such
as when an application form is filled at an enterprise’s website or when an email is received – in fact whenever
a ‘consumer event’ occurs – triggering an enterprise reaction.
These software products offer graphical interfaces for easy construction of
behavioral analytics filters in the form of business rules in the processing of consumer data streams for real-time monitoring
and triggering of specific offers or content. These are some of the deductive real-time streaming analytical
software firms which work with user-defined business rules:
Streambase.com provide software for rapidly building systems that analyze and act on real-time streaming data at a fraction of the cost
of building a data warehouse. Streambase uses what it calls Complex Event Processing technology
for low-latency filtering of consumer event data. The academic origins of CEP
began with research efforts at Cal Tech and Stanford University in the mid-1990s, with a focus on processing streaming
data by identifying complex sequences of events within specified time interval, and then triggering an appropriate action
or alert as a result of these real-time behaviors. The processing of messages for real-time analysis also involves the access to “clouds”
of data sources to enable the creation and maintenance of “behavioral analytical filters” for enterprises. With
the increasing speed and amount of behavioral data being generated 24/7 these type of software can be leveraged by enterprises
to improve their relevance and loyalty with consumers. Progress.com streaming behavioral analytical software
monitors customer events, identifies patterns and takes action instantly. Using Progress an enterprise
can achieve operational awareness with real-time, event-driven business activity monitoring (BAM). Progress,
as other of these streaming behavioral analytical software providers argue that by detecting business-relevant patterns in
real-time, they can helps enterprises quickly respond to business opportunities and identify threats wherever they may occur
and while there’s time to act. This
type of event driven software is being used across a wide range of applications in financial markets, manufacturing, retail
operations, and fraud detection and prevention. Yet another analytical software provider is Aleri.com their Streaming
Platform supports complex event processing, allowing enterprises to react instantly to changing conditions by analyzing incoming data in real-time. The software allows companies to rapidly build and deploy new
applications that derive insight from streaming event data, letting them instantly respond to changing conditions. Aleri
believes firms need to be event-driven and that delayed response to changing conditions can mean missed opportunities and
overlooked threats. Banks.com is using the Aleri CEP platform to gain real-time insight on customer activity and traffic trends to improve
everyday business operations and marketing campaigns in order to enhance customer relationship management and increase revenues.
Prior to using Aleri, Banks.com was making decisions on historical data, now however they are using Aleri to manage
their marketing and advertising programs in real-time, immediately responding to new trends and traffic patterns. The streaming
software allows Bank.com to monitor user behavior, traffic, and trends in real-time – to provide a more targeted approach
to their marketing and advertising programs – and to provide a more unique and relevant experience to their customers. There are also inductive streaming analytical software in which rules
are constructed from multiple analyzes performed horizontally across structured databases and unstructured documents, clickstreams
and emails. The inductive streamer products can be used for behavioral analytics by enterprises by first
analyzing streams of data from call sites, websites, email and transactional legacy systems in order to develop global models.
Once constructed those behavior analytical filters – models and business rules – can be deployed at websites,
call sites and other operational systems to react to consumer events as they occur enterprise-wide for real-time reactions
to consumer events as they take place.
At
the core of the inductive streaming analytical software products such as InferX.com are advanced algorithms which can perform behavioral analytics via segmentation, classification, clustering, concept extraction,
visualization, and prediction directly from consumer data streams. The inductive streaming analytical software
can be used to analyze multiple databases to determine the lifetime value of customers from cradle to grave the instant they
interact with an enterprise, whether at its website, call site or email.
Both streamers products support the processes of analyzing an assortment
of behaviors, demographics, lifestyle, transactional, Internet and geospatial information for the personalization of offers
to consumers at the various touch points irregardless of the channel. In behavioral analytics there is no latency between
analysis and action – instead it is one single integrated continuous process.
|
