|
|
|
|
Mooga is a patent pending viral, self learning mobile entertainment ecosystem incorporating Artificial Intelligence techniques (powered by iKen Studio) to understand, track, predict & recommend mobile content based on individual user tastes, downloads & popular contents. Mooga actively promotes viral spreading of content, allowing users to recommend content to their friends & get paid for it! Content sales increase exponentially as Mooga acts like a friend to the users & recommends preferred content to them. Mooga adapts to their tastes, likes & dislikes continuously, resulting in a vastly improved end user experience.
(http://www.infinitemoco.com)
|
|
|
|
|
|
A flag-ship product of iKen
Solutions: is a completely web-based development environment to develop and deploy applications,
knowledge-based decision support systems, websites
and BI (Business
Intelligence)
applications backed by or enhanced with artificial intelligence (AI) techniques especially integrated
architectures of expert system and case-based reasoning (neural networks and genetic algorithms will be added soon). This software is available in SaaS mode at iKen Studio website. (http://www.ikenstudio.com)
Download: 
Brochure
Presentation
|
|
|
iKen Studio Features
- Access, management and configuration through Web
- No desktop installation and management
|
- Generate automatic Java scripts and web pages
- No explicit database programming required
- Various development interfaces
- Use of simple language for writing rules
- Support large number of operators, functions and data types
- Existing C/C++ APIs can be used
|
- Support popular databases: MS-SQL Server, MySQL, MS-Access, Excel,Text, etc.
- Simultaneously connects and accesses data from multiple databases. Data can be integrated or merged.
- In-built extraction, mapping and transformation engine
- Data access and manipulation through flexible external dynamic queries
|
- All components and interfaces use XML
- SQL-XML and XML-SQL transformation
- Access to APIs and intelligent systems through web services
|
- Powerful expert system engine supporting large number of data types including matrix, trend, XML etc. and various SQL, matrix, list, chart, session management, cursor management, report functions
- Use of scripting language for implementing procedural logic
- Powerful case-based reasoning (CBR) engine supporting structured and conversational CBR applications and all R4 cycles of CBR
- Applications can be developed using hybrids of AI techniques
|
- Role-based access to various development interfaces
- Role and user based access to applications, databases and data
- Encryption to prevent unauthorized changes
- System tracks changes made by the users and save change history for later investigation
|
|
Intelligence without building costly data warehouses
To survive in hyper competitive world, every organization whether it is small or big is looking to make sense of information generated and effective use of precious organizational knowledge available to do better business, customer service and risk management. Since existing operational information systems focus on automating business processes, these do not have capabilities of intelligence, knowledge automation and decision support. Obviously, organizations look for solutions to these problems in variety of other technologies such as data warehousing, data mining and knowledge management which comes at a cost and time. These may not be affordable by many. There are many organizations where volume of data is not that too large to built data warehouse and have data mining technology.
Agile, adaptive, on-the-fly and operational intelligence
Organizations are moving towards enterprise solutions which use common operational database, data in this operational database itself can be used for real-time, agile, decision support and operational intelligence in distributed fashion rather than building separate data warehouses and doing analysis on reactive basis.
Add intelligence in every business process for everyone
Most of the software vendors who provide operational information systems like ERP, CBS, operational-CRM and independent functional information systems such as human resource management, learning management systems, inventory management, billing systems concentrate on automating well defined business processes and lack intelligence and decision support capabilities. Client organizations of such systems have to look for BI, decision support and analytics systems separately, derive intelligence out of them and integrate into operational systems. Having built-in intelligence, decision support systems and knowledge automation in such software products can definitely value-add to products and thereby to their clients. Business intelligence, analytical and decision support systems at tactical and top level need not be looked separately from that of operational and reporting systems. These can be very much part and parcel of overall information systems to have intelligence at every level for every one.
Automated and configurable intelligence
Customizing, integrating and automating such intelligence and knowledge in operational business systems may be a challenge, which may involve extensive data analysis and modeling (e.g. using techniques of data mining), and integrating results into operational systems etc. Also, systems such as transactions processing in banking and telecom need automated and configured intelligence on ongoing basis to determine customer and transaction risk associated on dynamic and real-time basis instead of one-time or on periodic basis. Specific intelligence based on requirement can be built, configured and automated on top of operational database management systems.
Conventional Approach
Traditional approach to intelligence (e.g. BI) is to build data warehouse and use data mining and OLAP technologies. Building data warehouses is costly and time consuming. This technology is meant to assist in strategic decision making. The data stored has been historical. BI systems treated separately than operational and management reporting systems. The techniques used for analysis are different than systems that implement intelligence (such as operational systems). Also scope of BI is limited to data mining and OLAP tools, however, there are many business processes that need intelligence and knowledge components.
iKen's Approach to Intelligence
We advocate to use intelligence every where and for everyone possible. It can be agile, operational and pro-active. We define intelligence in more broader sense, and, our approach is to make it part and parcel of every software product that implement business processes whether it is simple inventory management software or enterprise solutions like ERP and CRM.
iKen Studio itself has distinct components core engines implementing AI technologies, user-interfaces, database access with built-in transformation, mapping and integration engine. Based on requirement in software product, iKen Studio components can be tailor-made, implemented and configured. For example, if an eCommerce software product vendor needs customizable intelligent Q & A engine for users based on dynamic decision rules, iKen Studio's rule-based expert system component can be used. Similarly, other eCommerce solution may need iKen's personalization and recommendation engine as a part of their overall product offering to their clients. This value-adds to existing products tremendously because clients of these software products and solutions do not have to look at BI (and buy the tools) separately. iKen Studio can either be used as embedded/built-in/add-on or it can be used as Intelligence Middleware.
|
A common framework is used for analysis, development, configuration and
deployment.
e.g. a personalization and recommendation can be modeled, deployed and integrated with eCommerce solution, it can be configured to
(A) Dynamically learn patterns using collaborative filtering every one
hour
(B) Calculate the next best products
and contents to be shown to the user whenever user makes a purchase/download, etc.
|
Mostly focus is on analysis (finding out patterns, relationships, trends etc.). Implementation of results in operational systems is
always an
issue.
After performing analysis, results (e.g. patterns) need to be implemented into operational
systems, so that operational systems follow them at execution time. e.g. embedding customer risk profile into credit card transaction system to manage the transaction risk.
|
|
Intelligence is derived
on-the-fly (lazy learning) and,
analysis is more
proactive.
|
Mostly reactive and off-line analysis
(It is done at a given
point of time and applied in future).
|
|
Machine intelligence
is supplemented by
human intelligence to add heuristics,
business rules. meta
information like
tags, conceptual and contextual knowledge.
|
Analytics
has limitations on
adding heuristics,
etc.,
may be difficult when techniques such as neural networks
are used.
|
|
Its focus
is on both macro as well as
micro analysis. The
macro focus
is suitable for
analysis of activity
group of people
and transactions,
their patterns, associating events,
etc. The micro
analysis is useful
for analyzing
behaviours and
patterns of individual
customer or user.
e.g. using micro-analysis it is possible to carry out following tasks.
(A) What things an
individual customer likes or
does, when, where, etc.
(e.g. a customer
X buys product
P1 and P2 when
she comes on weekends)
(B)
Customer
Y would likely respond to
particular promotional material Y
when send through email on a
Saturday.
|
It
has mainly
macro-focus, the
focus is on group of customers and large number of transactions, finds out broader patterns.
e.g. (A)
Customers of particular
profile do certain
kind of things (like customers between
age group 20-25 with
income group 10,000-25,000 like
X kind of products, customers between
age group 40-50 with
income group 25,000-50,000 are more
loyal and pay bills on
time.)
(B)
X group of customers should be targeted for
promotional material
Y during on
first week of month M because they are likely to respond.
|
|
Analytics
can be programmed for individual
users (customers)
or group of users: business logic, user
or group specific rules etc. e.g.
(A) a credit card
customer can set
her/his own generic rules for
'kind' of transaction
she/he would be doing and would like to match each one with
that
(B) based on
number and type of
downloads of an
user in specific
categories, logic
can be set to to
push categories
and contents
(C)
based on business
policy rules, the
products can be
prioritized when
shown to the user
|
Not possible through
only data mining techniques. Needs to be implemented in
operational
systems. |
|
|
|
|
