Archive

Archive for November 15, 2010

Strategic Knowledge Engineering

November 15, 2010 Leave a comment

The nature of work has evolved towards service and knowledge related contexts; in this scenario we need intense focus over the processes and frameworks that would define on how we harness and leverage information that is generated over all processes for driving more value. Organizations seeking to extend themselves into this critical area of analytics needs to focus on shaping the process and the people; the vital components. Organizations should be aware of the characteristics of its relative business knowledge and its sources, features and usability. It needs to shape methods that can collate data from extensive source area, link disparate data sources for a collective sense. There needs to be a governing process that evaluates, merges, researches, develops,  the data and comes out with options for business optimization.

Splitting the stages into functional units tied down to specific goals.

1. Knowledge Sourcing
2. Knowledge Abstraction
3. Knowledge Framing
4. Knowledge Warehousing
5. Knowledge Engineering

Knowledge Sourcing can be described as the identifying and acquiring historic and real-time data that are available from varied sources in the annals of any given business process. Data usually is available in the databases, files, logs, documents and they hold information transactions. The data is accumulated over time and the stores swell with size; all the while presenting an opportunity to present intelligent or informative analysis that could drive a stack of benefits. Other than the normal reporting structures built to mark the progress over Production, Quality and all in between; specific insight is never sought. The inference – insight that can be extracted remains an opportunity and for long. This data stacks and repositories are analyzed and identified as channels / sources; this source is the feeder for the analytics and the outcomes, process of source identification and data acquisition needs to be clinical.

Knowledge Abstraction helps in framing the insights and is completely skill dependent; this human skill is at an expert level on the domain of choice [SME  – Subject Matter Expertise] and the process relies heavily on the understanding and knowledge of the people resources. The data set is categorized based on the business case or the problem statement, data and information framework built here are weighed and categorized in order to support the reasoning and outcomes. The frame is built over an objective where the entire pursuit of intelligence is architected. All information here is bridged, connectors, the domino effect and factoring are all part of this abstraction process. Abstraction has two distinct process loops; one

Knowledge Framing ensures that abstracted data is further developed and refined through higher process routines to achieve anchoring over statistical data. The anchoring is vital as this builds the entire exercise over reasoning, analytics and recommendations on numerical realities. There effort here is predominantly built over mining [drilling down] data blocks to identify patterns, strings, values  to build neural relation that will help garner deep insight into all the facets.

Knowledge Warehousing comes in as the vital next step where the structured information and knowledge is stored into prescribed data structures that acts as the foundation for all the processed data. These individual ‘marts’ contain data stacks that are structured  over certain perspectives. Here the data stacks are linked, merged, to evolve and position the data for all analytics and intelligence extracts. This warehousing of the structured knowledge is done using enterprise warehousing applications, there is also a process layer to help drive Data reporting based on rule engines and business case.

Knowledge Engineering Analytics is the function where we have all the analytics process and frameworks deployed to extract intelligence out of the data warehouse. There is a lot of factor building, dependency tracking, correlations exercises that are done over analytics suites that help understanding all the different perspectives from an analytics standpoint. Causes, factors, symptoms, diagnosis, recommendations, solutioning are all the indispensable next steps that add a tremendous value to business and business operations.

Jesu Valiant

Service & Analytics recommendations:
To initiate a SKE – Strategic Knowledge Engineering for any of your data stacks, there is a comprehensive solution stack shaped over a decade of analytics at the CSS Knowledge Engineering & Research Labs.

Email: enterprise.ke@csscorp.com

%d bloggers like this: