Structuring Industry Knowledge

Structure is always better than lack of structure. KDD provides a knowledge structure that is intuitive and has the potential to assist industry and academia in their quest for quality delivery. Purpose of this website is to bring KDD out from a concept to a stage where a knowledge worker can actually get benefitted. 

Here are the four main challenges in industry and academia where KDD can positively influence:   

• Project delivery issues related to gaps in requirement and test coverage and knowledge erosion due to attrition and not keeping the project artefacts up to date.  
• Long learning curve in acquiring industry knowledge (such as knowledge of insurance, banking and retail) as it is considered vast, subjective and fathomless.
• Enterprise knowledge is stored in artefacts of heterogeneous formats (i.e., audio, video, presentation, spreadsheet, document, blog, wiki) resulting in delays in maintaining and retrieving it.
• Lack of formal exposure of industry knowledge in academia - it is a known industry-academia gap. 

These challenges allude to a lack of proper understanding of the domain knowledge related to the task being performed. The main challenge in learning domain (industry) knowledge is – it is vast, subjective and mostly available in unstructured formats. Manuals, Specifications, blogs, audios, videos and presentations are examples of unstructured knowledge formats.

Knowledge Driven Development (KDD) proposes a domain-agnostic structure that quantifies industry knowledge. It also extends this structure to cover enterprise knowledge and reuses it to drive software development in an artefact less project delivery environment. 

Knowledge Driven Development (KDD) is a combination of a framework and a methodology based on the reuse of structured industry and enterprise knowledge that in turn streamlines execution activities leading to a high-quality working product.   

Knowledge Driven Development (KDD) = Domain Knowledge Framework (DKF) + Atomic Knowledge Model (AKM)

Domain Knowledge Framework (DKF) is a simple and common framework to help acquire knowledge of multiple industries.

Atomic Knowledge Model (AKM) is a project delivery methodology based on a single source of structured project knowledge (based on enterprise knowledge structure) that is easy to specify and maintain and drives project. 

• Structuring industry and enterprise knowledge makes it quantifiable.
• Domain agnostic industry and enterprise knowledge structure.
• Empowering IT project delivery without producing project delivery artefacts.
• One (all encompassing) structure addressing multiple dimensions of industry and enterprise knowledge - product, process, business rule, business data etc.
• Navigating from high level to detailed level knowledge seamlessly in four levels of hierarchy.

• KDD structure is a conduit to link the mission and vision of an organization to the detailed knowledge required to drive initiatives and projects.  
• KDD structure provides a single source of structured knowledge that may not only bring significant savings in project delivery without compromising on quality, but it may also reduce the dependency on Subject Matter Expert (SME) and save on onboarding cost of inducting a new employee.  
• KDD is an asset to the learning and development function of an organization. It not only accelerates gaining domain competency, but also makes training measurable and accountable towards the day job of a knowledge worker. 
• KDD formalizes exposing industry knowledge to the students of computer science and management (due to its domain agnostic structure) increasing their employability quotient. 

• There are two books and many papers and articles (link provided in the 'RESOURCES' section) that can be used to gain a basic understanding of KDD. 
• A demo version of the KDD tool will shortly be made available through this website to have a better visualization on how KDD can help a knowledge worker. 
• We offer various aids as detailed in the 'WHAT WE OFFER' section to help industry and academia to adopt this concept and get benefitted from it. 

Gen AI uses a combination of deep learning techniques and sophisticated neural network structures to understand and respond to a question. The more information GEN AI has access to, the better answer it will generate. The GEN AI structure is complex and not humanly readable.  

KDD is based on a combination of hierarchical and knowledge graph based structures. It acts as a container to store information in a structured way both at high level and detailed level that is fit for purpose for a knowledge worker performing their day job. The structure is intuitive and humanly readable. The structure quantifies the information and reduces redundancy, duplicity and inconsistency making it easily maintainable. It provides a mechanism where a knowledge worker can get the information they are looking for in up to 4 layers of hierarchy.  

To exploit GEN AI effectively, one needs to know how to frame a relevant question. And it is not easy as demonstrated by the emergence of prompt engineering and requires necessary domain knowledge. This gap can be filled by the KDD structures if its initial hierarchical levels are populated for the respective domains.  

The more information KDD structure holds, the more useful it becomes for a knowledge worker. GEN AI can be used to populate information into the KDD structures particularly at the (detailed) later levels of the hierarchy so that the full potential of KDD can be realized.