Archive for the ‘Knowledge Management System’ Category

Time has come to handle the knowledge-IC systematically, for use in strategic management of business. The knowledge over a period gets developed in the organisation and it resides in people, information files and databases. It is not always explicit but tacit in character and content, to know and acquire. To bring knowledge as critical input in the management process, it is necessary to have knowledge management systems. The knowledge management system therefore deals with definition, acquisition, construction, storage, delivery and application of knowledge. KMS handles two types of knowledge. One is tacit and other is explicit.

The knowledge has a structure and character.

While on the subject of why build KMS, it is necessary to understand ‘Knowledge’ in terms of its meaning, evaluation and migration.

Knowledge therefore is an essence of business management intelligence, residing in individuals, group of individuals, systems in the form of information set, models, processes and databases. Use of knowledge is critical to the organisation, hence knowledge creation, storage, distribution and delivery is very important calling for establishing formal KMS.

Knowledge Management System Architecture

KMS architecture deals with knowledge identification, generation and delivery for application in business.

Knowledge Generation:

After identification, definition and structuring, the knowledge process must be set for acquisition of knowledge. On acquisition, knowledge needs to be manipulated for understanding, presentation and usage.

Next step then is to integrate knowledge sets to build knowledge databases for access and distribution. Manipulation and integration processes would bring knowledge closer to its application at right place and in right structures.

The toughest job is to give meaningful definition and presentation to tacit knowledge for ease of use or application.

Many decision scenarios call for simultaneous application of tacit and explicit knowledge. Its generation as a set is difficult. It, however, can be achieved through training of concerned personnel in the organisation.

Knowledge Delivery:

One may create knowledge and place it in knowledge database, but owing to its nature, it needs to be protected and secured and also simultaneously made available to users for viewing, manipulating and application.

The system for access control, authorisation and authentication of knowledge for the purpose of update, alter, delete, etc. are necessary. Developing systems for packaging knowledge and for delivery for ready to use are also necessary.

Tools for Knowledge Management:

KM deals with knowledge generation, knowledge codification and refinement and knowledge transmission. The tools are available to handle all these functions of knowledge management.

The tools are:

  • Database management tools – For data management and seeking knowledge through SQL queries.
  • Data warehousing, Data mart, Data mining tools. – For business information creation and using Data Mining Tools, OLAP Tools to seek knowledge on different views and scenarios.
  • Process modelling and Management tools. – For recording standard process as an explicit knowledge for use in the organisation.
  • Work flow management tools. – For recording the process of workflow as an explicit knowledge for group workers.
  • Search engine tool. – For locating specific information through search algorithms.
  • Document management tools – like Lotus notes.
  • These tools are known as database management tools for documents. They are useful to search and manipulate documents to create new knowledge.

Knowledge-based Expert System (KBES):

Decision-making or problem-solving is a unique situation riddled with uncertainty and complexity dominated by the resource constraints and a possibility of several goals. In such cases, flexible systems (open systems) are required to solve the problems. Most of such situations, termed as unstructured situations, adopt two methods of problem solving, generalized or the Knowledge-based Expert System (KBES).

The generalized problem-solving approach considers the generally applicable constraints, examines all possible alternatives and selects one by trial and error method with reference to a goal. The knowledge-based problem-solving approach considers the specific constraints within a domain, examines the limited problem alternatives within a knowledge domain and selects the one with knowledge based reasoning with reference to a goal.

In a generalised approach, all alternatives are considered and the resolution of the problem is by trial and error, with no assurance, whether it is the best or the optimum, while, in the knowledge based approach, only limited alternatives are considered and resolution is made by a logical reasoning with the assurance of the local optimum. The generalised approach is dominated by a procedure or method, while the knowledge based approach is dominated by the reasoning process based on the knowledge.

Knowledge Base:

It is a database of knowledge consisting of the theoretical foundations, facts, judgments, rules, formulae, intuition, and experience. It is a structural storage with facilities of easy access.

Inference Mechanism:
It is a tool to interpret the knowledge available and to perform logical deductions in a given situation.

User Control Mechanism:
It is a tool applied to the inference mechanism to select, interpret and deduct or enter. The user control mechanism uses the knowledge base in guiding the inference process.

In the KBES, three components are independent of each other. This helps in modifying the system without affecting all the components. Like in the database application, where the data is independent of its application, in KBES, knowledge is independent from application, i.e., inference process. The KBES database, stores the data, the cause-and-effect relation rules, and the probability information on event occurrences.

For example, the knowledge base of Health Care would have a knowledge such as “obesity leads to high blood pressure,” “there are 60 per cent chances that smokers may suffer from cancer.” The KBES, therefore, stores and uses knowledge, accepts judgments, questions intelligently, draws inferences, provides explanation with reasons, offers advice and prompts further queries for confirmation.

In the KBES, the knowledge database uses certain methods of knowledge representation. These methods are – Semantic Networks, Frames and Rules.

The characteristic of a variety of tables are used to represent knowledge on table. A table in a drawing room inherits the characteristics of a table in a drawing room.


The second method of representing the knowledge is putting the same in frames. The concept of frame is to put the related knowledge in one area called a frame. The frame is an organised data structure of knowledge. The frames can be related to other frames. A frame consists of the slots representing a part of the knowledge. Each slot has a value which is expressed in the form of data, information, process and rules.


The third method of representing the knowledge is rule-based. A rule is a conditional statement of an action that is supposed to take place, under certain conditions. Some rules can be constructed in the form of ‘If Then’ statements.

Inference Mechanism:

Having created a knowledge database, it is necessary to create the inference mechanism. The mechanism is based on the principle of reasoning. When reasoning is goal driven, it is called Backward Chaining to goal and when it is data driven it is called Forward Chaining to goal.

For example, if there is a breakdown in the plant, then looking backward for the symptoms and causes, based on the knowledge database, is backward chaining. However, if the data which is being collected in the process of plant operations is interpreted with the knowledge base, it can be predicted whether the plan will stop or work at low efficiency. The data here is used to infer the performance of the plant and this is called forward chaining.

The choice between backward or forward chaining really depends on the kind of situation. To resolve a problem after the event, one has to go from goal (breakdown, stoppage, etc.) to data, i.e., it is a case of backward chaining. But if the question is of preventing a breakdown, then the data would be monitored in such a way if it is directing towards a goal (breakdown, stoppage), then it is a case of forward chaining.