quick search
GO

quick links
Consultancy Overview 
Finisco Knowledge Solutions consultancy services conform to the Data Mining Process Industry Standard (CRISP-DM) and enable you to harness our market leading skills to assist you in developing or refining engineered solutions particularly in the area of Predictive Analytics. Our consultancy services cover the following areas:

Task Decomposition
It is often that the task as defined by the business is at a very high level and needs to be decomposed into sub tasks. Some of these tasks may involve integration with the current environment while others may actually correspond to tasks within related business processes. The identification of these tasks and the development of a project plan is the key objective of Task Decomposition.

Current Systems Assessment
The starting point as always is the gaining of an understanding of the current systems and documenting the information available.

Information Needs Analysis
The objective of this task is to understand the different actors within the customer organisation and their requirements for information from the system. These actors may be internal to the organisation, for example, management, account managers, internal intermediaries or external to the organisation such as customers, service providers/intermediaries. An understanding of the needs of these actors will enable the development of a 'Requirements Document' for the system and will also determine the data requirements.

Business Process Analysis
The business processes need to be analysed with a dual purpose:

  • Consumers of information from the system
  • Generators of data that needs to be stored within the system

    The impact of the different processes on the system requirements and the data model used will be studied and documented within this task.
  • Legacy Systems Integration
    A key requirement is to ensure that the data being generated by the various legacy systems and other systems under development within the organisation are not creating data silos. It is essential that the data being generated by these systems are stored in a consistent data model. The new data requirements that emerge from the Information Needs Analysis will add to the complexity of the data model so it is essential that the data model be developed at an early stage.

    Knowledge Needs Analysis
    This task will focus on the knowledge needs with the organisation. The knowledge generation will focus on the knowledge required by the marketing, operational and management functions of the organisation.

    Data Preparation
    Using the Knowledge Needs Analysis as the basis, we are now in a position to develop a pre-processing regime for the data to ensure that any inconsistencies in the data are handled and the data is made ready to discover knowledge from. The transformations on the data are an essential part of the data mining exercise and they need to be productionised when the knowledge is deployed to transform, often in real-time, the data prior to application of the knowledge.

    Knowledge Discovery and Validation
    This task focuses on the actual task of discovering the knowledge required from the data and validating the knowledge, using sampling techniques to ensure that the patterns discovered are modelling the underlying model hidden in the data and not noise within the data. This may also entail test deployments of the knowledge and measuring the lift generated by the model.

    Knowledge Deployment
    This task deploys the knowledge discovered within a production environment. Despite the myth that the knowledge once deployed will generate increasing value for the business, an important factor in the deployment is the development of metrics for monitoring the effectiveness of the knowledge deployed so as to flag points where the process of knowledge discovery must be repeated, knowledge validated and deployed once again due to changes in the environment in response factors including the deployment of the previous model.

    Horizontal Issues
    There are a number of other issues that are not distinct tasks but affect the overall architecture of any system. These are:

  • Security
  • Data Privacy
  • Audit
  • Data Quality
  • Risk and Compliance Management
  • Standards Compliance/Integration/Future Planning
  • Data Extract, Transform and Load from old CIS to new CIS.
  •  We are delighted with the SEO services provided by Finisco, which has contributed to our companies top rankings in major search engines. We would recommend Finisco to any company wishing to improve their online visibilty.


    Clair Sherwood-Parkin
    Plimsoll