Question

As the Data Manager (DM) on the “TES-100” project, the CEO has decided to proceed with the study and wants you to consider using the new savvy company system of remote data entry/electronic data capture. Review the current literature (i.e. internet) regarding this topic and decide if this study would be RDE/EDC appropriate, then write a 2-3 page paper or 20-25 slides on why you have made the decision you have made including pros and cons.

Answer 1

LITERATURE REVIEW

Background

Data mining is a very important method to find information concerning your client behaviour towards your business offerings. It explores the unknown credible patterns those square measure vital for business success.

Data mining has usually misunderstood; folks suppose that it includes solely process of knowledge however is really way more than this i.e. it covers advanced tools and technologies.

According to Doug Alexander of the University of TX it's truly outlined because the “computer-assisted method of dig through and analysing huge sets of knowledge then extracting the which means of data”.

With data mining, Business Organizations square measure able to build a lot of correct business choices and incur a lot of profits. From business, selling advertising and introduction of recent merchandise or services, and everything in between.

Problem known

Data mining may be a artistic method which needs variety of various skills and information. presently there's no commonplace framework during which to hold out data processing comes. this suggests that the success or failure of an information mining project is extremely enthusiastic about the actual person or team carrying it out and prosperous follow cannot essentially be recurrent across the enterprise. data processing desires a regular approach which can facilitate translate business issues into data processing tasks, counsel applicable knowledge transformations and data processing techniques, and supply suggests that for evaluating the effectiveness of the results and documenting the expertise.

Methods and Tools used

The CRISP-DM (CRoss business commonplace method for knowledge Mining) project addressed elements of those issues by process a method model that provides a framework for concluding data processing comes that is freelance of each the business sector and therefore the technology used. The CRISP-DM method model aims to form giant data processing comes, less expensive, a lot of reliable, a lot of repeatable, a lot of manageable, and faster.

Why the information Mining business desires a regular method Model

For the market, there'll be several edges if a standard method model is accepted. The model will function a standard point of reference to debate data processing and can increase the understanding of crucial data processing problems by all participants, particularly at the customers’ facet. however most significantly, it'll produce the impression that data processing is a longtime engineering follow. Customers can feel more well-off if they're told the same story by totally different tool or service suppliers. On the a lot of sensible facet, customers will get a lot of affordable expectations on however the project can proceed and what to expect at the tip. handling tool and repair suppliers, it'll be a lot of easier for them to check totally different offers to choose the simplest. a standard method model also will support the dissemination of data and skill among the organization.

The CRISP-DM methodology is delineated in terms of a stratified method model, comprising four levels of abstraction (from general to specific): phases, generic tasks, specialised tasks, and method instances (see figure 1). At the highest level, the information mining method is organized into alittle range of phases. every part consists of many second-level generic tasks. This second level is termed generic, as a result of it's meant to be general enough to hide all attainable data processing things. The generic tasks square measure designed to be as complete and stable as attainable. Complete suggests that to hide each the entire method of knowledge mining and every one attainable data processing applications. Stable implies that we would like the model be valid for nevertheless unforeseen developments like new modelling techniques. The third level, the specialised task level, is that the place to explain however actions within the generic tasks ought to be meted out in specific things. for instance, at the second level there's a generic task referred to as build model. At the third level, we would have a task referred to as build response model that contains activities specific to the matter and to the information mining tool chosen.

The fourth level, the method instance level, may be a record of actions, decisions, associate degreed results of an actual data processing engagement. A method instance is organized in step with the tasks outlined at the upper levels, however represents what truly happened in an exceedingly specific engagement, instead of what happens normally.

The CRISP-DM methodology distinguishes between the Reference Model and therefore the User Guide. Whereas the Reference Model presents a fast summary of phases, tasks, and their outputs, and describes what to try to to in an exceedingly data processing project, the User Guide offers a lot of elaborate tips and hints for every part and every task among a part and depicts the way to do an information mining project.