Computational DataNotes
A business challenge in the information age: Corporations are dealing with torrents of data from their myriad business functions and operations. In many instances, businesses do not have the in-house expertise to analyse and mine their own data to extract the most pertinent/relevant information. Most importantly, there are no viable means to even visualize what the data looks like.
To extract value from (mostly) passive data , businesses require accessible tools to determine what value is inherent in their own data. For example, what constitutes normal vs. abnormal trends. The result is that quite often businesses do not recognise the significant value of the latent data. Data is partitioned in many spreadsheets disconnected from each other with lack of connectivity that blinds management to the value of their own data.
Spreadsheets do not represent the real value of the data: Due to anaemic training and limited functionality of spreadsheets, data is manipulated in simplistic ways. In the 21st century – Information Age – simple linear lists of numbers and/or names do not suffice to provide the most appropriate data analytics to operate successful businesses.
Case Study
Major US Fortune 500 PC manufacturer’s large sales lead database was classified to predict a new lead’s WON vs. LOST and the following is a list of Confusion Matrices as a tiny CloudNote.
The CloudNote is interactive, simply run your mouse over the matrices and no need to click. The actual classification accuracy data pops up!