John Morton has extensive experience in leveraging big data and analytics for business value. This includes roles as CTO for SAS and advising organizations on big data strategies. The document discusses differences between data, information, and knowledge. Data becomes information when processed, while knowledge allows predictions and depends on experience. It also outlines challenges in moving from data to knowledge, such as framing questions and addressing assumptions and data quality.
In the past :
I have come up from the mathematics and programming background accreting qualifications and honours through time.
Yes I have Maths, Astronomy and Information Systems engineering , including satallite communications in may background. I have spent over 20 years in the IT industry working on large scale industrialised solutions.
The last 10 years of so I have spent my life specifically looking at industrialised architecture and systems.
So if you can see something that no one else can, does that give you better knowledge and intelligence?
Or is that intelligence.
Information
Information refers to data that has been given some meaning by way of connection to other data. In computing terms it is data that has been processed. The ‘meaning’ applied to the data may not be useful. For instance, data stored in a database can be processed by some code to give information about something, for example a banking application can determine how a particular account balance increased by returning the record of the credit that occurred to that account, thus ‘information’ is retrieved about that transaction. Please note that without information, you will not have knowledge.
Knowledge
Knowledge is the concise and appropriate collection of information in a way that makes information useful. Knowledge refers to a deterministic process where patterns within a given set of information are discovered. We can also positively say that when a person memorizes some information about something, then they have knowledge about it. That knowledge will have some use, however that knowledge doesn’t in itself infer further knowledge. Take the example of school kids who memorize knowledge of the multiplication table (times table), for instance like the result of 3 times 3 is 9 (3*3=9), because they have amassed knowledge of the table. However, the kids will not be able to respond when asked the result of 2300*150 as that entry isn’t in the table. It takes true analytical ability and the ability to reduce the question to empirical factual knowledge, not just some memorized set of knowledge, to solve the problem.
The selection of mining methodology is very important in case of data mining. Different mining techniques are available based on the type of data to be mined. Also, factors including type of knowledge required as outcome of data mining, amount of data noise, size of data to be mined etc are also to be considered before selecting data mining method. The correct selection of data mining method is very important to get correct results.
Performance issues are also to be resolved as performance is the most expected factor in case of data mining. Though there are many different methods available for data mining, most of them do not perform as expected especially in case of large quantities of data. Also, data mining needs a skilled user to provide noise free data, select the data mining algorithm, make proper conclusions out of the output etc. Thus, data mining cannot stand alone by itself. Moreover, the predictions made with the help of data might not necessarily become true always. It could get affected by many other factors that were not available during the data mining process.