Data Mining isn’t necessarily a new concept in Nigeria, Machine Learning is. It is like a new surface that needs excavation. Both are different concepts often used interchangeably. In most cases the implementation of both are used together.
Considering a lot of factors, Nigeria isn’t ready to adopt some of these tech innovations but its has no option. They have to adopt and implement them due to the benefits they bring to the table.
Presently, Nigeria may have been suffering from shortage of skills in this area. That is why Data Science Nigeria is in the fore front of transferring data science, data mining and machine learning skills to Nigerians. Boot-camps are being organized for these young Nigerians to beef up their skills in Data Science and Machine Learning.
These are mouthwatering fields in terms of financial rewards. And the demands for people the are skilled in those fields are becoming increasingly high. The knowledge gap is huge and the goal is to reduce the gap.
However, let me highlight the major differences between Data Mining and Machine Learning.
Data Mining:
Data Mining deals with the extraction of data or data patterns from a large set of data. It is easy, you mine for specific data’ from a large quantum of data sets. In other words, the Data Mining algorithm looks out for a specific data patterns given data set.
Machine Learning:
Machine Learning is the design and development of machine that can learn itself from a given set of data to achieve a desirable outcome without it being explicitly coded. Machine Learning literally means ‘a machine that learns on its own’.
Machine learning algorithms are of a few type like supervised learning, unsupervised etc.
Applications: spam mail detection, optical character recognition etc
According to Hao Zhang, a Professional in Deep Learning, says Data Mining is a computational process of knowledge discovering from large data sets.
His excerpt:
Specifically, data mining is the non-trivial process of identifying valid, novel, potentially useful, and ultimately understandable patterns from huge volume of data.
Roughly speaking, machine learning and database are two supports for data mining. Database offers data management techniques, while machine learning offers data analysis techniques.
Traditional machine learning research does not regard massive data as a target. Many techniques are designed to handle small and medium-sized data. If these techniques are used directly for massive data, the results may be poor or the algorithm may not work. Therefore, data mining must does some specific and non-trivial transformations to these techniques.
In fact, algorithms run in polynomial time may be considered very good in machine learning, but when facing massive data, data mining may not accept algorithms with time complexity. Therefore efficient data structure and data scheduling strategy, which database is good at, may be used to transform the machine learning algorithms.
