Data Mining: Applications, Machine Learning, and Current Technologies

This report explores data mining and machine learning, detailing their applications in current technologies. It begins with an introduction to data mining, explaining its process of extracting patterns from large datasets using methods that intersect machine learning, statistics, and database systems....Read More

Running Head: DATA MINING 1
Data Mining in Current Technologies
Name of student
School

DATA MINING 2
INTRODUCTION
This is the professional act of actually examining large or small datasets in databases that have
already been developed and are actually in existence. The main aim of doing data mining is to
aid in discovering patterns in large data sets and this is with the methods that intersect machine
learning, statistics, and database systems. There are more than one database systems. Some of the
database systems from which data is mined during data mining processes include, Microsoft's
SQL, Oracle's SQL, Mango DB, MySQL among others. These systems are preferred for different
reasons and different ones are compatible with different data science software, therefore, taste
and preference of the data science software of choice will be influenced by most of the things.
DATA MINING AND MACHINE LEARNING
After data has been mined, from any of the database software listed above plus many more, there
is usually the assignment, personal or imposed by an organization, lecturer, or self-driven to get
to find the actual insights that the data variables of the dataset have. These insights can be a
correlation between variables, variables' variability in matters correlation, getting the classes
under which variables fall through supervised and supervised algorithms. One needs to note the
fact that, in data mining, the process of actually extracting data from databases is just a first step
and the most skeleton step that is first done before other important steps are reached. Data
mining at large is an interdisciplinary area that involves both computer science and mathematical
formulas and applications. What this means is that after the actual retrieve of data from the
relevant sources, then there will follow an application of analyzing datasets by the use of coding
software and this is where the computer science discipline comes in.

DATA MINING 3
The mathematical discipline comes in only when directing the appearance of the results that are
being sorted for. What this means is that when one wants the mean of a variable then they can
easily, in the codes indicate that the mean is what is the need and that is the result that will be
given back. The second most important scenario is when there is the interpretation of the results
that have been arrived after an analytical code has been written down. This part is the most
important as it largely requires those that have a very strong analytical background (Lu, Setiono
and Liu, 2017).
In all the mentioned above, data mining's data extraction leads to the attainment of data
otherwise known as big data. Machine learning would result from all of that has been gotten as
big data as the same data will have to be analyzed, what this means is that big data is a subset of
machine learning. In machine learning though, the data retrieved will be analyzed through
different machine learning types. The types are supervised unsupervised, semi-supervised and
reinforced machine learning. All these machine learning types are existent just because data is to
be analyzed to aid in a decision that is to be made.
APPLICATION OF DATA MINING
Data mining tied up with machine learning aid in customer satisfaction as well as profits boosts.
Looking at every serviced based organization and the best example to take in this case is the
telecommunication sector. In this sector, there are lots of risks and what this includes losing
customers to other firms in the same sector. What this translates to is an actual loss of profits. To
help curb this menace, such companies have set up databases of different customers that they
might think would churn against those that promise to be very loyal to the firm. After this then

DATA MINING 4
classification using either logistic regression or any other understandable classification algorithm
can be used to get to classify those that are loyal and those that are not loyal. This model's
accuracy can be tested by the use of a confusion matrix whereby the classification of the
customers is checked if is done properly. From the classification, the firm can, therefore, know
very well where to invest in and how to convince respective customers not to churn out hence
helping keep up profits because it is cheaper to maintain existing customers than to attain new
ones (Zheng, 2015).
The second application of data mining is artificial intelligence whereby some algorithms have
been written that have aided in the development of the currently self-driven cars. Most people
have therefore cut on costs of actually having to employ drivers that would rather make them
spend more and more money in wages and salaries directed to the drivers. Those that drive
themselves as well so not have to hustle in the bid of having to direct the cars as wanted as the
cars only direct themselves. This saves energy in the long run.
CONCLUSION
Data mining and machine learning is the directive of most of the organizations as there is a
change in times and technologies have been advanced as well. Therefore, firms should up their
game in employment as well as update of all autos used for production as this would improve
profits in the long run.
References

DATA MINING 5
Lu, H., Setiono, R. and Liu, H., 2017. Neurorule: A connectionist approach to data mining.
arXiv preprint arXiv:1701.01358.
Zheng, Y., 2015. Trajectory data mining: an overview. ACM Transactions on Intelligent Systems
and Technology (TIST), 6(3), p.29.