# Introduction to Big Data Analytics

Elina Loseva
06.12.2018.

### RATIONAL MOTIVATION

• Statistics and data analyses helps us to see overall picture of complex phenomenon
• Drawing conclusions from particular data usually will improve our knowledge of society itself (because we make world and around us, but not always can we make constructive deductions as good as programs, machines and mathematical functions) .
• Big data analytics can help to make predictions that can ensure efficiency of our lifes, organization functions, technology usage, etc.

### EMOTIONAL MOTIVATION

IF all the points in previous slide didn't convince You, then think about Alan Turing - one of the most impressive man of 20th century, who was not only great at math, cryptography (thus saving millions of lives at WW2) and is father of computing science. also he made some important contributions to the theory of Bayesian statistics.

Basically he founded some of the things I am dealing with know. And that makes me feel closer to this course.

### EVALUATION

Positive aspects (I):

• Course gives us knowledge about things that currently are very popular and it is always better keep Your knowledge up-to-date, and not learn outdated stuff.
• All materials are stored in earnig environment, they are accesible at any time.
• Lecturers in this course was understanding and patient, even when we had no idea about the simplest mathematical equations.

### EVALUATION

Positive aspects (II):

• Practical work in R Studio is big bonus.
• It is always great to hear practical examples from real life. That was one of the biggest advantage in this course. We heard things not only from theoretical aspects, but from personal experience. That helps to clear our narrow vision.

### EVALUATION

Negative aspects:

The subject of this course is rather complex, especially for person with no technical and mathematical background, I would suggest preparing extra information for total begginers, explaining so called “obvious” things (formulas, funcions of graphs, semantics of certain terms).

### PERSONAL THOUGHTS

Even if at fisrt this course is hard to grasp, it doesn't mean it it not interesting, on the contrary, understanding took some time, but we gained knowledge about modern technologies and data configurations. For me, person with background in humanities, it was great excercises for the left side of the brains. Even if from time to time, questions going around in my head seemed a bit dumb, nevertheless, they were answered with no condescending tone.

### FAVOURITE THING

I think, that environment and communication with learning staff is very important component, and in this course all lecturers were very competent and positive, and is the thing that I enjoyed most of all, light and pleasant atmosphere, that doesn't clug Your brains and the incoming infomation is easier to understand. And ofcourse the tirp to the lab was also a great experience. Most interesting lecture of all for me was about Artificial Neuron Networks, even though it is complex and a bit apocalyptical (Something from Terminator and Matrix combined), it was very exceptional.

### DISLIKE

Maybe a little bit parralyzing was my knowledge gap in this subject particularly. I would have prefered a little bit history of Big Data analytics, just to clarify the development of the scientifical aspects. Altough there was a lot of interesting facts in this course, just a hint - there's never enough.

### CLUSTER PLOT AND CODE

library(cluster)
library(fpc)

data(iris)

data_for_clustering <- iris[, -5]
clusters_iris <- kmeans(data_for_clustering, centers = 3)
plotcluster(data_for_clustering, clusters_iris\$cluster)


### CONCLUSION

Overall, outline of this course is logically arranged, course wasn't simple, but a little personal effort on our side could really do a lot of good in this course. Practical experience working with R made all the theoretical aspects easier to understand.

THANK YOU FOR YOUR TIME AND ATTENTION!