My article will be divided into four parts. First of all, I'm going to give you the general introduction to mobile operators and tariffs. Secondly, I’m going to give you the general idea of clustering, Data mining and machine learning. Thirdly, I will focus on main algorithms which we usually use in Big Data. And finally, I’ll summarize the main points of my presentation and make conclusions.
A mobile network operator is a provider of mobile phone communication that controls the necessary network infrastructure and processes the billing. A country typically has 3 - 4 operators. Each of them has its own network and its own customers.
For example, in Moscow there are three main operators: MTS, Megafon and Beeline. You can see it in picture 1.
Figure 1 – number of subscribers of telecom operators
The rapid development of information technology, particularly advances in data collection, storage, and processing methods, has allowed many organizations to collect vast amounts of data that need to be analyzed. The volumes of this data are so large that the capabilities of experts are no longer sufficient.
Today, the direction associated with the intellectualization of methods of data processing and analysis is developing intensively.
Intelligent data analysis systems are designed to minimize the decision maker's efforts in data analysis and in setting up analysis algorithms. Many Intelligent data analysis systems allow not only to solve classical decision-making tasks, but are also able to identify cause-effect relationships, hidden patterns in the system being analyzed.
Big Data –is a collection of data that is huge in volume, yet growing exponentially with time. It is a data with so large size and complexity that none of traditional data management tools can store it or process it efficiently. Big data is also a data but with huge size.
Figure 2 – Big Data
Data mining is the process of sorting through large data sets to identify patterns and relationships that can help solve business problems through data analysis. Data mining techniques and tools enable enterprises to predict future trends and make more-informed business decisions.
Figure 3 – Fata Mining
Clustering is used to identify groups of similar objects in datasets with two or more variable quantities.
Figure 4 – clustering
Machine learning is a branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.
Figure 5 – Machine learning
All methods used in Data Mining can be categorized into groups:
- Methods in which the analyzed data are stored:
- Methods in which the analyzed data are not stored:
- Descriptive methods:
- Predictive methods:
- Statistical methods
- Cybernetic methods
I researched the tariffs of the main mobile operators and in the process of my research, I also created a graph that shows the growth of clients by year.
Figure 6 – consumer communication dynamics
Also based on my results I created a graph which shows the percentage of Internet use and calls by year for the main telecom operators in Moscow and we can see that the Internet is more popular, in consequence we can say that tariffs with a large package of Internet in our time is in great demand.
Figure 7 – Chart of internet and phone usage
My research also revealed the values of the average speed of the Internet, depending on the operator.
Figure 8 – the average speed of the Internet
A similar test was made by the Ookla. Ookla compared 6,334,730 user-initiated tests taken on the Speedtest IOS and android mobile from all the major mobile carriers in Russia during the award period to determine who showed the fastest mobile network speeds.
This picture shows the results of the ookla research.. the results of this research show approximately the same results that were obtained in my research using dada mining algorithms.
In conclusion, I would like to say that the use of algorithms allows you to analyze many areas of business. such analysis can help companies to understand their mistakes in doing business, including errors in working with large amounts of data, which in turn will help companies to grow.
Список литературы
- Li, Y., Chen, S. and Zhao, S. (2020) “Short text classification and clustering based Mobile Application Traffic Identification Method,” Journal of Physics: Conference Series, 1616(1), p. 012109. https://doi.org/10.1088/1742-6596/1616/1/012109
- Alho, L. et al. (2020) Machine learning based mobile network Throughput Classification, arXiv.org. https://arxiv.org/abs/2004.13148
