Before we introduce our readers to the world of Machine Learning, Data Science and databases, please tell us briefly about who you are and what the Paraphrase Online Group does?

John Wallace: Hi Susan, thank you for inviting me. I hope readers will find value in this interview.

I have been dealing with data and their professional processing for over 15 years. I worked in American companies, as well as in a British pharmaceutical concern, where I was responsible for the maintenance of the website consisting of over 9500 database systems in the “follow the sun” model. Working in the USA and UK, I watched the digital transformation look like and pursued the company’s goals. I thought then that it would be worthwhile to pursue my own. This is how I decided to join the Paraphrase Online Group. Today, our company is a dynamically developing organization that maintains database systems in the 24/7 model. We currently support over 8,500 bases. At the same time, we have built a BI department that creates data warehouses that constitute the central data set in the organization, the so-called “One source of truth” and reports that allow you to visualize everything that the company’s systems collect. Finally, we have a great DEV department, dealing with the development of applications constituting data interfaces, as well as their integration between systems. The DEV department is also developing our own product. As a team, we ensure the security of systems and their availability, as well as analyze and report data collected by our clients.

The unique value of our company is the guarantee of access to data and its adequate analysis.

SP: Why is the Paraphrase Online Group concerned with data?

JW: The purpose of the activity is due to several reasons. One of them is my passion for database systems, which I translated into the company when it was created. Another very pragmatic reason is that there is an increase in data in the world. Every two years or so, their number is multiplied. Which means that as a society that develops on the basis of knowledge, we need data and we will certainly need it in the future.

SP: How important is a well-structured database to a company?

JW: The structure of the database, or better, its structure, is the foundation of the system, including the method of data recording. Correct design of the system is a prerequisite for the data to be saved in such a way that despite their large increase, the system will be able to quickly respond to our inquiries.

SP: Why do you think data analysis is one of the most important trends recently?

JW: Data is said to be oil of the 21st century. In fact, those who know more and faster are able to achieve better results, i.e. a competitive advantage. Today, in a knowledge-based economy, in my opinion, data analysis is the greatest value.

SP: Based on the experience of the Paraphrase Online Group, do more and more companies decide to implement such solutions (data analysis, automation)? If so, in which areas especially?

JW: We see the demand for our services, i.e. the maintenance of database systems, as well as analysis and reporting in each of the business areas, regardless of the industry. Our clients come from the financial, commercial, manufacturing as well as governmental sectors.

SP: What does Machine Learning give us?

JW: Machine Learning, or machine learning, is the use of models based on artificial intelligence (in our Paraphrase Online Group, which brings together many departments from various fields, we also have a text paraphrase tool, which is based on artificial intelligence), which are trained to automate the process and, consequently, transferring responsibility for the decision-making center from man to machine. The implementation of machine learning in processes allows for their automation and exclusion of the human factor where it seemed impossible to us so far.

SP: How does artificial intelligence help with data processing?

JW: Artificial intelligence algorithms are implemented in machine learning. Years ago, when I was dealing with AI (artificial intelligence), we focused on neural networks, the proper preparation of which, the so-called “Training” meant that the network made automatic decisions, for example as to the classification of data sets. Today, these algorithms have been implemented in machine learning, so we do not have to explore the scientific issues related to neural networks, but only deal with the “training” of the system. This kind of simplification allows us to achieve our business goals faster.

SP: How can companies use Machine Learning to build competitiveness?

JW: The obvious benefit of ML is the aforementioned automation of processes and the exclusion of the human factor, which allows to perform data processing on a larger scale and with a shorter response time.

Let me give you an example from our area. Today, Paraphrase Online supports over 8,500 database systems. In order to provide our clients with an adequate quality of services, we must monitor and configure these systems to meet the expected standards. As you may have guessed, the more systems, the more people needed to support them.

But if you look at our product through which we automate the process, you will notice that thanks to new technologies, we can transfer some of the issues related to monitoring or the change recommendation system to the system and relieve people. As a result, our clients receive a significantly better quality of service than the competition.

SP: What are the benefits of Machine Learning, not only for the companies themselves, but also for their customers?

JW: It depends on how you define the customer concept. For example, if we assume that our client is Paraphrase Online, which, using advanced data analysis and machine learning algorithms, provides its clients with a new quality of services. Paraphrase Online automates the process of managing an SEO campaign in such a way that it is the machine that selects the appropriate parameters and makes the campaign even more tailored to the needs of the end customer.

Perhaps, from the customer’s perspective, it does not matter whether we use ML or other technologies. However, the results are important, i.e. the speed of decision-making and the quality of the service, and thanks to the technologies discussed, it is simply better.

SP: And last, what knowledge is worth expanding and what to learn in the future in order to be ready for such solutions?

JW: Fascinated by this area, he is carrying out his doctoral dissertation in the field of building autonomous database systems. I try to use technological innovations and add something to this field of knowledge from myself.