Machine learning (ML) is quickly evolving and spreading into many different areas of business and life in general. For this reason, growing number of companies of all size look how to utilize the potential of ML to enhance their products and services and get a competitive advantage.
One of such companies is young Spanish start-up based in Galicia, called OpSeeker.
It develops tools for financial mentoring of potential small investors by using knowledge based on behavioral finance. The initial challenge when starting an ML project in a small company is to select a suitable tool from the multitude currently available. To select a proper tool at the beginning can save precious resources as it will make all the subsequent efforts more effective. However, small companies may be lacking time or in-house expertise to perform this selection of a right tool-set.
Therefore, one of two goals of this thesis is to develop an easy-to-use methodology that would allow such companies quickly get a recommendation of a technology to use, based just on the properties of the project the company is working on and resource limitations.
This methodology will be then used as a basis for practical part of this thesis - to se-
lect proper tools for one of the Opseeker ́s projects - using ML for classifying users of their tool, based on the user’s behavior. The motivation behind this is to better understand the different groups of users in order to personalize the service for them and therefore provide a higher value.
The practical part of this thesis consists of executing a research among potential users of OpSeeker’s users to identify three different groups of users based on their stance towards investing. The data collected in this research will be used to train a classification algorithm that will serve as a basis for personalization of OpSeeker’s services in order to provide higher value for the its customers and users.