The decrease in entropy or surprise achieved by modifying a dataset is known as information gain, and it is frequently employed in decision tree development. In order to estimate information gain, the entropy of a data sample before and after a modification is used (Kurniabudi et al., 2020). This phenomenon is possible by increasing the amount of data, as well as improving the methods of their analysis, which significantly elevates the possibility of assessing and predicting the conditions of the banking system. Information gain can be determined as a methodology that can be practically applied to the digitalization of payments in the banking sphere (Jadhav et al., 2018; (Jeske et al., 2021)). The first real-life example of the application of information gain is the use of not only demographic and economic data but also access to analysis of social media profiles of customers.
New data sources allowed the bank to significantly improve targeted analysis to improve products and marketing strategies, as well as to acquire new channels of communication with customers (Laksamana, 2018). The second example is the introduction of machine learning principles into the processes of data analysis used by the banking system. In comparison with the old approaches, this system allowed to significantly increase the possibility of risk assessment, which optimized the bank’s activities (Leo et al., 2019). The introduction of new approaches allowed to increase the bank’s revenue by 14%, and clientele by 16% over a six-month period. This growth was made possible by collecting more diverse data and analyzing more information to develop more relevant strategies.
The potential benefits offered by the use of information gain contribute to my professional development as a banker. First of all, it became possible for me to communicate more informedly with clients in order to determine their needs and form the best offers based on the data available about them. A potential benefit of implementing machine learning is that the banking system can now analyze more data, which allows it to identify current trends. On the basis of a larger set of data, the bank can form large arrays to analyze the patterns of behavior of banking consumers (Provost and Fawcett, 2013). For me, as for a banker, this is a paramount benefit, as I can better analyze the banking environment and help in the development of the organization based on the available data. From an initial perspective, these improvements in the future will allow me to delve into the study of various aspects of the digitalization of the banking system, which will open new career prospects for me.
First of all, I used decision trees to optimize offers for clients based on their personal data. On the basis of multiple data obtained during the analysis of various sources of information, I was able to obtain the most significant points that fit into the needs of the client. Secondly, I used decision trees when providing services to clients to optimize the response time to their questions. Using decision trees within the appropriate software, it is possible to reduce the number of steps for making a decision, which greatly simplifies communication and solving customer problems. Activity data allowed me to make more accurate decisions based on personalized data, as well as avoid mistakes when working with customer information and its processing. Thus, decision trees increased the effectiveness of both organizational and personal decision-making.
Reference List
Jadhav, S., He, H. & Jenkins, K. (2018) ‘Information gain directed genetic algorithm wrapper feature selection for credit rating.’ Applied Soft Computing, 69, pp.541–553.
Jeske, T., Würfels, M. and Lennings, F. (2021) ‘Development of digitalization in production industry – impact on productivity, management and human work.’ Procedia Computer Science, 180, pp.371–380.
Kurniabudi et al., (2020) ‘CICIDS-2017 dataset feature analysis with information gain for anomaly detection.’ IEEE Access, 8, pp.132911–132921.
Laksamana, P. (2018) ‘Impact of social media marketing on purchase intention and brand loyalty: evidence from Indonesia’s banking industry.’ International Review of Management and Marketing, 8(1), pp.13-18.
Leo, M., Sharma, S. and Maddulety, K. (2019) ‘Machine learning in banking risk management: a literature review.’ Risks, 29, pp.1-22.
Provost, F. and Fawcett, T. (2013) Data Sciences for Business: What you need to know about Data Mining and Data-Analytics Thinking, 1st edn. O’Reilly Media: Sebastopol, CA.