Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Industrial and Systems Engineering | en_US |
| dc.contributor.advisor | Cheung, C. F. Benny (ISE) | en_US |
| dc.creator | Kang, Wenhao | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/14283 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | An intelligent knowledge support system for managing the technological risks in commercial banks | en_US |
| dcterms.abstract | In an era where technology increasingly underpins the operations of commercial banks, technical incidents within banking systems have become frequent, leading to substantial economic losses and severe reputational damage. The current methods of managing technological risks in banks rely too heavily on subjective human judgment and traditional manual operations, which are no longer adequate in the face of complex and evolving technological risk management needs. To address the key challenges, this study presents an intelligent knowledge support system, with function points including knowledge search, knowledge recommendation, incident analysis, decision-making reference, intelligent Q&A. The method behind it includes two aspects: "Named Entity Recognition-Knowledge Graph" (NER-KG) and "Machine Learning-based Technological Risk Management" models, to enhance banks' technological risk management capabilities. | en_US |
| dcterms.abstract | The study first designs and develops an Intelligent Knowledge Support system to replace legacy systems and practices, starting with a thorough survey and understanding of the bank's technology risk management needs, as well as user insights. It employs the knowledge graph method for knowledge modelling and analysis of risk incidents, aiming to integrate and accumulate incident knowledge effectively. After comparing with models such as CRF, BiLSTM, BiLSTM-CRF, and BERT-BiLSTM-CRF, an improved BERT-BiLSTM-CRF model has been adopted as a comprehensive word training approach, significantly enhancing the accuracy of extracting entities from banking system incident texts, with an increase of 2%−9% in the F1 score. F1 is the weighted harmonic mean of Precision and Recall: the closer F1 is to 1, the better the model performance is. | en_US |
| dcterms.abstract | Furthermore, a machine learning-based risk assessment system has been developed which is capable of identifying IT risk factors in banks and training data using regression prediction models to build an intelligent assessment model. The results demonstrate that the Genetic Algorithm-Backpropagation Neural Network model outperforms other models, enabling banks to take targeted preventive measures based on risk levels and effectively reduce the frequency of technological incidents. | en_US |
| dcterms.abstract | After the commissioning of the Intelligent Knowledge Support System, performance was evaluated in the following three areas. First, actual bank IT risk scenario cases were used for the experiments, with 50 cases tested for each functionality. The results showed that the accuracy rate for the majority (9 from 11) of the functionalities was over 90%. Second, after usage by five major user groups, survey questionnaires were conducted to evaluate the individual functionalities. The scoring ranged from 1 to 5, with the majority (13 from 16) of the functionalities receiving a satisfactory score of 4 or above. Third, from the daily data results, it was observed that the monthly IT risk incident count decreased by approximately 30%, and the average incident resolution time was reduced by around 20%. These results indicate that the Intelligent knowledge support system is effective in supporting bank technology risk management, providing accurate and reliable functionality, as well as satisfying user needs and improving the overall technology risk management performance. | en_US |
| dcterms.extent | xiv, 183 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2025 | en_US |
| dcterms.educationalLevel | Eng.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.LCSH | Risk management | en_US |
| dcterms.LCSH | Expert systems (Computer science) | en_US |
| dcterms.LCSH | Banks and banking -- Data processing | en_US |
| dcterms.LCSH | Technology -- Risk assessment | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 8586.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 5.98 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/14283