Download papers presented in the conference and proceedings below:


 Han Viet Thuan1, Phan Thanh Duc2,

1National Economics University, Hanoi, Vietnam

2Faculty of Management Information Systems,
The Banking Academy of Vietnam
, Hanoi, Vietnam

Abstract. In the context of the economic and social development of our world today, data and information resources are growing at an extremely fast pace. Big Data, an inherent feature of cloud computing, will offer unprecedented opportunities for both traditional and multimedia networks. Exploiting Big Data has become an indispensable trend in the socio-economic development process, which leads to profound changes in the financial and banking sector.

Vietnam is in the process of deep integration with the world, including the integration of science and technology. With more than 30 million people accessing the Internet, more than 15 million smartphone users, there is a great opportunity to exploit Big Data to enhance economic and social activities in this country.

France is a developed country, the sixth-largest economy in the world, with a gross domestic product of about 1.7 trillion euros. The application of Big Data in the financial and banking sector has brought tremendous and practical benefits.

This article presents the study and application of Big Data in banking sector in France and proposes some lessons for Vietnam.    

Keywords: Big Data; 5Vs Mod;, Information Value; Banking; Cloud Computing


Applying Big Data to Improve Customer Experience In The Commerial Banks

Nguyen Van Thuy

The Banking Academy of Vietnam, Hanoi, Vietnam

Abstract. The article focuses on the basics of customer experience at commercial banks and the application of Big Data in enhancing the customer experience, thereby providing recommendations to commercial banks of Vietnam.

Keywords: Big Data; Customer Experience



Big data analysis for credit scoring

Nguyen Huu Hai1, Nguyen Truong Thang2, Nguyen Viet Anh3,, 

1,2,3Institute of Information Technology – Vietnam Academy of Science and Technology

Abstract. Credit is a commonly used method to finance personal and corporate projects. The risk of debt default has prompted lenders to use a credit scoring system to help them make more informed decisions about credit expansion. Currently, most banks use a traditional credit scoring model based on borrower’s past financial history. Therefore, people without financial history might be excluded from the credit system. In this paper, we introduce an approach to building a credit scoring model for borrowers from telecommunication and social networking data. These data allow evaluating the personal behaviors as well as the socioeconomic status of users. This is a new approach, replacing traditional credit scoring methods when there is no financial history data. This is also a way to increase access to credit for users who have not been approached by banks or financial institutions. This method can also be used as an extra source of information to improve credit scores based on traditional credit scoring techniques.

Key words: Big data, bank, finance, credit scoring.




Nguyen Duc1

Performance Management unit, PVComBank. Previously,

PVComBank, Hanoi, Vietnam

Abstract. In the last few years, most of top 12 banks in Vietnam have implemented Data Warehouse, which mean they realized that data is very importance with banks’ activities. Data is a strategic asset also amount of data is increasing dramatically daily which lead to the demanding of having an independence unit who just focus on managing and exploring data. Setting up BICC became a hot topic inside banking industry and still many discussions to find the best way to establish BICC. In this article, the writer with almost 10 years of experience in Data Governance, Data Management, Data Warehouse and BICC, share idea how to establish BICC inside banks and key factors banks need to focus more in order to success\



Business process enhancement with process mining in commercial banks

Mai Tan Tai1, Chu Van Huy2,

1,2Faculty of Management Information Systems

The Banking Academy of Vietnam, Hanoi, Vietnam

Abstract. Improving business process plays an important role in management operations of commercial banks. In order to do so, it is necessary to apply advanced techniques of data mining and analysis. Being built on data mining and process modelling and analysis, process mining would be a crucial weapon for commercial banks to have an insight of their own organisation such as business processes, organisational structure and operational performance. This paper aims to introduce process mining concepts, tools and techniques as well as demonstrate a framework with a particular example, which gives a holistic view of process mining application into commercial banks.

Keywords: Process Mining, Business Process Management, Big Data.




Nguyen Thi Hoi1, Dam Gia Manh2 ,

1,2Information System and E- Commerce Faculty, Thuongmai University, Vietnam

Abstract. Today, with the explosion of social media such as social networks, blogs, content sharing sites, … every second, every minute there are many entries published. Huge source of information from social media is one of the unstructured data sources that form the current system of Big Data. Therefore, many researchers and interested in the social media, as well as Big Data has launched a number of ways to filter, sort, search or make posts similarities together based on fragments text, the short description or a specific attribute of the entry, … the question is how to estimate the similarities of the two entries on the social media was posted? … In this paper, we propose two issues: Firstly is modeling the entries published in some popular social media based on their attributes such as title, topic or  category, tags and content, …; And secondly it presents a model for  stimating the semantic similarity among these entries using library encyclopedia Wikipedia. Our model can be applied in data processing of unstructured, semi-structured, or compare the similarities between the user’s information in the online processing system and the user recommendation system.

Keywords: Text similarity, Entry similarity, Similarity measurement, Social media




Bui Thi Hong Nhung1, Ngo Thuy Linh2, Nguyen Thi Thu Trang3,

1, 2, 3Faculty of Management Information Systems

The Banking academy of Vietnam, Hanoi, Vietnam

Abstract. Process mining is a relative young research discipline that is a bridge between data mining and business process modeling. The goal of process mining is to discover, monitor and improve real processes by extracting knowledge from event logs recorder by a variety of systems. In the task of process discovery, there are some problems such as generate spaghetti-like process models when the input event logs are less structured, more flexible. An approach to overcome this is to cluster the input event logs into simpler event sub-logs that can be adequately represented by a process model. Traditional approaches to trace clustering were transformed the traces into a vector space and using some distance metrics such as Euclidean distance, Jaccard distance, Generic edit distance etc. But in these approaches there are lack of order and relationship between the activities in each traces. In recently research, we proposed a new trace clustering solution based on the idea of using the distance graph model for trace representation. Experimental results proved the effect of the proposed solution on two measures of Fitness and Precision when compared to contemporary approaches. In this paper, we continue to add new research on the idea of using the distance graph model for trace representation.

Keywords: event log; process mining; trace clustering; distance graph model; fitness measure;



A New Secure Sum Protocol

Vu Duy Hien1

1Faculty of Management Information Systems

The  Banking Academy of Vietnam. Hanoi, Vietnam

Abstract. Secure Multiparty Computation-SMC is a main method to construct Privacy Preserving Data Mining solutions. In SMC techniques, Secure Sum Protocol-SSP is the most significant. For SMC solutions and SSPs, the researcher’s challenge is optimizing three parameters: accuracy, performance and privacy. However, most of proposed SSPs are  not high performance or not enough private. In this paper, we present a new SSP which balances two parameters performance and privacy. Besides, we also construct a general binding model between these parameters above.

Keywords: Secure Sum, Secure Multiparty Computation, PPDM, Data Mining


Mathematical Model Of PageRank Algorithm

Nguyen Thi Thuy Anh1, Le Thi Hong Nhung2 and Ngo Thuy Linh3,,

1, 2, 3Faculty of Management Information Systems

The Banking academy of Vietnam, Hanoi, Vietnam

Abstract. With the growing momentum of e-commerce leading to a boom in the number of websites. Therefore, the Search Engine is a useful tool to help users in the world of the website. One of the ways to help Google’s search engine succeed today is to use the PageRank algorithm as part of the search engine ranking system in order to prioritize URL paths within. Search results page. The basic foundation of the PageRank algorithm is to evaluate the importance of web pages based on the link between them [2,4]. This paper presents the mathematical model of the PageRank algorithm and uses Excel to simulate the PageRank algorithm. This paper consists of 3 parts: Part 1 – The introduction of Pagerank algorithm, Part 2 – Theoretical Background, Part 3 – Simulate PageRank algorithm using Excel

Key word: Google; PageRank



Some Statistical Methods for Analyzing Big Data and Applications: A Survey

Le Si Dong1, Ta Quoc Bao2, Ha Binh Minh3,,

1,2,3Banking University Ho Chi Minh City
36 Ton That Dam, District 1, Ho Chi Minh City, Vietnam

Abstract. In this paper we introduce the concepts and the characteristics of big data. We review some most important statistical methods used for analyzing big data: Classification methods, Copulas methods, Approximate Stream Regression, and Symbolic data analysis. Furthermore, we discuss some applications of big data and provide some existing open problems that challenge to data scientists.



A Solution for Building Intelligence Estate Information System Based on Big Data

Nguyen Thi Thu Ha1, Truong Huy Hoang2, Nguyen Ngoc Linh3, Trinh Tung4,,,

1, 2 Electric Power University, 235 Hoang Quoc Viet, Vietnam

3 National University of Civil Engineering, 55 Giai Phong, Vietnam

4Academy of Policy and Development

Abstract. In every country, the real estate market plays an important role in the economy. It accounts for about 40 per cent of the each country’s total wealth. Therefore, the good management of the real estate market helps stabilize and develop the economy. In this paper, we present a solution for building a intelligence real estate information system which can manage real estate data and analysis automatic based on Big data from Internet. The real estate transaction information from the websites is aggregated on the system’s database. After that, we perform information extraction, information storage, information retrieval, analysis of information and then displayed the processing results through reports and charts. We also built an application on IOS and performed experimental. The results show that, it is really significant for developing this model on the wide range.

Keyword: Data analysis; information retrieval; information extraction; big data; information system; real estate; developing the economy.



Phan Thanh Duc1, Le Ngoc Tu2,

Nguyen Quynh Anh3, Tran Phuong Lan4, Chu Thi Huyen Trang5,,,,

1, 3, 4, 5Faculty of Management Information Systems

The Banking Academy of Vietnam, 2Saslesforce Corporati

Abstract. Recently, Big Data is a term mentioned as a solution for collecting and processing huge sources of information. The increase in volume, velocity, and variety of information as well as the authentication are making the data more and more difficult to analyze, which makes the implementation of Big Data projects become an urgent need. There have been a number of studies on Big Data technology worldwide, in Vietnam, however, an appropriate approach to the implementation of Big Data project seems not to get enough attention.

It is the fact that traditional systems approaches are no longer appropriate to Big Data project. Nevertheless, the deployment of Big Data technology for CRM in the banking industry in Vietnam has not been thoroughly reviewed and evaluated. Therefore, in order to implement effectively Big Data technology into the CRM system, the article will focus on analyzing the problems in customer relationship management in Commercial Banks in Vietnam, the recommendations for approaching the implementation of Big Data for CRM project to solve the identified problems will be given afterwards.

Keywords: Big Data; CRM; System approach;



Apply Mask To Enhance Detection Of Financial Fraud For Enterprise

Nghiem Thi Lich1,  Nghiem Thi Toan2,

1Dept. of economic information system and e-commerce, Thuong mai University, Ha Noi

2Dept. of mathematics, Ly Nhan Tong high school, Bac Ninh

Abstract. Fraudulent high financial reporting has dominated the market. Financial fraud detection (FFD) has been important in the fight against the severe consequences of financial fraud. The discovery and prevention of financial fraud has attracted great interest from researchers. This paper presents MASK method, which is new method in data mining techniques, applied in fraud detection. The algorithm is tested on the data obtained from the UCI standard data warehouse, and other actual datasets. Results showed MASK is superior methodology to adjust the data as Over-sampling, Random Under-sampling, and Borderline SMOTE when applying the classification algorithm SVM, C50, and RF. Mask method is an effective classification technique for unbalanced data, which enhance detection of financial fraud in enterprise.

Keywords: Financial fraud detection, classification, imbalanced dataset, MASK



Applying SVM Networks In Foreign Exchange Market

Nguyen Thi Thu Thuy1

1Informatics Department, ThuongMai University, Hanoi, Vietnam

Abstract. ForReign Exchange Market is one of busy financial market in the world. It has been worked 5 days a week. The exchanged problem is how to predict a strategy of buying/selling exchange rate between alternative money such as USD vs EUR. This is based on the numerous historical data from Forex market. Neural Network, in particular to Support Vector Machine can improve the predicted results. The alternative parameters and number of support vectors are used to produce appropriate models. In this paper, 6 alternative models using Gaussian and polyminial kernel functions are experimented. The use of Poly models have shown better results compared to the ones of GsRBFs.

Keywords: ForReign Exchange Market, Exchange, Support Vector Machine (SVM), Support, Prediction.


Big data and the ability to promote the marketing effectively in Vietnam enterprises

Chu Van Huy1, Mai Tan Tai2,

1Faculty of Management Information Systems

The Banking academy of Vietnam, Hanoi, Vietnam

Abstract. One of the objects proposed to enterprises is to “promote the ability of approaching, supplying the products-services to customers”. Business departments in enterprises, especially, the marketing department which have many efforts at specializing works. However, the interaction and development in market share in business have not satisfied the expectations. The big data exploitation technology (Big data) are now considered as a breakthrough solution, a turning point in the ability of approaching, understanding customers effectively and firmly. This paper mentions the necessary, strategy for applying, proposing the model of Big data implementation to get an advantage of promoting the business in Vietnam enterprises.

Keywords: Big data; Enterprise information systems; Business intelligence; Machine learning, Service oriented architecture.




Le Van Hungg1, An Phuong Diep2,

1,2Faculty of Management Information Systems

The Banking academy of Vietnam, Hanoi, Vietnam

Abstract. The bank benefits mainly from lending but this activity also contains many risks. Therefore, it is very important to collect and store customer information so that data mining can be used to find out the rules. They give the bank many advantages in making the right decision to reduce the risk and increase the efficiency of bank lending. In this study, we would like to introduce the Business Intelligence toolkit and the application of bank lending decision support.

Keywords: Big data, association rule, decision,bank, business intelligence.



Deep Learning for Credit Scoring in the Era of Big Data

Le Quy Tai1, Giang Thi Thu Huyen2,

1L3S Research Center, Leibniz University Hannover, Germany

2Management Information System Faculty, The Banking Academy of Vietnam

Abstract. Credit scoring plays a crucial role in the risk management process of commercial banks. Recently, Deep Learning techniques have driven remarkable improvements in classification performances in some applications of machine learning fields such as image classification and automatic speech recognition. The ability to analyze and learn massive amounts of data is a key benefit of Deep Leaning models. In this paper, we adapt two deep learning architectures to credit scoring: 1) Sequential Deep Neural Network; 2) Convolutional Neural Network. We use four metrics to evaluate the performance of these algorithms on three different data sets. We find that all the neural nets achieve better scores than traditional approaches (kNN, Naive Bayes, Decision tree and SVM).

Keywords: deep learning; neural network; credit scoring; big data



Factors that influence the adoption of Internet Banking in Vietnam

Tran Thi Hue1, Nguyen Thanh Thuy2,

1, 2Faculty of Management Information Systems

The Banking academy of Vietnam, Hanoi, Vietnam

Abstract. The purpose of this article is to determine the factors that affect the adoption of Internet Banking services in Vietnam. Different factors are conceptualized in the provided model and the relationship between them and the adoption of Internet Banking is also considered. A total of 500 respondents in Vietnam were sampled for research: 350 were Internet Banking user, 150 were not Internet Banking user. Factor analyses and regression technique are used to study the relationship. The results of the model tested clearly that use of Internet Banking in Vietnam is influenced most strongly by convenience, risk, security and prior Internet knowledge. The research also propose that demographic factors impact significantly Internet Banking behaviors, specifically, occupation and instruction. Finally, this paper suggests that an understanding the factors affecting intention to use Internet Banking is very important to the practitioners who plan and promote new forms of banking in the current competitive market.
Key word: Internet Banking; adoption; factor analysis; e-commerce




Cao Thi Nham1, Vo Van Luong2,

1,2 International School

 Duy Tan University, Danang, Vietnam

Abstract.  In recent years, banks have been competing by offering many services through different channels such as internet banking, mobile banking, telephone banking, ATM, etc. Meanwhile, human is becoming more and more busy. They do not even have time to do individual tasks. Therefore, it’s necessary to select the most suitable services to introduce to each customer. This will save time as well as increase customer satisfaction and efficiency in developing bank services. The aims of this paper are proposing an architecture for deploying a recommender system based on collaborative filter technique.

Keywords: collaborative; recommender system




Le Van Hung1, An Phuong Diep2,

1,2 Faculty of Management Information Systems

The Banking Academy of Vietnam, Hanoi, Vietnam

Abstract: In the socio-economic development of Vietnam today, the financial forecast is more and more concerned. Investing in the stock market requires experience and knowledge of the investors. Data mining techniques which are used to forecast market fluctuations are good hints for investors to make informed trading decisions. In this article, we present the ARIMA forecasting model and make a trial prediction for the stock price of Ocean Group Joint Stock Company

Keywords: data mining, forecast, ARIMA, stock, investor;



The application of Big Data in the commercial banks’ credit activities

Do Thi Van Trang1, Bui Ngoc Phuong2

1,2 Faculty of Finance, The Banking Academy, Hanoi, Vietnam

Abstract. The Fourth Industrial Revolution has a strong impact on all sectors, for example, economy, politics, society, finance and banking. One of the main technologies of this revolution is the Big Data which is playing an important role in the long-term development strategy of commercial banks. Big Data is applied to a large number of commercial banks’ activities, for instance, satisfying the customers’ demand, managing risk and capital resources, as well as enabling banks to detect and prevent frauds or errors. This article mentions the application of Big Data in corporate finance analysis in lending activity of commercial banks.

Keywords: Big Data, credit activities, and commercial banks.



Big data approach toward measurement of customer lifetime value and market segmentation: The case of telecommunication sector in Vietnam

Ha Hien1, Nguyen Kim Thanh2, Tu Van Binh3,,

1,2,3Fastdata Solutions, Hanoi, Vietnam

Abstract. With application of model of customer lifetime value (CLV) and its calculating is based on source of data mining from big data of a telecommunication company in Vietnam, who is asked to be hidden its name, because of confidential condition. Anyway, data source is provided by the department of information technology of the company, which a sample size collected is 245,355 prepaid subscribers who are active customers by a cell phone. Sampling is randomly selected from entire customer base of more than 34 million subscribers from population of the telecommunications company. As stated, these customers were active from July 31, 2016 to returns pre (before). Months considered to collect to measure customer survival/churn behavior are followed for 6 months, this means from January 2016 as the origin of time to June 31, 2016 as the observation terminated time. Finding out CLV of each prepaid subscriber is carried out and its continuous application on market segment is an important step to support marketing strategy of the company to maintain the market place and limit churn rate.



Social Network Analysis for Vietnamese Commercial Banks: Applications, Principles and Challenges

Dinh Trong Hieu

Faculty of Management Information Systems

Banking academy of Vietnam, Hanoi, Vietnam

Abstract. Social network analysis is increasingly focused. Understanding the benefits and challenges of using social networking in business will help commercial banks take the right steps and further improve overall business performance.

Keywords: Social network, social listening, social network analysis, big data, data mining, commercial bank.



MS. Luu Minh Tuan
Assoc. Prof. Han Viet Thuan
PhD. Nguyen Trung Tuan
National Economics University
Abstract. The development of information technology (IT) has brought useful values to the society – economy, as well as creating the complex issues about information security, especially the risk of the information loss. Recently, the types of illegal intrusion of the IT systems at universities, colleges, research institutes (are called universities) steal the data to be increasingly and affected level is more serious. The information security is essential, plays an important role deciding on the universities’s development. At National Economics University (NEU), information security in management, training and scientific researching activities is very urgent and necessary. One of the most effective information security solutions is the use of the cryptosystems to secure documents and data. This article aims to study the application of the Elgamal public key cryptosystem to documents information security at NEU.

Keywords: public key cryptosystem, Elgamal cryptosystem, documents information security