J Comput Virol Hack Tech DOI 10.1007/s11416-016-0288-9

ORIGINAL PAPER

A study on the service and trend of Fintech security based on text-mining: focused on the data of Korean online news Guozhong Li1 · Jian Sheng Dai1 · Eun-Mi Park2 · Seong-Taek Park3

Received: 18 August 2016 / Accepted: 16 December 2016 © Springer-Verlag France 2017

Abstract The purpose of this research is to draw the direction and tasks for Korean Fintech industry taking into considerations of the importance of successful operation of Fintech and security. News data from Naver the most famous Korean portal site have been collected and analyzed and 20 most frequently used keyword were ranked both in 2015 and 2016. Payment, platform, banking, etc of Fintech service has become keywords since Fintech started its service in earnest in 2015, and security, enterprise, support, finance etc. are the keywords in 2016. The results show the difference in terms of importance. The results of this research also provide guidelines of security principles for enterprises who provide Fintech services. Keywords Security · Fintech · Text-mining · Keyword network · Crawling

B

Seong-Taek Park [email protected] Guozhong Li [email protected] Jian Sheng Dai [email protected] Eun-Mi Park [email protected]

1

Department of Management Information Science and Information System, Kunming University of Science and Technology, No. 68, Wenchang Road, 121 Street, Kunming 650093, Yunnan Province, China

2

Department of Business Administration, Kyungpook National University, 80 Daehak-ro, Buk-Gu, Daegu 41566, South Korea

3

Department of Management Information Systems, Chungbuk National University, Seong Taek Park, 1 Chungdae-ro, Seonwon-Gu, Cheongju, Chungbuk 28644, South Korea

1 Introduction The rapid development of ICT (Information & Communication Technology) has brought great changes to our daily life. With the emergence of internet, offline business is no longer the only alternative, online business transaction has become so popular these days. However, the emergence of smartphone make business transaction possible on mobile, as there is no time and space limitation. The recent emergence of O2O (Offline to Online) has resulted in the appearance of Fintech which provide the services of both online and offline payment. Fintech is the compound of finance and technology and is a new financial service based on ICT. In traditional payment cash, transfer process and credit card are the key factors, while payment services on mobile devices has become more convenient with the introduction of app card which makes credit card payment easier and real time transfer possible. The important thing is all the relevant parts have centered on financial institutions. However, Fintech has recently received much attention and mobile payment market is at the core of this industry. Therefore, non-financial institutions become necessary factors in today’s financial payment market in addition to financial institutions. Fintech will bring about explosive impact on consumers’ financial life and financial market. Different from the existing financial institutions, ICT enterprises develop convenient and diversified financial services based on internet technology to provide to the consumers, which are threatening the existing financial institutions. In case of Korea, Kakao Pay, Naver Pay, Samsung Pay, etc, are primary financial service of ICT enterprises, and Korean domestic financial institutions manage to enter into technological partnership with ICT enterprises to strengthen Fintech

123

G. Li et al.

industry. If non-financial institutions’ entering into the financial industry will lead to eating away the existing financial enterprises’ indigenous market share, this will threaten financial enterprises’ indigenous industry. There should be precautions to prevent financial accidents (voice phishing, sooping, pharming, etc) that are happening in Korea. According to the investigation results of domestic usage pattern of payment means [1], 72.3% of the individuals who don’t use internet payment worry about information leakage and privacy security. Different from the existing financial institutions, consumers become more anxious about security and accidents because online payment is the specialized bank that authenticate the users and exchange personal information online. Therefore, it is crucial to make thorough preparations to prevent security and accidents such as personal information leakage, illegal dealings, computer system paralysis, etc. This research use Fintech text ming skills to look at the trend of Fintech industry in Korea. We collected news dataJAVA (Jsoup) from Naver, the biggest portal site of Korea. R was used to analyze the big data to extract the nouns of news data that have been collected. The purpose of this research is to draw the direction and tasks for Korean Fintech industry taking into considerations of the importance of successful operation of Fintech and security. This research consists of following sections. Section 2 includes definitions, backgrounds and preceeding research in regard to Fintech and securities. In section 3, data collection and research procedures were introduced. In section 4, a comparison between results of Fintech security were made. Section 5 includes results of research and implications.

2 Research background 2.1 Fintech As the compound of finance and technology, Fintech is a general term for financial service and financial industry via the fusion of financial and ICT. Mobil, SNS, bigdata and other ICT technologies have been used to provide differentiated financial service. However, mobile simple payment service and mobile banking application card have recently become popular as innovative financial service. In addition, financial services such as payment settlement based on technology hold by innovative non-financial institutions are directly provided to users. Samsung pay, Naver pay, Apple pay, Kakao pay and Alipay are the representative examples of such financial services. NFC is the fundamental technology of Fintech and personalization is the most important feature of Fintech. Therefore, Fintech can provide location based O2O (online to offline)

123

financial service in accordance with individual’s behavior pattern. In addition, it is the core of Fintech which use various algorism methods such as statistics, machine learning and deep learning to conduct an analyzation. Therefore, it is possible to provide a 24 hours real time online and offline service [2]. Fintech industry area is composed of 4 parts [3]. 1. Remittance settlement: Customers benefits from the convenience of payment service which is easy to use and the low charge. 2. Financial data analysis: New added value is created by collecting and analyzing the diversified data which is related with individuals and enterprise customers. 3. Financial software: Using the more evolved smart technology, more efficient and innovative financial tasks and financial service related with S/W are provided than the traditional way. 4. Platform: Without the intervention of world enterprises and financial enterprises customers can freely transact and the transactions platform can be divided into four types. 2.2 Security Various form of attacks on IT infrastructure and computer networks have been serious problems to enterprises, and many of them making great efforts to prevent these attacks [4]. Smart card providers try to make transactions safety by storying data with security technology [5]. Security risk factors are affecting cloud service users in choosing service providers in future [6]. Protecting the essential or confidential data are becoming more an more important due to the important roles data plays in modern enterprise activities [7]. Fintech service has focused on the users’ convenience, therefore the use of Fintech is quite easy and convenient. However, because of problems such as account extortion such as CSRF (Cross-site request forgery), DDOS attack, Session hijacking, users’ privacy have been exposed to danger. According to the periodically published research concerning data leakage/infringement (BLI: Breach Level Index), there have been 237 accidents of personal information and business information leakage in the second quarter of 2014 all over the world [8]. This has lead to 1750 million customers records, individual information, financial information leakage to outside. Retail industry accounts for 83% as the biggest share. Less than 1% of the enterprises or institutions involved in the 237 accidents introduced access control system using data encryption and means of authentication. However, the equipment of authentication system will not guarantee thorough safety.

A study on the service and trend of Fintech security based on text-mining: focused on the data...

Interpark of Korea has equipped with Personal Information Management System in 2015, however, there have been over 10 million accidents of personal information leakage in 2016. In the context of Fintech, the security problems can be classified from the perspective of authentication, service and regulations. Taking advantage of the weakness of smartphone which has build on the open source platform, it is vulnerable to ID piracy, bypassing the additional authentication, phishing and pharming. Therefore, there is probability of attacking on the weakness of strangers or sociotechnology.

Table 1 Analytical Process Data collection

Object: Naver news Period: 2015.01.01.∼2016.07.31. Tool used: JAVA

Noun extraction

Nouns extracted are ranked in order of frequency and by years

Comparative analysis

Comparative analysis of Fintech news between 2015 and 2016 deduction of results and implication

Tool used: R

2.3 Preceding research Moon and Kim [9] suggest the additional security methods to be used such as FIDO (Fast IDentity Online) which is a convenient authentication method; block chain the best security technology of today used in bitcoin; FDS that detects the illegal dealings; and certificate verification that was once obligatorily used [9]. Park and Jin [10] suggest to establish the telecommunication protocols(machine registration, user authentication, payment settlement) for a secure transaction in the context of Fintech [10]. Park and Kim [11] analyzed the characteristics of Fintech service and potential security vulnerabilities and suggested counterplans to guarantee the safety of Fintech [11]. [12] carried out a comparative analysis on Fintech’s trend and payment service [12]. Choi [13] used abnormal detection and regression analysis- the basis of deep learning to check applicable technologies in all fields, then, presented solutions to Fintech securites such as abnormality trade detection, biometric authentication, phishing, pharming detection, identification and illegal use of others’ name etc [13]. Based on the top ten mobile risks published by Open Web Application Security Project (OWASP), [14] managed to find the solutions to 10 vulnerabilities in mobile context to apply to mobile Fintech applications [14]. Go et al. [15] cleaned up the standard of Fintech development and Fintech legislation status in both domestic and abroad based on the classification of device, server, authentification technology and communication technology etc., and also discussed the future trend of Fintech [15].

3 Research methodology 3.1 Analytical process This research conducted text mining analysis using the news data from Naver the Korean portal site that includes the

Fig. 1 Analytical procedure

contents of Fintech security (Table 1). Figure 1 shows the analytical procedure. First, Jsoup-the html parsing library was used for data collection, which is able to be operated in Java. Jsoup was imported in eclips and transformed the source of Naver news into news data. The data collected was analyzed with ‘R’the tool for bigdata analysis and then nouns were extracted. Using the visualization tool of Tagxedo, the procedure of visualization were conducted with the extracted nouns that have high relevance with Fintech security and service, whereas the nouns have low relevance were deleted.

3.2 Data collection and analysis The news data for this research use were collected by JAVA [16]. The collected data of Naver news used in this research includes 10,475 news from January 1st, 2015 through December 31st, 2015 and 4,452 news from January 1st, 2016 through July 31st, 2016. Fintech security was used as the keyword to collect corresponding news and news that referred to the keyword were extracted and the rest were not extracted. A total of 14,924 news were collected (Table 2). Volume of 2016 is less than 2015. This is because 12 mouths’ data were collected in 2015 while only 7 months data were collected in 2016. Additionally, the attention to Fintech was on the climax in 2015 and produced abundant news, while in 2016 the attentions to Fintech declined.

123

G. Li et al. Table 2 The scope of Naver news data calculation Classification

Contents

Channel

Naver news

Calculation conditions

‘Fintech security’ keyword title and body

Volume of calculation

2015

2016

Total

10,472

4,452

14,924

Calculation period

2015-01-01 00:00:00∼2016-07-31 23:59:59 (total 57 days)

Fig. 2 Volume of calculation (Fintech security)

This figure shows the amount of news that has been searched with Fintech security as the keyword from 1st quarter of 2015 through 3ed quarter of 2016 (Fig. 2).

4 Analytical results 4.1 Extraction of noun of news data Having deleted the overlapped news data collected from the Naver news, the frequency of each words have been calculated with R package [16,17]. Figure 3 shows the ranking of the nouns frequency from 1 to 20 during the period of 2015 and 2016. 7 months’ news were collected in 2016, therefore the total volume collected are less than 2015 as 12 months’ news were collected in 2015. The figure on the right side was the text file of the data collected. Only the nouns were extracted in the form of figure as seen on the right side after being analyzed with KONLP package in Rstudio. 4.2 2015 data In 2015, payment was found the most frequently referred keyword (Table 3). This is because Fintech has opened the mobile simple payment market and that has led to such results. Bank is the second most frequently referred keyword. It is the ICT

123

enterprises not the bank that take the lead and bank ranked the second as keyword. The platform ranked the third. Kakao pay, Naver pay, Samsung pay etc. have emerged and received much attention from the market. This is why platform has been so frequently referred. Service is the fourth most frequently referred noun. This is because many customers have paid great attention to this new service-mobile simple payment and regard it as the new rising star in the financial market. Fintech is the fifth most referred noun and it is followed by financial, technology, open, construct, domestic. Security was the twelfth most referred noun. Though security is the essential part of Fintech, it is not referred much as expected. This is because the most customers have paid much attention on Fintech the newly started service instead of security. This has led to the result of such deduction. Kakao pay ranked 19th as Korean’s first mobile simple payment service (Fig. 4). To sum up the results of keyword analysis of 2015, the Fintech service were at the beginning stage, therefore, the relevant keyword such as the existing functions, environment, service and so on have received more attention that led to higher ranks. While, the other keywords such as security, authentication and so on ranked low regardless their important roles. 4.3 2016 data The analytical results of 2016 are as follows (Table 4). Security is the most referred keyword. There have been many financial accidents in Korea due to users lack of security awareness regardless of having received both online and offline certification. The biggest difference of 2016 from 2015 is security. Though less news have been collected than 2015, security has been referred most implying security is the priority for better development of Fintech market. Enterprise appeared in the second position. Fintech area is not only open to financial institutions, it also opens to ICT enterprises. Kakao pay, Samsung pay, Naver pay and other existing relevant companies are entering into this area, which reflects that Fintech is running on the growth highway, which has led to such results. In addition, many financial institutions are entering into the Fintech industry area concerned about lagging behind in competition. They aligned with the existing ICT enterprises and financial institutions for further growth. For instance, KB financial expanded their alliance scope to 16 startups. And, Nonghyup bank is going to be the first provider of cloud service in financial section that support security service for Fintech startups on August, 2016. Start ranked 8th. A lot number of startups are entering into the Fintech industry in the worldwide. There is also a startup fever of

A study on the service and trend of Fintech security based on text-mining: focused on the data... Fig. 3 Extraction of noun (data cleaning)

Table 3 Result of 2015 data Ranks

Keyword

Frequency

1

Payment

21,229

2

Bank

19,059

3

Platform

18,448

4

Service

17,222

5

Fintech

16,174

6

Finance

14,402 11,795

7

Technology

8

Open

11,219

9

Construct

10,361

10

Domestic

9,366

11

Feasible

9,326

12

Security

9,017

13

Internet

8,773

14

Enterprise

8,580

15

Card

8,350

16

Pay

8,103

17

Metier

6,109

18

IT

5,963

19

Kakao

5,910

20

Authentication

5,593

Fig. 4 Result of 2015 Fintech security

4.4 The analysis of keyword network upbuilding the existing items or establishing new service platforms interlocking with government policies. P2P financial area is becoming the key issue (Fig. 5). To sum up the results of keyword analysis of 2016, security is the essential keyword in the stage of vitalization of Fintex service. ICT enterprises, government, financial institution are entering the market with diversified service which have made service, technology and operation appeared in the lower ranks.

As seen in Figs. 6 and 7 pay, security were found to have higher indegree compared with other keywords in accordance with the analytical results of 2015. Higher indegrees means a stronger power of pulling other keywords. In contrast, bank and platform have relatively low indegree. Payment, bank and platform were found to have higher outdegree and it means that a higher relevance compared with other keywords.

123

G. Li et al. Table 4 Result of 2016 data Ranks 1

Keyword

Frequency

Security

246,123

2

Enterprise

90,045

3

Support

90,045

4

Fintech

82,041

5

Center

70,035

6

Financial

66,033

7

Economy

46,023

8

Start

46,023

9

Create

46,023

10

Convergence

44,022

11

Industry

40,020

12

Cyber

38,019

13

Service

38,019

14

Technology

32,016 32,016

15

Market

16

Economy

26,013

17

Domestic

26,013

18

Britain

24,012

19

Operating

24,012

20

Global

20,010

Fig. 6 Keyword network of 2015 Fintech (indegree)

Fig. 7 Keyword network of 2015 Fintech (outdegree)

Fig. 5 Result of 2016 Fintech security

The key word market and technology have much higher indegree compared with other keywords in accordance with the analytical results of 2016 as seen in Figs. 8 and 9. In contrast, security and Fintech have relatively low indegree. Security, Fintech and Finance were found to have high outdegree.

123

Fig. 8 Keyword network of 2016 Fintech (intdegree)

5 Conclusions This research used crawling method collected news data concerning security during the period of and analyzed the keywords selected. The scope of news is 12 months of 2015

A study on the service and trend of Fintech security based on text-mining: focused on the data...

Emotional analysis, topic modeling analysis and other techniques that are part of text ming methods might be used for further research. Besides the Naver news, SNS data could also be crawled for analysis and which would reflect more diversified opinions and draw more significant results can be expected. There is also possibility to analyze the keyword by industry and topics in future. Acknowledgements The work described in this paper was supported by the project “the Foundation for Talents Fostering of Kunming University of Science and Technology” (Project Number: KKSY201408092).

Fig. 9 Keyword network of 2016 Fintech (outdegree)

References

when the Fintech is at the initial stage and 7 months of 2016 when Fintech enters the vitalization stage. To sum up the results of keyword analysis of 2015, the Fintech service were at the beginning stage, therefore, the relevant keyword such as the existing functions, environment, service and so on have received more attention that led to higher ranks. While, the other keywords such as security, authentication and so on ranked low regardless their important roles. To sum up the results of keyword analysis of 2016, security is the essential keyword in the stage of vitalization of Fintex service. ICT enterprises, government, financial institution are entering the market with diversified service which have made service, technology and operation appeared in the lower ranks. Recently, more and more existing financial institutions are doing their best to lead in Fintech market. However, security is the first priority and financial service suppliers need to establish security principles to guarantee stability of Fintech service. There is necessity of taking positive measures to prevent the attacks on personal information leaked, bypassing of authentication, phishing, pharming, and attacks on the weak connection between finance and ICT. This research analyzed the keyword concerning Fintech service and trend. Different from the previous research, the implication of this research is that this article collected news data and analyzed the trend of Korean Fintech security and service using the technology of text mining and visualization. This research will contribute to the switch of awareness of Fintech service and security by the market in today’s ongoing expanding of Fintech market. However, the limitation in this research is that news articles is the only source of data collection for analysis.

1. Bank of Korea: Payment using behavioral findings and implications. Financing payment report 2015-1. 2015 (2014) 2. Noh, K.S., Kim, E.C., Moon, Y.E., Park, S.T., Lee, S.H., Lim, K.H., Jung, K.H.: E-commerce. Life and Power Press, Gyeonggi (2015) 3. UK Trade & Investment: Fintech the UK’s unique environment for growth. Business is Great (2014) 4. Luh, R., Marschalek, S., Kaiser, M., Janicke, H., Schrittwieser, S.: Semantics-aware detection of targeted attacks: a survey. J. Comput. Virol. Hacking Tech. (2016). doi:10.1007/s11416-016-0273-3 5. Bouffard, G., Lanet, J.: Reversing the operating system of a Java based smart card. J. Comput. Virol. Hacking Tech. 10(4), 239–253 (2014) 6. Kim, D.Y., Li, G., Park, S.T., Ko, M.H.: A study on effects of security risks on acceptance of enterprise cloud service: moderating of employment and non-employment using PLS multiple group analysis. J. Comput. Virol. Hacking Tech. 12(3), 151–161 (2016) 7. David, B., Larget, D., Scherrer, T.: The security of databases: the Access case. J. Comput. Virol. Hacking Tech. 9(2), 95–107 (2013) 8. Breach Level Index: http://breachlevelindex.com 9. Moon, S.T., Kim, K.N.: An analysis of Fintech service technology and security trend. Rev. Korean Soc. Internet Inf. 16(2), 23–32 (2015) 10. Park, J.O., Jin, B.W.: A study on authentication method for secure payment in Fintech environment. J. Inst. Internet Broadcast Commun. 15(4), 25–31 (2015) 11. Park, J.K., Kim, I.J.: Countermeasure and security vulnerability of Fintech services. Korea Inf. Process. Soc. Rev. 22(5), 36–45 (2015) 12. Ahn, J.K., Yoon, Y.J., Ahn, E.H., Kim, H.U.: A comparative study of domestic Fintech trend and pay service using social media mining: focused on Kakao pay and Samsung pay the Korea Society of Management Information System. In: Collection of Conference Dissertations (2016) 13. Choi, D.S.: AI and Fintech security. Rev. KIISC 26(2), 35–38 (2016) 14. Cho, J.S., Choi, H.N., Kim, S.Y., Kim, H.S.: Security analysis on mobile finance apps. Commun. Korean Inst. Inf. Sci. Eng. 34(4), 34–41 (2016) 15. Go, W., Oh, S.T., Lee, T.J.: Fintech security standardization trend. Commun. Korean Inst. Inf. Sci. Eng. 34(4), 13–19 (2016) 16. Park, E.M., Seo, J.H., Ko, M.H.: The effects of leadership by types of soccer instruction on big data analysis. Cluster Comput. (2016). doi:10.1007/s10586-016-0609-2 17. Ko, M.H., Kim, B.S., Kim, D.Y., Park, S.T.: Comparison between transition year in Irelans versus self-direction learning semester in Korea through online news data analysis. Indian J. Sci. Technol. 9(24), 1–7 (2016)

123