• Journal of Digital Convergence
• /
• v.17 no.7
• /
• pp.285-292
• /
• 2019

With the popularization of PC, SNS and IoT, a lot of data is generated and the amount is increasing exponentially. Artificial neural network learning is a topic that attracts attention in many fields in recent years by using huge amounts of data. Artificial neural network learning has shown tremendous potential in speech recognition and image recognition, and is widely applied to a variety of complex areas such as medical diagnosis, artificial intelligence games, and face recognition. The results of artificial neural networks are accurate enough to surpass real human beings. Despite these many advantages, privacy problems still exist in artificial neural network learning. Learning data for artificial neural network learning includes various information including personal sensitive information, so that privacy can be exposed due to malicious attackers. There is a privacy risk that occurs when an attacker interferes with learning and degrades learning or attacks a model that has completed learning. In this paper, we analyze the attack method of the recently proposed neural network model and its privacy protection method.

• ETRI Journal
• /
• v.33 no.6
• /
• pp.935-944
• /
• 2011

As a growing number of individuals are exposed to surveillance cameras, the need to prevent captured videos from being used inappropriately has increased. Privacy-related information can be protected through video encryption during transmission or storage, and several algorithms have been proposed for such purposes. However, the simple way of evaluating the security by counting the number of brute-force trials is not proper for measuring the security of video encryption algorithms, considering that attackers can devise specially crafted attacks for specific purposes by exploiting the characteristics of the target video codec. In this paper, we introduce a new attack for recovering contour information from encrypted H.264 video. The attack can thus be used to extract face outlines for the purpose of personal identification. We analyze the security of previous video encryption schemes against the proposed attack and show that the security of these schemes is lower than expected in terms of privacy protection. To enhance security, an advanced block shuffling method is proposed, an analysis of which shows that it is more secure than the previous method and can be an improvement against the proposed attack.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.119-130
• /
• 2022

Crime is a common social problem that affects the quality of life. As the number of crimes increases, it is necessary to build a model to predict the number of crimes that may occur in a given period, identify the characteristics of a person who may commit a particular crime, and identify places where a particular crime may occur. Data privacy is the main challenge that organizations face when building this type of predictive models. Federated learning (FL) is a promising approach that overcomes data security and privacy challenges, as it enables organizations to build a machine learning model based on distributed datasets without sharing raw data or violating data privacy. In this paper, a federated long short- term memory (LSTM) model is proposed and compared with a traditional LSTM model. Proposed model is developed using TensorFlow Federated (TFF) and the Keras API to predict the number of crimes. The proposed model is applied on the Boston crime dataset. The proposed model's parameters are fine tuned to obtain minimum loss and maximum accuracy. The proposed federated LSTM model is compared with the traditional LSTM model and found that the federated LSTM model achieved lower loss, better accuracy, and higher training time than the traditional LSTM model.

• Journal of Families and Better Life
• /
• v.23 no.4 s.76
• /
• pp.67-77
• /
• 2005

The purpose of this study was to identify housing attributes that make residents feel 'at home' in ALFs in Southwest Virginia. For this purpose, residents' needs, experiences, and opinions of the physical environment, the social environment and the organizational environments such as policies and programs of ALFs were identified. As a multi-case study, five ALFs in Southwest Virginia were studied using constant comparative methods of data analysis. In addition to face-to-face interviews with 25 residents and 5 administrators of five ALFs, observations were conducted with personal journal. Each facility was designed to be a single-family house or multi-family dwelling in outside appearance. Most of the respondents were satisfied with their current dwelling as a 'home' in terms of homelike attributes such as 'autonomy/ privacy', personalization,' safety and security,' services and care,' independence,' social interaction/friendship,' family support,' and 'rules and regulations.' In spite of high satisfaction with the facility, however, many people did not think of their current dwelling as a real 'home'. As the biggest difference between living in their own homes and living in the ALF, people feinted out a lack of independence and social interaction. Residents of ALFs may have reordered their priorities in their current life situation so that safety, security, and caie were more important to them than fooling "at home."

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.1
• /
• pp.63-74
• /
• 2023

In order to ensure safety to invasion of privacy, Federated Learning(FL) that learns using parameters is emerging. However a paper that leaks training data using gradients was recently published. Our paper implements an experiment to leak training data using gradients in a federated learning environment, and proposes a method to improve reconstruction performance by improving existing attacks that leak training data. Experiments using Yale face database B, MNIST dataset on the proposed method show that federated learning is not safe from invasion of privacy by reconstructing up to 100 data out of 100 training data when performance of federated learning is high at accuracy=99~100%. In addition, by comparing the performance (MSE, PSNR, SSIM) of pixels and the performance of identification by Human Test, we want to emphasize the importance of the performance of identification rather than the performance of pixels.

• Information Systems Review
• /
• v.25 no.4
• /
• pp.161-181
• /
• 2023

As the post-COVID-19 consumer life environment is rapidly becoming non-face-to-face, changing non-face-to-face financial life services are having a significant impact on consumers' daily lives. People who do not have access to digital devices and services that have become essential goods are at risk of being left behind in the "digital blind spot," where they are marginalized not only in their daily lives but also in society and the economy as a whole (Kim Min-Jeung A, Kim Min-Jung B, Park Joo-Yung, 2022). In this study, we examined the effects of perceived control factors, Cognitive control, behavioral control, and decisional control, on intention to use digital finance. For this study, we surveyed 133 customers who are aware of and intend to use digital finance. The results show that cognitive control, behavioral control, and decisional control have significant effects on intention to use digital finance. In this relationship, the moderating effect of privacy concerns differs from the effect of decision control on intention to use digital finance. These findings suggest that digital financial services firms should consider whether to reduce or increase customer control. Based on these findings, we discuss marketing strategies and implications for digital financial services companies.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.331-332
• /
• 2006

To enhance security and privacy in biometrics, changeable (or cancelable) biometrics have recently been introduced. The idea is to transform a biometric signal or feature into a new one for enrollment and matching. In this paper, we proposed changeable biometrics for face recognition using on PCA based approach. PCA coefficient vector extracted from an input face image. The vector is scrambled randomly and removed. When a transformed template is compromised, it is replaced by a new scrambling rule. In our experiment, we compared the performance between when PCA coefficient vectors are used for verification and when the transformed coefficient vectors are used for verification.

• Journal of Intelligence and Information Systems
• /
• v.10 no.1
• /
• pp.65-75
• /
• 2004

It is notorious that PIN(Personal Identification Number) is used widely for user verification and authentication in networked environment. But, when the user Identification and password are exposed by hacking, we can be damaged monetary damage as well as invasion of privacy. In this paper, we adopt face recognition-based authentication which have nothing to worry what the ID and password will be exposed. Also, we suggest the remote authentication and verification system by considering not only 2-Tier system but also 3-Tier system getting be distributed. In this research, we analyze the face feature data using the SVM(Support Vector Machine) and PCA(Principle Component Analysis), and implement artificial intelligence face recognition module in distributed environment which increase the authentication speed and heightens accuracy by utilizing artificial intelligence techniques.

• Journal of Digital Convergence
• /
• v.11 no.4
• /
• pp.57-69
• /
• 2013

Progresses in ICT make the processing and exchange of personal data across international borders often necessary and relatively easy. The challenge lies in protecting fundamental rights and freedoms of individuals, notably the right to privacy and the right to personal information, while encouraging the free and secure flow of information across borders for the continued expansion of online transactions. The key to establishing a functioning international solution for personal data protection is to strike a right balance between the two camps which currently dominate the debate - the advocates of individual privacy rights on one side exemplified by the EU, and the proponents of self-regulation and economic efficiency on the other, represented by the U.S. In the face of a growing tension between the two sides each equipped with their own ideals, a practical solution may lie in utilizing established institutions of standardization such as ISO and IEC as a ground upon which an agreement can take its root.

• Korean Institute of Interior Design Journal
• /
• v.26 no.4
• /
• pp.84-92
• /
• 2017

Just as the front gate is located at the meeting point between the house and the street, the Subject and the Other face each other the same way. This study examines the relationship between House(subject) and Stree (other) at the boundary of the Front Gate-Face. Pursuing the aspects of the changing Front Gate-Face accordingly to the attitude of the Subject facing the Other, this study tries to analyze the possibilities and significance of the hospitality Front Gate-Face with the ethical point of view of Levinas. As architectural instance, results of examining the Front Gate-Face of traditional houses in Korea, China and Japan are as follows. Front Gate-Face of China is characterized by self-centered introversion to interact with the external world (the other). Front Gate-Face of Japan is characterized by a humble submission to the group. Front Gate-Face of Korea shows however more flexible relationship orientations in terms of hospitality, compared to Japan or China. When looking through hospitality factors, accordingly to the above mentioned Korean hospitality characteristics, the possibilities seem not be exclusively bordered inside the conceptual category perimeter suggested by Levinas' concept of hospitality. It is almost impossible for the nowadays ever-strong privacy culture to not allow room for the architectural structure of an absolute hospitality toward others. However, this impossibility not being absolute, still yields a space for a significant possibility to explore.