• 제목/요약/키워드: security technology

검색결과 6,825건 처리시간 0.033초

LIME과 SHAP 모델 공유에 의한 모델 해석 (Model Interpretation through LIME and SHAP Model Sharing)

  • 김용길
    • 한국인터넷방송통신학회논문지
    • /
    • 제24권2호
    • /
    • pp.177-184
    • /
    • 2024
  • 데이터가 빠른 속도로 증가하고 있는 가운데 가능한 최고의 정확도를 달성하기 위해 모든 종류의 복잡한 앙상블 및 딥 러닝 알고리즘이 사용되고 있다. 그렇지만, 이러한 모델이 알 수 없는 데이터를 예측/분류/인식/추적하는 방법과 관련하여 예측, 분류, 인식, 추적이 항상 신뢰할 수 있는 것은 아니다. 데이터 부족, 불균형 데이터 세트, 편향된 데이터 세트 등과 같은 다양한 이유가 학습 모델에 의해 포착되는 결정에 영향을 미칠 수 있다. 이와 관련하여 현재 모델의 설명 가능성에 관한 연구가 관심을 끌고 있다. 현재 설명 가능성 기법과 관련하여 LIME과 SHAP가 보편적으로 사용되고 있지만, 출력 결과들은 다소 상이한 측면을 나타내고 있다. 이에 본 연구에서는 LIME과 SHAP을 결합하는 방식을 소개하고, 데모와 관련해서 IEEE CIS 데이터 세트에서 거래를 사기로 분류할 때 LightGBM 및 Keras 모델이 내린 결정에 대한 설명 가능성을 분석한다.

The impact of modern airport security protocols on patients with total shoulder replacements

  • Michael D. Scheidt;Neal Sethi;Matthew Ballard;Michael Wesolowski;Dane Salazar;Nickolas Garbis
    • Clinics in Shoulder and Elbow
    • /
    • 제26권4호
    • /
    • pp.416-422
    • /
    • 2023
  • Background: Advancements in airport screening measures in response to 9/11 have resulted in increased false alarm rates for patients with orthopedic and metal implants. With the implementation of millimeter-wave scanning technology, it is important to assess the changes in airport screening experiences of patients who underwent total shoulder arthroplasty (TSA). Methods: Here, 197 patients with prior anatomic and reverse TSA completed between 2013 and 2020 responded to a questionnaire regarding their experiences with airport travel screening after their operation. Of these patients, 86 (44%) stated that they had traveled by plane, while 111 (56%) had not. The questionnaire addressed several measures including the number of domestic and international flights following the operation, number of false alarm screenings by the millimeter-wave scanner, patient body habitus, and presence of additional metal implants. Results: A total of 53 patients (62%) responded "yes" to false screening alarms due to shoulder arthroplasty. The odds of a false screening alarm for patients with other metal implants was 5.87 times that of a false screening alarm for patients with no other metal implants (P<0.1). Of a reported 662 flights, 303 (45.8%) resulted in false screening alarms. Greater body mass index was not significantly lower in patients who experienced false screening alarms (P=0.30). Conclusions: Patients with anatomic and reverse TSA trigger false alarms with millimeter-wave scanners during airport screening at rates consistent with prior reports following 9/11. Patient education on the possibility of false alarms during airport screening is important until improvements in implant identification are made. Level of evidence: IV.

Network Anomaly Traffic Detection Using WGAN-CNN-BiLSTM in Big Data Cloud-Edge Collaborative Computing Environment

  • Yue Wang
    • Journal of Information Processing Systems
    • /
    • 제20권3호
    • /
    • pp.375-390
    • /
    • 2024
  • Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.

Lignocellulolytic Enzymes Production by Four Wild Filamentous Fungi for Olive Stones Valorization: Comparing Three Fermentation Regimens

  • Soukaina Arif;Hasna Nait M'Barek;Boris Bekaert;Mohamed Ben Aziz;Mohammed Diouri;Geert Haesaert;Hassan Hajjaj
    • Journal of Microbiology and Biotechnology
    • /
    • 제34권5호
    • /
    • pp.1017-1028
    • /
    • 2024
  • Lignocellulolytic enzymes play a crucial role in efficiently converting lignocellulose into valuable platform molecules in various industries. However, they are limited by their production yields, costs, and stability. Consequently, their production by producers adapted to local environments and the choice of low-cost raw materials can address these limitations. Due to the large amounts of olive stones (OS) generated in Morocco which are still undervalued, Penicillium crustosum, Fusarium nygamai, Trichoderma capillare, and Aspergillus calidoustus, are cultivated under different fermentation techniques using this by-product as a local lignocellulosic substrate. Based on a multilevel factorial design, their potential to produce lignocellulolytic enzymes during 15 days of dark incubation was evaluated. The results revealed that P. crustosum expressed a maximum total cellulase activity of 10.9 IU/ml under sequential fermentation (SF) and 3.6 IU/ml of β-glucosidase activity under submerged fermentation (SmF). F. nygamai recorded the best laccase activity of 9 IU/ml under solid-state fermentation (SSF). Unlike T. capillare, SF was the inducive culture for the former activity with 7.6 IU/ml. A. calidoustus produced, respectively, 1,009 ㎍/ml of proteins and 11.5 IU/ml of endoglucanase activity as the best results achieved. Optimum cellulase production took place after the 5th day under SF, while ligninases occurred between the 9th and the 11th days under SSF. This study reports for the first time the lignocellulolytic activities of F. nygamai and A. calidoustus. Furthermore, it underlines the potential of the four fungi as biomass decomposers for environmentally-friendly applications, emphasizing the efficiency of OS as an inducing substrate for enzyme production.

The Effect of Synbiotic Supplementation on Bone Complications, Anemia, and Gastrointestinal Function in Hemodialysis Patients: A Double-Blind Randomized Clinical Trial

  • Yasaman Azamian;Hadi Abdollahzad;Shahab Rezaeian;Mohammad Hossein Rouhani;Mohammad Hossein Fatehi
    • Clinical Nutrition Research
    • /
    • 제13권4호
    • /
    • pp.272-283
    • /
    • 2024
  • Probiotics affect biomarkers indicative of bone formation, such as alkaline phosphatase (ALP), calcium status, bone mineralization, bone turnover markers and metabolism. This study aims to investigate the effects of synbiotic on gastrointestinal (GI) disorder, bone complications and anemia in hemodialysis (HD) patients. In this randomized, double-blind, placebo-controlled clinical trial study, HD patients received 2 symbiotic (n = 19) or placebo (n = 17) capsules daily for 12 weeks. GI function, serum levels of bone-specific biomarkers, and serum levels of anemia-specific biomarkers were assessed at the beginning and the end of study. GI function was assessed with gastrointestinal symptom rating scale questionnaire. The data were analyzed using SPSS. At the end of this study, parathyroid hormone levels decreased significantly in the synbiotic group (p = 0.039); however, in comparison to placebo group, the difference was not significant. Decrease of ALP levels in the synbiotic group were not statistically significant. However, a significant difference was seen between the 2 groups at the end of intervention (p = 0.037). Improvement in GI symptoms was observed in both groups, but the reduction rate was higher in the synbiotic group. Additionally, at the end of the study, a significant difference between the 2 groups was observed (p < 0.05). No statistically significant difference was observed in the levels of other factors within each group and between the 2 groups (p > 0.05). Symbiotic supplements after 12 weeks led to an improvement in GI function and ALP levels in HD patients. Further investigation into bone-mineral disorders in HD patients is necessary.

다양한 산업에서의 키 관리 시스템 비교 분석 (Comparison of key management systems across different industries)

  • 권우주;장항배
    • Journal of Platform Technology
    • /
    • 제12권3호
    • /
    • pp.55-61
    • /
    • 2024
  • 디지털 환경이 복잡해지고 사이버 공격이 정교해지면서 데이터 보호에 대한 중요성이 부각되고 있다. 데이터 유출, 시스템 침입, 인증 우회와 같은 다양한 보안 위협이 증가하면서 안전한 키 관리가 대두되고 있다. 키 관리 시스템(Key Management System, KMS)은 암호 키 생명주기 절차 전체를 관리하며, 여러 산업에서 사용되고 있다. 공공, 금융을 포함하는 다양한 산업의 환경에 맞는 요구 사항을 고려한 키 관리 시스템이 필요한 상황이다. 본 논문은 대표적인 산업에서 사용하는 키 관리 시스템을 비교 분석하여, 산업별 키 관리 시스템의 특성을 도출하는 것이 목적이다. 연구 방법은 문헌 및 기술 문서 분석, 사례 분석으로 정보를 수집하여, 산업 분야별 비교 분석을 진행하였다. 본 논문의 결과는 산업 환경에 맞는 키 관리 시스템을 도입하거나 개발 시 실질적인 가이드를 제공할 수 있을 것이다. 한계점은 분석한 산업 분야가 부족하고, 실험적 검증이 부족하였다. 이에 향후 연구에서 다양한 분야의 키 관리 시스템을 포함하여, 실험을 통한 구체적인 성능 테스트를 진행하고자 한다.

  • PDF

DRM기반 의료취약지 진료만족도 조사 연구 (A study on Satisfaction with Treatment in Medically vulnerable areas based on DRM)

  • 이경화
    • 보건의료생명과학 논문지
    • /
    • 제9권2호
    • /
    • pp.213-220
    • /
    • 2021
  • 의료취약지에 거주하는 농촌 주민들에게 적절한 의료서비스를 제공하고, 이를 통한 보건의료자원의 불균형을 해소하기 위한 정책이 필요하다. 일반적으로 해당 주민들에게 의료서비스 확보를 위하여 설문 방법을 활용하며, 조사자와 참여자간의 설문문형과 개인정보를 디지털 컨텐츠로 배포 및 수집하기 위해 보안 정책이 필요하다. 제안한 논문에서는 조사자의 설문문형과 참여자의 개인정보를 안전하게 관리하기 위한 방안으로 DRM(Digital Rights Management)기술을 도입하며, 기존의 License Server가 모든 도메인들에게 라이센스를 발행, 갱신, 취소하는 문제점을 개선하기 위하여 패밀리도메인으로 묶은 DM만이 인증서를 발급받은 후 라이센스 관리를 수행하는 DRM기반 설문조사 정책에 편리성을 제공한다.

Titanium alloys: A closer-look at mechanical, gamma-ray, neutron, and transmission properties of different grade alloys through MCNPcode application

  • Ghada ALMisned;Omer Guler;Duygu Sen Baykal;G. Kilic;H.O. Tekin
    • Nuclear Engineering and Technology
    • /
    • 제56권9호
    • /
    • pp.3501-3511
    • /
    • 2024
  • Titanium alloys play a vital role in optimizing the effectiveness and security of nuclear reactors, strengthening structural durability, and facilitating the effective handling of nuclear waste. The aim of this study is to investigate the gamma-ray, neutron, and transmission properties of four common titanium alloys through the examination of the deposited energy amount in the liquid sodium coolant material, in relation to the mechanical properties of these alloys. MCNP (version 6.3) is utilized for designing the titanium pipes. Next, the pipes were re-designed considering the elemental mass fractions and densities of the investigated titanium alloys. Grade 26 sample is reported with the highest values of mass attenuation coefficients and the lowest HVL values among those investigated alloys. Grade 26 is reported to have the lowest TF value, whereas Grade 12 demonstrated the highest TF value. The highest Effective Removal Cross Section (ΣR, 1/cm) value against fast neutrons is reported for Grade 26. The utilization of Grade 26 sample as pipe material resulted in the lowest deposited energy amount (MeV/g) and subsequent lowest contamination in the coolant material. Out of the alloys that were chosen for analysis, it has been determined that Grade 26 exhibits the highest level of strength. It can be concluded that the Grade 26 alloy exhibits desirable characteristics for applications in nuclear technologies that require superior gamma-ray and neutron absorption properties, as well as exceptional mechanical properties. Nevertheless, it is essential to emphasize the importance for ongoing studies to enhance the existing material properties of Grade 26, with the aim of achieving improved safety and efficacy in nuclear applications.

IoT 기반 유해 가스 환경 제어 시스템 개발 (Development of IoT-based Hazardous Gas Environment Control System )

  • 김철훈;류대현;최태완
    • 한국전자통신학회논문지
    • /
    • 제19권5호
    • /
    • pp.1013-1018
    • /
    • 2024
  • 본 연구에서는 산업 현장에서의 유해 가스로 인한 재해를 예방하기 위해 IoT(Internet of Things) 기술을 활용한 실시간 모니터링 시스템을 개발하고 그 성능을 평가하였다. 개발된 시스템은 유해 가스를 실시간으로 감지하고 신속한 경보를 발령하여 98% 이상의 데이터 정확도와 3초 미만의 반응 시간을 달성하였다. 시스템은 센서 노드, 중앙 처리 장치, 사용자 인터페이스로 구성되며, 클라우드 기반 원격 감시 및 관제 프로그램을 통해 작업 환경과 작업자 상태를 실시간으로 모니터링한다. 본 연구는 산업 현장의 안전 관리를 효과적으로 수행할 수 있는 새로운 접근 방식을 제시하였으며, 향후 다중 가스 감지 기능의 개선, AI 기반의 예측 모델 개발, 보안 강화 등을 통해 더욱 발전된 모니터링 시스템으로 진화할 것으로 기대된다.

AI와 윤리: 문헌의 종합적 분석과 정보시스템 분야의 향후 연구 방향 (Navigating Ethical AI: A Comprehensive Analysis of Literature and Future Directions in Information Systems)

  • 민진영
    • 지식경영연구
    • /
    • 제25권3호
    • /
    • pp.1-22
    • /
    • 2024
  • AI의 사용이 일상 생활의 많은 부분에서 현실화 되어감에 따라 AI가 가져오는 긍정적인 기회와 혜택이 주목 받는 한편, AI가 초래할 수 있는 윤리적 문제들에 대한 염려도 커지고 있다. 정보시스템 분야는 기술이 비즈니스와 사회에 미치는 영향을 연구하는 분야로서 AI가 인류 사회에 바람직한 영향을 미칠 수 있도록 기여해야 한다. 따라서 AI와 윤리 관련한 다양한 연구들을 살펴보고 정보시스템 분야의 연구가 나아가야 할 방향을 탐색할 필요가 있다. 본 연구는 이를 위해 먼저 2020년부터 현재까지의 문헌을 수집하여 연구자의 코딩과 토픽 모델링을 통해 연구 주제를 범주화 하였다. 분석 결과 AI 윤리 원칙, 윤리적 AI 디자인 및 개발, 윤리적 AI 도입 및 활용, 윤리적 AI 사용의 네 가지로 연구 주제를 범주화하고, 각 범주 별로 문헌을 고찰하여 연구 현황을 짚은 후, 정보 시스템 분야에서의 AI 윤리에 대한 향후 연구 방향을 제언하였다.