• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.133-144
• /
• 2020

고령화로 인해 증가하는 노인 비율만큼이나 치매를 앓는 노인 수 또한 빠르게 늘고 있는데 이는 사회적, 경제적 부담을 발생시킨다. 특히, 간병인의 근무 시간 손실 및 간호 부담으로 인한 의료 비용 증가와 같은 간접비용을 포함하는 치매 관리 비용은 수년에 걸쳐 기하급수적으로 증가하고 있다. 이러한 비용을 줄이기 위해 치매 환자를 돌보기 위한 관리 시스템 도입이 시급하다. 따라서 본 연구는 항상 치매 환자를 돌볼 수 없는 환경이나 독거노인을 관리하기 위한 센서 기반 이상 행동 탐지 시스템을 제안한다. 기존 연구들은 단지 행동을 인지하거나 정상 행동 여부를 평가하는 정도였고 센서로부터 받은 데이터가 아닌 이미지를 처리하여 행동을 인지한 연구도 있었다. 본 연구에서는 실데이터 수집에 한계가 있음을 인지하여 비지도 학습 모델인 오토인코더와 지도 학습 모델인 장·단기 기억 모형을 동시에 사용했다. 비지도 학습 모델인 오토인코더는 정상 행동 데이터를 학습하여 정상적인 행동에 대한 패턴을 학습시켰고 장·단기 기억 모형은 센서로 인지 가능한 행동을 학습시켜 분류를 좀 더 세분화했다. 테스트 결과 각각의 모델은 약 96%, 98% 이상의 정확도를 도출하였고 오토인코더의 이상치가 3% 이상을 갖는 경우 장·단기 기억 모형을 통과하도록 설계했다. 이 시스템을 통해 혼자 사는 노인이나 치매 환자를 효율적으로 관리할 수 있으며 돌보기 위한 비용 또한 절감할 수 있을 것으로 전망된다.

• 한국인공지능학회지
• /
• 제11권2호
• /
• pp.19-27
• /
• 2023

In recent times, an exponential increase in Internet traffic has been observed as a result of advancing development of the Internet of Things, mobile networks with sensors, and communication functions within various devices. Further, the COVID-19 pandemic has inevitably led to an explosion of social network traffic. Within this context, considerable attention has been drawn to research on network traffic analysis based on machine learning. In this paper, we design and develop a new machine learning framework for network traffic analysis whereby normal and abnormal traffic is distinguished from one another. To achieve this, we combine together well-known machine learning algorithms and network traffic analysis techniques. Using one of the most widely used datasets KDD CUP'99 in the Weka and Apache Spark environments, we compare and investigate results obtained from time series type analysis of various aspects including malicious codes, feature extraction, data formalization, network traffic measurement tool implementation. Experimental analysis showed that while both the logistic regression and the support vector machine algorithm were excellent for performance evaluation, among these, the logistic regression algorithm performs better. The quantitative analysis results of our proposed machine learning framework show that this approach is reliable and practical, and the performance of the proposed system and another paper is compared and analyzed. In addition, we determined that the framework developed in the Apache Spark environment exhibits a much faster processing speed in the Spark environment than in Weka as there are more datasets used to create and classify machine learning models.

• 한국정보통신학회논문지
• /
• 제23권8호
• /
• pp.903-909
• /
• 2019

치주질환은 상당수의 성인들이 가지고 있는 질환이지만 아직 분자적인 수준에서의 발생 기작과 치료 방법에 대해서는 많은 것이 밝혀져 있지 않다. 본 연구에서는 치주질환 조직과 정상 조직에서 얻어진 유전자 발현 데이터를 이용하여 치주질환 조직과 정상 조직 사이에 분자적 차이가 있는지를 확인한다. 특히 기계학습 알고리즘을 이용하여 유전자 발현양 기반 치주질환 조직과 정상 조직의 분류가 가능한지를 확인하고, 각 조직에서 발현양 차이가 나는 유전자들이 주로 어떤 기능을 하는 것인지 살펴본다. t-SNE를 이용한 분석 결과 정상 조직과 치주질환 조직 샘플이 명확히 구분되어 군집화 될 수 있음이 확인되었다. 또한, 결정 트리, 랜덤 포레스트, 서포트 벡터 머신을 이용한 분류 알고리즘을 적용한 결과 불균형 데이터임에도 높은 정확도와 민감도, 특이도를 보였으며, 염증 반응 및 면역 반응 관련 유전자들이 주로 두 집단 간에 차이를 보임이 확인되었다.

• Fisheries and Aquatic Sciences
• /
• 제25권3호
• /
• pp.151-157
• /
• 2022

Various approaches have been applied to transform aquaculture from a manual, labour-intensive industry to one dependent on automation technologies in the era of the fourth industrial revolution. Technologies associated with the monitoring of physical condition have successfully been applied in most aquafarm facilities; however, real-time biological monitoring systems that can observe fish condition and behaviour are still required. In this study, we used a video recorder placed on top of a fish tank to observe the swimming patterns of rock bream (Oplegnathus fasciatus), first one fish alone and then a group of five fish. Rock bream in the video samples were successfully identified using the you-only-look-once v3 algorithm, which is based on the Darknet-53 convolutional neural network. In addition to recordings of swimming behaviour under normal conditions, the swimming patterns of fish under abnormal conditions were recorded on adding an anaesthetic or lowering the salinity. The abnormal conditions led to changes in the velocity of movement (3.8 ± 0.6 cm/s) involving an initial rapid increase in speed (up to 16.5 ± 3.0 cm/s, upon 2-phenoxyethanol treatment) before the fish stopped moving, as well as changing from swimming upright to dying lying on their sides. Machine learning was applied to datasets consisting of normal or abnormal behaviour patterns, to evaluate the fish behaviour. The proposed algorithm showed a high accuracy (98.1%) in discriminating normal and abnormal rock bream behaviour. We conclude that artificial intelligence-based detection of abnormal behaviour can be applied to develop an automatic bio-management system for use in the aquaculture industry.

• Tuberculosis and Respiratory Diseases
• /
• 제86권3호
• /
• pp.226-233
• /
• 2023

Background: Inactive or old, healed tuberculosis (TB) on chest radiograph (CR) is often found in high TB incidence countries, and to avoid unnecessary evaluation and medication, differentiation from active TB is important. This study develops a deep learning (DL) model to estimate activity in a single chest radiographic analysis. Methods: A total of 3,824 active TB CRs from 511 individuals and 2,277 inactive TB CRs from 558 individuals were retrospectively collected. A pretrained convolutional neural network was fine-tuned to classify active and inactive TB. The model was pretrained with 8,964 pneumonia and 8,525 normal cases from the National Institute of Health (NIH) dataset. During the pretraining phase, the DL model learns the following tasks: pneumonia vs. normal, pneumonia vs. active TB, and active TB vs. normal. The performance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to determine active TB by DL model and radiologists. Sensitivities and specificities for determining active TB were evaluated for both the DL model and radiologists. Results: The performance of the DL model showed area under the curve (AUC) values of 0.980 in internal validation, and 0.815 and 0.887 in external validation. The AUC values for the DL model, thoracic radiologist, and general radiologist, evaluated using one of the external validation datasets, were 0.815, 0.871, and 0.811, respectively. Conclusion: This DL-based algorithm showed potential as an effective diagnostic tool to identify TB activity, and could be useful for the follow-up of patients with inactive TB in high TB burden countries.

• 한국인지과학회:학술대회논문집
• /
• 한국인지과학회 2000년도 한글 및 한국어 정보처리
• /
• pp.133-137
• /
• 2000

구문태그가 부착되지 않은 코퍼스를 사용하여 문법규칙의 확률을 훈련하는 비통제 학습(unsupervised learning) 방법의 대표적인 것이 CNF(Chomsky Normal Form)의 CFG(Context Free Grammar)를 입력으로 하는 inside-outside 알고리즘이다. 본 연구에서는 의존문법을 CNF로 변환하는 기법에 대해 논하고 의존문법을 위해 변형된 inside-outside 알고리즘을 논한다. 또한 이 알고리즘을 사용하여 실제 훈련한 결과를 보이고, 의존규칙과 구문구조 확률을 같이 사용하는 hybrid방식 구문분석기에 적용한 결과를 보인다.

• 디지털산업정보학회논문지
• /
• 제8권4호
• /
• pp.35-41
• /
• 2012

Recently, universities have difficulties in operating the normal curriculum because fresher's basic academic ability is declined. It causes campus misfits so managing students is also not an easy matter. The education system that focuses only on college entrance exams is one of the reasons why this phenomenon occurred. Activity with self-directed Learning Community to know learning level themselves and execute systematic studying habit is essential for improving this problem. This activity can help students understanding and having interest in class and be motivated to study. But it had burdened tutors with submitting activity report in written form. In this paper, we suggest the Mobile Based Activity Report Submission System which can be the solution of the problem that the Self-directed Learning Community System has. This system reduces the emotional burden to write the reports and manages them efficiently.

• 디지털산업정보학회논문지
• /
• 제17권3호
• /
• pp.9-15
• /
• 2021

In this paper, the Embedded MEMS system to the power apparatus used Edge Impulse machine learning tools and therefore an improved predictive system design is implemented. The proposed MEMS embedded system is developed based on nRF52840 system and the sensor with 3-Axis Digital Magnetometer, I2C interface and magnetic measurable range ±120 uT, BM1422AGMV which incorporates magneto impedance elements to detect magnetic field and the ARM M4 32-bit processor controller circuit in a small package. The MEMS embedded platform is consisted with Edge Impulse Machine Learning and system driver implementation between hardware and software drivers using SensorQ which is special queue including user application temporary sensor data. In this paper by experimenting, TensorFlow machine learning training output is applied to the power apparatus for analyzing the status such as "Normal, Warning, Hazard" and predicting the performance at level of 99.6% accuracy and 0.01 loss.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• Journal of information and communication convergence engineering
• /
• 제21권4호
• /
• pp.281-286
• /
• 2023

Diagnosing reading disorders involves complex procedures to evaluate complex cognitive processes. For an accurate diagnosis, a series of tests and evaluations by human experts are required. In this study, we propose a quantitative tool to diagnose reading disorders based on natural reading behaviors using minimal human input. The eye movements of the third- and fourth-grade students were recorded while they read a text at their own pace. Seven machine learning models were used to evaluate the gaze patterns of the words in the presented text and classify the students as normal or having a reading disorder. The accuracy of the machine learning-based diagnosis was measured using the diagnosis by human experts as the ground truth. The highest accuracy of 0.8 was achieved by the support vector machine and random forest classifiers. This result demonstrated that machine learning-based automated diagnosis could substitute for the traditional diagnosis of reading disorders and enable large-scale screening for students at an early age.