• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.830-832
• /
• 2019

IoT 기술의 도래로 인하여 실생활에 사용되는 사물들에 Sensor가 부착되어 시간마다 Sensor data가 발생하는 세상이 열리게 되었다. 이러한 IoT Device들에 부착되어 있는 sensor를 통하여 수집이 된 data는 방대한 양을 가지기 때문에 Deep Learning에 적용하는데 충분하며 아주 중요한 역할을 한다. 이러한 IoT Device들은 우리의 실제 생활에 아주 가까이 다양한 환경으로 접할 수 있다. 예를 들어 스마트시티, 스마트팩토리, 스마트홈 등이 있다. 이러한 것들은 우리의 일상생활에 편리함과 직결되어 있다. 본 논문에서는 Smart home 환경에서의 Multi Resident Activity Recognition이다. Smart home의 가구에 부착되어 있는 센서에서 발생된 센서데이터를 활용하여 1) Training Similarity Network, 2) Embedding, 3) Clustering, 4) Recognizing 네 단계 프로세스를 거쳐 문제를 해결한다. 그 결과, 우리가 제안한 프로세스를 통하여 차원 축소 효과와 Un-seen data를 효과적으로 처리할수 있게 된다.

• The Journal of Information Systems
• /
• v.30 no.3
• /
• pp.201-216
• /
• 2021

Purpose The purpose of this study is to analyze the trend of patent technology in textile materials using text mining methodology based on Dynamic Embedded Topic Model and Structural Topic Model. It is expected that this study will have positive impact on revitalizing and developing textile materials industry as finding out technology trends. Design/methodology/approach The data used in this study is 866 domestic patent text data in textile material from 1974 to 2020. In order to analyze technology trends from various aspect, Dynamic Embedded Topic Model and Structural Topic Model mechanism were used. The word embedding technique used in DETM is the GloVe technique. For Stable learning of topic modeling, amortized variational inference was performed based on the Recurrent Neural Network. Findings As a result of this analysis, it was found that 'manufacture' topics had the largest share among the six topics. Keyword trend analysis found the fact that natural and nanotechnology have recently been attracting attention. The metadata analysis results showed that manufacture technologies could have a high probability of patent registration in entire time series, but the analysis results in recent years showed that the trend of elasticity and safety technology is increasing.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.599-614
• /
• 2021

This paper proposes a digital signage system based on an intelligent recommendation model. The proposed system consists of a server and an edge. The server manages the data, learns the advertisement recommendation model, and uses the trained advertisement recommendation model to determine the advertisements to be promoted in real time. The advertisement recommendation model provides predictions for various products and probabilities. The purchase index between the product and weather data was extracted and reflected using correlation analysis to improve the accuracy of predicting the probability of purchasing a product. First, the user information and product information are input to a deep neural network as a vector through an embedding process. With this information, the product candidate group generation model reduces the product candidates that can be purchased by a certain user. The advertisement recommendation model uses a wide and deep recommendation model to derive the recommendation list by predicting the probability of purchase for the selected products. Finally, the most suitable advertisements are selected using the predicted probability of purchase for all the users within the advertisement range. The proposed system does not communicate with the server. Therefore, it determines the advertisements using a model trained at the edge. It can also be applied to digital signage that requires immediate response from several users.

• Journal of the Korean Society of Safety
• /
• v.36 no.1
• /
• pp.56-63
• /
• 2021

The narrative texts of industrial accident reports help to identify accident risk factors. They relate the accident triggers to the sequence of events and the outcomes of an accident. Particularly, a set of related keywords in the context of the narrative can represent how the accident proceeded. Previous studies on text analytics for structuring accident reports have been limited to extracting individual keywords without context. We proposed a context-based analysis using a Natural Language Processing (NLP) algorithm to remedy this shortcoming. This study aims to apply Word2Vec of the NLP algorithm to extract adjacent keywords, known as word embedding, conducted by the neural network algorithm based on supervised learning. During processing, Word2Vec is conducted by adjacent keywords in narrative texts as inputs to achieve its supervised learning; keyword weights emerge as the vectors representing the degree of neighboring among keywords. Similar keyword weights mean that the keywords are closely arranged within sentences in the narrative text. Consequently, a set of keywords that have similar weights presents similar accidents. We extracted ten accident processes containing related keywords and used them to understand the risk factors determining how an accident proceeds. This information helps identify how a checklist for an accident report should be structured.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.108-111
• /
• 2022

Applying deep learning to machine vision systems for defect detection of products requires vast amounts of training data about various defect cases. However, since data imbalance occurs according to the type of defect in the actual manufacturing industry, it takes a lot of time to collect product images enough to generalize defect cases. In this paper, we apply a Siamese neural network that can be learned with even a small amount of data to product defect detection, and modify the image pairing method and contrastive loss function by properties the situation of product defect image data. We indirectly evaluated the embedding performance of Siamese neural networks using AUC-ROC, and it showed good performance when the images only paired among same products, not paired among defective products, and learned with exponential contrastive loss.

• Journal of Korea Multimedia Society
• /
• v.15 no.3
• /
• pp.299-311
• /
• 2012

Recently, H.264 SE(scalable extension) has become a standard of next generation multimedia service which is one source, multi-user service in the telecommunication environment of different kinds of networks and terminal equipments. But existing DRM schemes for multimedia service are not fit for H.264 SE system. Because the amount of transmitted multimedia data is changed considering network environment and terminal equipments' performance by the system, but in the existing DRM schemes, the amount of handled multimedia data are not variable according to network environment and terminal equipments' performance. In this paper, an information security scheme combined video watermarking and encryption is presented for H.264 SE. Amount of watermarks and embedding positions are calculated by the frame number of enhancement layers which are created according to the state of networks and terminal equipments. In order to minimize delayed time by video watermarking and encryption, the video data are watermarked and encrypted in the H.264 SE compression process. In the experimental results, we confirmed that proposed scheme is robust against video compression, general signal processing and geometric processing.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.447-453
• /
• 2020

In this paper, we propose a system to extract effective speaker representations from a speech signal using a deep learning method. Based on the fact that speech signal contains identity unrelated information such as text content, emotion, background noise, and so on, we perform a training such that the extracted features only represent speaker-related information but do not represent speaker-unrelated information. Specifically, we propose an auto-encoder based disentanglement method that outputs both speaker-related and speaker-unrelated embeddings using effective loss functions. To further improve the reconstruction performance in the decoding process, we also introduce a discriminator popularly used in Generative Adversarial Network (GAN) structure. Since improving the decoding capability is helpful for preserving speaker information and disentanglement, it results in the improvement of speaker verification performance. Experimental results demonstrate the effectiveness of our proposed method by improving Equal Error Rate (EER) on benchmark dataset, Voxceleb1.

• International Journal of Computer Science & Network Security
• /
• v.21 no.1
• /
• pp.143-151
• /
• 2021

The enormous prevalence of transferring official confidential digital documents via the Internet shows the urgent need to deliver confidential messages to the recipient without letting any unauthorized person to know contents of the secret messages or detect there existence . Several Steganography techniques such as the least significant Bit (LSB), Secure Cover Selection (SCS), Discrete Cosine Transform (DCT) and Palette Based (PB) were applied to prevent any intruder from analyzing and getting the secret transferred message. The utilized steganography methods should defiance the challenges of Steganalysis techniques in term of analysis and detection. This paper presents a novel and robust framework for color image steganography that combines Linear Congruential Generator (LCG), simulated annealing (SA), Cesar cryptography and LSB substitution method in one system in order to reduce the objection of Steganalysis and deliver data securely to their destination. SA with the support of LCG finds out the optimal minimum sniffing path inside a cover color image (RGB) then the confidential message will be encrypt and embedded within the RGB image path as a host medium by using Cesar and LSB procedures. Embedding and extraction processes of secret message require a common knowledge between sender and receiver; that knowledge are represented by SA initialization parameters, LCG seed, Cesar key agreement and secret message length. Steganalysis intruder will not understand or detect the secret message inside the host image without the correct knowledge about the manipulation process. The constructed system satisfies the main requirements of image steganography in term of robustness against confidential message extraction, high quality visual appearance, little mean square error (MSE) and high peak signal noise ratio (PSNR).

• Journal of Information Processing Systems
• /
• v.19 no.3
• /
• pp.289-301
• /
• 2023

Due to the development and dissemination of modern technology, anyone can easily communicate using services such as social network service (SNS) through a personal computer (PC) or smartphone. The development of these technologies has caused many beneficial effects. At the same time, bad effects also occurred, one of which was the spam problem. Spam refers to unwanted or rejected information received by unspecified users. The continuous exposure of such information to service users creates inconvenience in the user's use of the service, and if filtering is not performed correctly, the quality of service deteriorates. Recently, spammers are creating more malicious spam by distorting the image of spam text so that optical character recognition (OCR)-based spam filters cannot easily detect it. Fortunately, the level of transformation of image spam circulated on social media is not serious yet. However, in the mail system, spammers (the person who sends spam) showed various modifications to the spam image for neutralizing OCR, and therefore, the same situation can happen with spam images on social media. Spammers have been shown to interfere with OCR reading through geometric transformations such as image distortion, noise addition, and blurring. Various techniques have been studied to filter image spam, but at the same time, methods of interfering with image spam identification using obfuscated images are also continuously developing. In this paper, we propose a deep learning-based spam image detection model to improve the existing OCR-based spam image detection performance and compensate for vulnerabilities. The proposed model extracts text features and image features from the image using four sub-models. First, the OCR-based text model extracts the text-related features, whether the image contains spam words, and the word embedding vector from the input image. Then, the convolution neural network-based image model extracts image obfuscation and image feature vectors from the input image. The extracted feature is determined whether it is a spam image by the final spam image classifier. As a result of evaluating the F1-score of the proposed model, the performance was about 14 points higher than the OCR-based spam image detection performance.

• The Journal of Korean Obstetrics and Gynecology
• /
• v.37 no.2
• /
• pp.75-108
• /
• 2024

Objectives: The purpose of this study is to review the acupuncture treatment for Polycystic ovary syndrome (PCOS) in women. Methods: We searched articles in 3 search engines with keywords related to 'Polycystic ovary syndrome', 'PCOS', and 'Stein-leventhal' in February 2024. Clinical researches and case reports that used acupuncture on PCOS were included. Animal studies and non clinical data were excluded. Data on acupuncture treatment such as methods, site, duration, frequency, and period were analyzed. Results: Of 60 selected articles, there were 51 randomized controlled trials, 5 clinical trials and 4 case reports. Studies were conducted using manual acupuncture, electro acupuncture, auricular acupuncture, thread embedding acupuncture, warm needling and laser acupuncture. Most studies used more than one acupoint, and there were 78 acupoints selected for acupuncture treatment for PCOS. The most commonly used acupoint was 三陰交 (SP6)(n=50). By analyzing the network of acupoints, 關元 (CV4), 氣海 (CV6), 中脘 (CV12), 三陰交 (SP6), 血海 (SP10), 天樞 (ST25), 足三里 (ST36) were located in center of the network. The mean treatment time, number of treatments, and duration were 28.63±4.48 minutes, 34.52±29.26 times, and 98.18±38.25 days. Conclusions: The results of this study could be useful in establishing the evidence for performing standardized acupuncture treatment for Polycystic ovary syndrome.