• Title/Summary/Keyword: Missing Data Handling

Search Result 38, Processing Time 0.026 seconds

Smart device research for the prevention of missing child (미아 방지용 스마트 디바이스 구현에 관한 연구)

  • Ahn, Jong-Chan;Kim, Young-Kil
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2007.10a
    • /
    • pp.437-440
    • /
    • 2007
  • Recently embedded system developed a lot. Physically, embedded systems range from portable devices such as digital watches and MP3 players, to large stationary installations like traffic lights, factory controllers, or the systems controlling nuclear power plants. This paper focuses on implementation of portable device which is applicable to the child-kidnap or missing child prevention system in residential area or public area. To be specific, this device is to transmit video data which comes from the camera in the device into the host PC via WLAN. Embedded hardware platform consists of s3c2440 with ARM9 core, WindowsCE OS and other sensors. OS enables the platform to do multitasking jobs which are handling GPS data, taking video, capturing audio via microphone in the device and transfer all kind of realtime data to the host PC.

  • PDF

Design of Sensor Data's Missing Value Handling Technique for Pet Healthcare Service based on Graph Attention Networks (펫 헬스 케어 서비스를 위한 GATs 기반 센서 데이터 처리 기법 설계)

  • Lee, Jihoon;Moon, Nammee
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2021.05a
    • /
    • pp.463-465
    • /
    • 2021
  • 센서 데이터는 여러가지 원인으로 인해 데이터 결측치가 발생할 수 있으며, 결측치로 인한 데이터의 처리 방식에 따라 데이터 분석 결과가 다르게 해석될 수 있다. 이는 펫 헬스 케어 서비스에서 치명적인 문제로 연결될 수 있다. 따라서 본 논문에서는 펫 웨어러블 디바이스로부터 수집되는 다양한 센서 데이터의 결측치를 처리하기 위해 GATs(Graph Attention neTworks)와 LSTM(Long Short Term Memory)을 결합하여 활용한 데이터 결측치 처리 기법을 제안한다. 펫 웨어러블 디바이스의 센서 데이터가 서로 연관성을 가지고 있다는 점을 바탕으로 인접 노드의 Attention 수치와 Feature map을 도출한다. 이후 Prediction Layer 를 통해 결측치의 Feature 를 예측한다. 예측된 Feature 를 기반으로 Decoding 과정과 함께 결측치 보간이 이루어진다. 제안된 기법은 모델의 변형을 통해 이상치 탐지에도 활용할 수 있을 것으로 기대한다.

Data Cleaning and Integration of Multi-year Dietary Survey in the Korea National Health and Nutrition Examination Survey (KNHANES) using Database Normalization Theory (데이터베이스 정규화 이론을 이용한 국민건강영양조사 중 다년도 식이조사 자료 정제 및 통합)

  • Kwon, Namji;Suh, Jihye;Lee, Hunjoo
    • Journal of Environmental Health Sciences
    • /
    • v.43 no.4
    • /
    • pp.298-306
    • /
    • 2017
  • Objectives: Since 1998, the Korea National Health and Nutrition Examination Survey (KNHANES) has been conducted in order to investigate the health and nutritional status of Koreans. The food intake data of individuals in the KNHANES has also been utilized as source dataset for risk assessment of chemicals via food. To improve the reliability of intake estimation and prevent missing data for less-responded foods, the structure of integrated long-standing datasets is significant. However, it is difficult to merge multi-year survey datasets due to ineffective cleaning processes for handling extensive numbers of codes for each food item along with changes in dietary habits over time. Therefore, this study aims at 1) cleaning the process of abnormal data 2) generation of integrated long-standing raw data, and 3) contributing to the production of consistent dietary exposure factors. Methods: Codebooks, the guideline book, and raw intake data from KNHANES V and VI were used for analysis. The violation of the primary key constraint and the $1^{st}-3rd$ normal form in relational database theory were tested for the codebook and the structure of the raw data, respectively. Afterwards, the cleaning process was executed for the raw data by using these integrated codes. Results: Duplication of key records and abnormality in table structures were observed. However, after adjusting according to the suggested method above, the codes were corrected and integrated codes were newly created. Finally, we were able to clean the raw data provided by respondents to the KNHANES survey. Conclusion: The results of this study will contribute to the integration of the multi-year datasets and help improve the data production system by clarifying, testing, and verifying the primary key, integrity of the code, and primitive data structure according to the database normalization theory in the national health data.

Analysis of Phenological Changes by Phenocams on Some Major Species Distributed in Wetland and Forest Ecosystems in Korea (Phenocam을 활용한 국내 습지 및 산림생태계 대표 수종의 계절적 변화 분석)

  • Minki Hong;Hyohyemi Lee;Jeong-Soo Park
    • Ecology and Resilient Infrastructure
    • /
    • v.10 no.4
    • /
    • pp.226-236
    • /
    • 2023
  • As climate change intensifies, the importance of studying plant phenology has increased, leading to a surge in research employing automated video recording devices like Phenocams. In this study, using the Phenocams operated by the National Institute of Ecology, we examined the trends in plant phenological changes across diverse ecosystem types in South Korea and analyzed their correlations with climate factors. The patterns of plant phenological changes varied by region and tree species. Pinus thunbergii and Pinus densiflora typically show an overall increase in their growth period, positively correlating with temperatures and precipitation during winter. However, uniquely, for Abies koreana on Hallasan Mt., a higher amount of precipitation in August leads to an earlier end of season (eos), and the correlation analysis with the recent phenomenon of dying A. Koreana seems necessary. beyond the analysis, solutions for handling missing data issues during the data collection process were proposed. Furthermore, to expand future research scope and encompass diverse ecosystem types, a suggestion to combine Phenocam research with satellite observations was presented.

Reporting Quality of Research Studies on AI Applications in Medical Images According to the CLAIM Guidelines in a Radiology Journal With a Strong Prominence in Asia

  • Dong Yeong Kim;Hyun Woo Oh;Chong Hyun Suh
    • Korean Journal of Radiology
    • /
    • v.24 no.12
    • /
    • pp.1179-1189
    • /
    • 2023
  • Objective: We aimed to evaluate the reporting quality of research articles that applied deep learning to medical imaging. Using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines and a journal with prominence in Asia as a sample, we intended to provide an insight into reporting quality in the Asian region and establish a journal-specific audit. Materials and Methods: A total of 38 articles published in the Korean Journal of Radiology between June 2018 and January 2023 were analyzed. The analysis included calculating the percentage of studies that adhered to each CLAIM item and identifying items that were met by ≤ 50% of the studies. The article review was initially conducted independently by two reviewers, and the consensus results were used for the final analysis. We also compared adherence rates to CLAIM before and after December 2020. Results: Of the 42 items in the CLAIM guidelines, 12 items (29%) were satisfied by ≤ 50% of the included articles. None of the studies reported handling missing data (item #13). Only one study respectively presented the use of de-identification methods (#12), intended sample size (#19), robustness or sensitivity analysis (#30), and full study protocol (#41). Of the studies, 35% reported the selection of data subsets (#10), 40% reported registration information (#40), and 50% measured inter and intrarater variability (#18). No significant changes were observed in the rates of adherence to these 12 items before and after December 2020. Conclusion: The reporting quality of artificial intelligence studies according to CLAIM guidelines, in our study sample, showed room for improvement. We recommend that the authors and reviewers have a solid understanding of the relevant reporting guidelines and ensure that the essential elements are adequately reported when writing and reviewing the manuscripts for publication.

Safety countermeasures for the marine casualties of fishing vessels in Korea (우리나라의 어선 해양사고에 대한 안전대책)

  • Kang, Il-Kwon;Kim, Hyung-Seok;Shin, Hyeong-Il;Lee, Yoo-Won;Kim, Jeong-Chang;Jo, Hyo-Jae
    • Journal of the Korean Society of Fisheries and Ocean Technology
    • /
    • v.43 no.2
    • /
    • pp.149-159
    • /
    • 2007
  • Marine casualties of fishing vessels were analyzed to reduce the sacrifice of human life using data of the Korean Maritime Safety Tribunal from 1995 to 2004 in Korea. The occurred number of fishing vessel casualties were likely to be higher portion than non-fishing vessels, but the occurring ratio of fishing vessel casualties were marked 2.96 times lower than that of non-fishing vessel casualties. The occurring ratios of bigger fishing vessel casualties were higher than smaller ones. Most marine casualties were resulted from the human factors such as poor watchkeeping, negligent action for engine and etc. The trend of marine casualties showed that the machinery damage hold the first and collision accidents took the second, but on a point of cause of them, operating errors took first and poor handling or inspection of machinery held the second place. Because those two casualties took major portion, and very important problems for safety of fishing vessels, so we ought to try to reduce the factors before everything else. In addition, since collision, sinking and capsizing in marine casualties have led to death, missing and injury of lives, it is necessary for navigation operators to take more educations and training intended to reduce the marine casualties systematically and continuously.

Quality of Radiomics Research on Brain Metastasis: A Roadmap to Promote Clinical Translation

  • Chae Jung Park;Yae Won Park;Sung Soo Ahn;Dain Kim;Eui Hyun Kim;Seok-Gu Kang;Jong Hee Chang;Se Hoon Kim;Seung-Koo Lee
    • Korean Journal of Radiology
    • /
    • v.23 no.1
    • /
    • pp.77-88
    • /
    • 2022
  • Objective: Our study aimed to evaluate the quality of radiomics studies on brain metastases based on the radiomics quality score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) checklist, and the Image Biomarker Standardization Initiative (IBSI) guidelines. Materials and Methods: PubMed MEDLINE, and EMBASE were searched for articles on radiomics for evaluating brain metastases, published until February 2021. Of the 572 articles, 29 relevant original research articles were included and evaluated according to the RQS, TRIPOD checklist, and IBSI guidelines. Results: External validation was performed in only three studies (10.3%). The median RQS was 3.0 (range, -6 to 12), with a low basic adherence rate of 50.0%. The adherence rate was low in comparison to the "gold standard" (10.3%), stating the potential clinical utility (10.3%), performing the cut-off analysis (3.4%), reporting calibration statistics (6.9%), and providing open science and data (3.4%). None of the studies involved test-retest or phantom studies, prospective studies, or cost-effectiveness analyses. The overall rate of adherence to the TRIPOD checklist was 60.3% and low for reporting title (3.4%), blind assessment of outcome (0%), description of the handling of missing data (0%), and presentation of the full prediction model (0%). The majority of studies lacked pre-processing steps, with bias-field correction, isovoxel resampling, skull stripping, and gray-level discretization performed in only six (20.7%), nine (31.0%), four (3.8%), and four (13.8%) studies, respectively. Conclusion: The overall scientific and reporting quality of radiomics studies on brain metastases published during the study period was insufficient. Radiomics studies should adhere to the RQS, TRIPOD, and IBSI guidelines to facilitate the translation of radiomics into the clinical field.

Elementary School Dietitian's Awareness and Performance of Food Supply Management in Gyeonggi North Province (경기 북부 지역 초등학교 영양사의 식자재 공급관리에 대한 인식도 및 실천도)

  • Eo, Geum-Hee;Park, Young-Sim
    • The Korean Journal of Food And Nutrition
    • /
    • v.21 no.4
    • /
    • pp.562-571
    • /
    • 2008
  • The principal objective of this study was to assess dietitian's awareness and performance with regard to food suppliers' selection guidelines and purchase guidelines for the receipt of safe food materials. A questionnaire was administered to 203 dietitians working at an elementary school in Gyeonggi North province, and 190 responses were ultimately returned. Excluding responses with incomplete answers and significant missing data, 161 responses(79.3%) were ultimately utilized for data analysis. We determined that meat/poulty and seafood were purchased mainly by manufacturer's branch(59.8% and 78.3%), and processed food and kimchi were generally purchased by producer's cooperatives(47.7% and 44.9%). 78.3% of the contracts were made via informal purchasing and the frequency of contracts was less than 3 times per year(53.4%). Market studies were conducted individually(54.7%), and by group(47.2%). Dietitian and parents volunteered(50.3%) or dietitians, and school and foodservice staffs(45.3%) participated - together in receiving and inspection. School contracted with $4{\sim}5$(39.1%) and $6{\sim}8$ food suppliers(29.8%). Dietitians(42.3%) or school councils(40.2%) evaluated food suppliers once per semester(60.2%). The majority of dietitians(96.3%) conducted sanitary education for food suppliers once per semester (68.3%). All 13 guidelines which were usually used to evaluate food suppliers were thought to be important(more than 4) based on a 5 scale(1: never important, 5: very important). Among them, 'sanitary storage status of food products(4.85)', 'have a chill car(4.83)' and 'proper control of refrigerator, freezer and storage room(4.81)' were the most crucial guidelines. As dietitians evaluated food supplier's-observances of 13 guidelines, 'have a chill car(4.89)', 'hold medical examination of employees(4.89)' and 'hold liability insurance for handling product(4.80)' were fulfilled relatively well. The guidelines for conducting education for their employees(4.45) and keeping their establishment and equipment sanitary(4.79) were important, but were not observed well(3.39 and 3.37). Additionally, the difference between the importance and the observance score of the 'HACCP certificate' were fairly significant(4.44 and 3.54). Dietitians tended to report that purchasing management related to the removal of foodborne illness factors(4.71), return confirmation(4.50) and purchase specification(4.32) were important. The difference between importance and performance was highest in the process of changing food suppliers when the contracts were violated more than three times. Dietitian's age, education work experience and employment type had no observable effects on how important they considered purchase management to be, but their work experience significantly affected performance in terms of the removal of foodborne illness factors(p<0.05). As supplier management is the most critical factor, it is clearly desirable to educate suppliers at specialized training centers and to supervise suppliers in accordance with the established sanitary guidelines.