최근 기후변화 등으로 인한 꿀벌의 폐사가 증가하고 있으나 관련 데이터가 부족하여 이에 대한 연구가 어려움 을 겪고 있어 학습용 인공지능 데이터를 구축하여 양봉 산업 발전에 기여하고자 한다. 학습용 데이터로 생애주기 별 5단계(알, 애벌레, 번데기, 숫벌, 여왕벌), 종봉별 4가지(이탈리안, 카니올란, 한봉, 호박벌), 발생질병 1종(백묵 병) 총 10가지 클래스를 데이터 수집장소 6곳(장성, 포천, 칠곡, 완주, 의령, 장흥)에서 학습용 데이터를 274,206장 구축하였다. 수집된 데이터는 원시데이터, 원천데이터 가공, 라벨링 데이터 결합, 가공데이터 검수 등을 통해 만들어졌으며 관측지에서 온습도, CO, CH4, NH3 등 환경데이터를 200,000건 확보하여 데이터 라벨링을 수행하 였다. 데이터는 AI Hub(www.aihub.or.kr)에서 다운받을 수 있다. 확보된 데이터는 꿀벌의 생애 단계별, 종별, 건강 상태별 이미지 데이터로 구성되어 있어 양봉 관리 자동화, 꿀벌 질병 예측, 양봉 기술 개발 및 연구 등 다양한 분야에 활용될 수 있을 것으로 예상된다.

식용곤충은 미래식량 자원으로써 우수한 가치를 지니고 있어 해외에서는 사육자동화, IoT 및 AI 기술적용, 수직재배시스템 구축 등 많은 연구가 진행되고 있지만 국내에서는 대규모 사육농가나 곤충스마트팜 기술개발 이 부족하여 이를 위한 AI/빅데이터 인프라 구축이 시급한 실정이다. 학습용 인공지능 데이터는 식용곤충으로 활용되고 있는 장수풍뎅이, 흰점박이꽃무지, 갈색거저리, 백강잠, 메뚜기, 풀무치의 생애 주기별 총 6종의 RGB 사진데이터와 분광이미지 데이터 408,000장을 구축하였으며 온도, 습도, CO,, 암모니아, 조도, 수분 등 환경 데이 터 200,000세트를 수집하였다. 수집된 데이터는 원시데이터 수집, 원천데이터 가공, 라벨링 데이터 결합, 가공데 이터 검수 등을 통해 만들어졌으며 관련 데이터는 AI Hub(www.aihub.or.kr)에서 다운받을 수 있다. 확보된 식용곤 충 6종의 데이터는 곤충 종별 성장단계, 환경 변수에 따른 최적의 사육환경 조성, 생산시기 예측, 스마트대량사육 시스템 개발, 제품 가공시 추적이력제 도입, 식용곤충 스마트팜 기술 개발 및 연구 등 다양한 분야에 활용될 수 있을 것으로 예상된다.

PURPOSES : This study introduces the use of instantaneous speed collected by in-vehicle sensing devices for space mean speed and time mean speed estimation. METHODS : Using probe vehicles’ instantaneous speed data and GPS location at each designated time interval enables analysts to calculate the link-based space mean speed and time mean speed for a certain time-space domain directly. This study proves the equations mathematically and compares the results of time mean speed using Wardrop’s equation and section average speed by trips. RESULTS : This study introduces the concept of link-based average speed using both time mean and space mean speed definitions. First, an unbiased space mean speed was explained for a roadway section can be calculated from high-resolution vehicle trajectory data without necessarily continuously tracking a vehicle. Second, this study proves that the average section speed by occupancy resulted in identical values to the true space mean speed formulation by Wardrop. In addition, the average section speed by trip was nearly identical to the time mean speed. CONCLUSIONS : Each section on a signalized intersection corridor yielded various speed distributions whose parameters varied as measurements were taken closer to the intersection proper. By exploring the space mean speed with speed distribution, the intersection impact area can be defined more accurately.

In this study, acoustic and viscosity data are collected in real time during the ball milling process and analyzed for correlation. After fast Fourier transformation (FFT) of the acoustic data, changes in the signals are observed as a function of the milling time. To analyze this quantitatively, the frequency band is divided into 1 kHz ranges to obtain an integral value. The integrated values in the 2–3 kHz range of the frequency band decrease linearly, confirming that they have a high correlation with changes in viscosity. The experiment is repeated four times to ensure the reproducibility of the data. The results of this study show that it is possible to estimate changes in slurry properties, such as viscosity and particle size, during the ball milling process using an acoustic signal.

The conversation analysis approach of using audio and video recordings of naturally occurring conversation contrasts with other methods of collecting data in the social and cognitive sciences. This paper intends to describe the ways in which conversation analysts collect their data suitable for analysis, including the steps involved in preparing to record, the recording of naturally occurring social interaction, and also the processing of recordings after they have been made. Before actually making the recordings, researchers must first choose what type of conversation they will collect and what device to use. Researchers also need to obtain consent for recording from the participants in a conversation analytic study. While making the recordings, researchers must figure out many practical issues such as who, when, how, what, how much, and where to record. After recording, researchers need to store data in audio or video formats. Researchers also archive it, using codes for identification that facilitate later finding, quoting and retrieving of any file. This paper focuses on these choices as they are made before, during, and after recordings and places specific emphasis on the effects these choices have on the recordings. It also illuminates how these techniques are closely related to the methodological and theoretical assumptions of the conversation analysis approach.

가상현실 콘텐츠는 ICT 기술의 발전과 5G의 고용량 전송, 빠른 속도의 네트워크 서비스가 상용화로 2019년 빠른 성장을 하고 있다. 하지만 가상현실 콘텐츠의 사용성 평가에 대한 기준과 연구는 미흡한 실정이다. 본 연구에서는 가상현실 콘텐츠를 사용하는 사용자의 반응과 상태를 측정 할 수 있는 객관성이 있는 데이터에 관한 연구를 진행한다. 연구의 방법으로는 가상현실 게임 콘텐츠 사용자의 실시간 인체정보 수집 프로세스를 설계하고의 변화를 비교·분석하였다. 이를 위해 인체정보 측정에 아이트래킹을 통한 시선 정보와 심박수 데이터를 수집하고 분석하여 시각화 솔루션을 설계하고 프로토타입을 구현하였다. 가상현실 게임 콘텐츠 사용자 인체정보 데이터를 기반으로 한 사용성 평가에 대한 가이드를 제안한다. 향후연구로는 다양한 인체정보 측정 및 프로토타입의 고도화에 있다.

Over the course of the past decades, technological advancements accompanied a plethora of new types of data and consumer insights (e.g., Erevelles, Fukawa, & Swayne, 2016). Companies value opportunities provided by the availability of large data sets for their business strategy. Customers, however, are wary, as these analyses require the collection and storage of large amounts of personal information. Therefore, it is vital for companies to understand what customers perceive to be fair with regard to their personal data (e.g., Malhotra, Kim, & Agarwal, 2004). However, research still lacks deeper insights into customers’ expectations of fair data handling (Marketing Science Institute, 2016). Yet, only few studies have covered the field of expectations regarding fair data collection and use (Earp, Antón, Aiman-Smith, & Stufflebeam, 2005; Milne & Bahl, 2010). Importantly, however, previous studies have frequently neglected how companies’ fulfillment of customers’ expectations translates into subsequent consumer behavior. Moreover, we have yet to understand if companies’ actual behavior meets customers’ expectations. Grounded on psychological contract and justice theory, we investigate how customers want their data to be handled and in which ways they want to be informed about its usage, while also exploring how customer expectations translate into subsequent behavioral intentions. Additionally, we shed light on current company behavior, thus analyzing if customers’ expectations of fair data collection and usage are aligned with company perspectives. Responding to calls for a mixed methods approach in business research (e.g., Harrison, 2013; Woodside, 2010), we undertook qualitative and quantitative studies to address our research goals. In Study 1, we conduct in-depth interviews with customers and experts to gain an overview of customer expectations with regard to fair data collection and usage. Based on these findings, we conducted a quantitative study (Study 2) investigating each of the customer expectations identified in the prior study. The findings of Study 2 reveal that customers expect a simplification of privacy statements as well as easier control options for their data. Moreover, customers are willing to switch to a competitor, if it better fulfills expectations. Study 3 applies a content analysis of company homepages and privacy statements. Aligning the results from Study 2 and Study 3, we demonstrate that companies currently do not sufficiently meet customers’ expectation of fair data collection and usage.

본 논문에서는 국제해사기구의 데이터수집 시스템 도입에 따른 MRV 지원 및 국제해운 에너지 효율 포탈 시스템에 대한 기술적 검증을 수행하였다. 데이터 수집 시스템과 연료 사용량 데이터 수집 방법론을 포함하는 SEEMP 가이드라인의 주요 내용을 검토하고, EU MRV와 차이점을 분석하여 MRV에 대한 국내 해운선사의 대응전략을 제시하고 이에 대한 MRV 지원시스템의 기술적 적합성을 검토하였다. MRV 지원시스템은 배출량 데이터를 원시단계에서 최종단계까지 통합 관리함으로서 검증을 위한 비용과 업무 효율성을 높일 수 있으며, 현재 해운선사의 보고절차를 유지하면서 데이터 변환 기능을 통해 표준양식으로 배출량 데이터를 수집하고 보고 할 수 있다. 또한, 포탈시스템에 대한 접근권한을 구분하여 해운선사의 데이터 수집과 보고, 검증자의 데이터 검증업무를 지원할 수 있으며 전자적인 방법으로 보고서 제출이 가능하여 MRV 국제 규제에 대한 대응이 가능하다.

Collection of ecological data such as the temporo-spatial distribution of a species is very difficult, due to broad distribution over large areas, phenology, and lack of resources for field survey. Citizen science, which is a cooperative scientific endeavor between researchers and interested citizens, is ideal for collecting large-scale ecological data. However, lack of proper equipment, species identification, and/or communication between researchers and participants are hindrance for a successful citizen science project. Here, we introduce the concept and methods of large-scale ecological data collection using smartphone apps. Most of the ecological data typically consist of sound or video recording, picture, geographic coordinate, and notes. There are several apps that can collect some or all of these ecological data. Furthermore, the result of a survey can be reported to researchers using Google Docs. The data collected by non-specialists can be validated by cross-checking of the survey report by Google Docs and the ecological data sent by apps. Finally, we report the results of a citizen science project in which temporo-spatial distributions of cicada species in Korea were studied via smartphone apps and Google Docs.

In this study, the bridge management system framework for the application of UAV collection data was proposed. The framework of UAV Bridge Management System(UBMS) is designed based on algorithm of Bridge Management System(BMS), administered by MOLIT. The UAV Bridge Management System will help facility managers to conduct safety diagnosis by providing rapid procedure.