• Title/Summary/Keyword: 텍스트 연구

Search Result 3,492, Processing Time 0.029 seconds

Crafting a Quality Performance Evaluation Model Leveraging Unstructured Data (비정형데이터를 활용한 건축현장 품질성과 평가 모델 개발)

  • Lee, Kiseok;Song, Taegeun;Yoo, Wi Sung
    • Journal of the Korea Institute of Building Construction
    • /
    • v.24 no.1
    • /
    • pp.157-168
    • /
    • 2024
  • The frequent occurrence of structural failures at building construction sites in Korea has underscored the critical role of rigorous oversight in the inspection and management of construction projects. As mandated by prevailing regulations and standards, onsite supervision by designated supervisors encompasses thorough documentation of construction quality, material standards, and the history of any reconstructions, among other factors. These reports, predominantly consisting of unstructured data, constitute approximately 80% of the data amassed at construction sites and serve as a comprehensive repository of quality-related information. This research introduces the SL-QPA model, which employs text mining techniques to preprocess supervision reports and establish a sentiment dictionary, thereby enabling the quantification of quality performance. The study's findings, demonstrating a statistically significant Pearson correlation between the quality performance scores derived from the SL-QPA model and various legally defined indicators, were substantiated through a one-way analysis of variance of the correlation coefficients. The SL-QPA model, as developed in this study, offers a supplementary approach to evaluating the quality performance of building construction projects. It holds the promise of enhancing quality inspection and management practices by harnessing the wealth of unstructured data generated throughout the lifecycle of construction projects.

Fine-tuning BERT-based NLP Models for Sentiment Analysis of Korean Reviews: Optimizing the sequence length (BERT 기반 자연어처리 모델의 미세 조정을 통한 한국어 리뷰 감성 분석: 입력 시퀀스 길이 최적화)

  • Sunga Hwang;Seyeon Park;Beakcheol Jang
    • Journal of Internet Computing and Services
    • /
    • v.25 no.4
    • /
    • pp.47-56
    • /
    • 2024
  • This paper proposes a method for fine-tuning BERT-based natural language processing models to perform sentiment analysis on Korean review data. By varying the input sequence length during this process and comparing the performance, we aim to explore the optimal performance according to the input sequence length. For this purpose, text review data collected from the clothing shopping platform M was utilized. Through web scraping, review data was collected. During the data preprocessing stage, positive and negative satisfaction scores were recalibrated to improve the accuracy of the analysis. Specifically, the GPT-4 API was used to reset the labels to reflect the actual sentiment of the review texts, and data imbalance issues were addressed by adjusting the data to 6:4 ratio. The reviews on the clothing shopping platform averaged about 12 tokens in length, and to provide the optimal model suitable for this, five BERT-based pre-trained models were used in the modeling stage, focusing on input sequence length and memory usage for performance comparison. The experimental results indicated that an input sequence length of 64 generally exhibited the most appropriate performance and memory usage. In particular, the KcELECTRA model showed optimal performance and memory usage at an input sequence length of 64, achieving higher than 92% accuracy and reliability in sentiment analysis of Korean review data. Furthermore, by utilizing BERTopic, we provide a Korean review sentiment analysis process that classifies new incoming review data by category and extracts sentiment scores for each category using the final constructed model.

A Study on the Work Process of Creating AI SORA Videos (AI SORA 동영상 생성 제작의 작업 과정에 관한 고찰)

  • Cho, Hyun Kyung
    • The Journal of the Convergence on Culture Technology
    • /
    • v.10 no.5
    • /
    • pp.827-832
    • /
    • 2024
  • The AI program Sora is a video production model that can be used innovatively and is the starting point of a major paradigm shift in video planning and production in the future. In this paper, through consideration of the characteristics, application, and process of the AI video production program, the characteristics of the AI design video production method were understood, and the production algorithm was considered. The detailed consideration and characteristics of the work creation process for the video graphic AI video generation program that will be intensified every year were examined. Next, the method of generating a customized video with a text prompt and the process of innovative production results different from the previous production method were considered. In addition, the design direction through the generation of AI images was studied through the review of the strengths and weaknesses of the image details of the recently announced AI music video results. By considering the security of the AI generation video Sora and looking at the internal process of the actual AI process, it will be possible to present indicators for the future direction of AI video model production and education along with the direction of the design designer and education system. In the text and conclusion, we analyzed the strengths and weaknesses and future status of OpenAI Sora image, concluded how to apply the Sora model's capabilities, limitations, quality, and human creativity, and presented problems and alternatives through examples of the Sora model's capabilities and limitations to increase human creativity.

'One's own sense of soverignty's Realization and Cultivation through 『Gyeokmongyogyeol』 (『격몽요결』을 통한 '주체적인 나'의 자각과 도야(陶冶) - 「수신장(修身章)」을 통한 청소년 철학하기 방법론 제시 -)

  • 황정희
    • 유학연구
    • /
    • v.42
    • /
    • pp.57-81
    • /
    • 2018
  • At present, while the East Asian thoughts are world-widely attract public attention, the education of philosophy in our society remains at the stage of mostly borrowing the Western texts and methods. Book of general education Gyeokmongyogyeol, written by Yulgok for the novices to go into neo-Confucianism, covers moral training, filial duty, courtesy and conduct of life over all. As for him, the study was not merely acquisition of knowledge, but also the whole process of feeling and realizing one's own dream in one's own life and cultivating one's own human nature. Such education philosophy of Yulgok is likely to be closely related with modern philosophical practice. Philosophical practice is a very dynamic and practical process in which wisdom acquired from philosophy is applied to reality. This study assumes that the youth cannot feel happy from their own studies while they spend lots of time and make great efforts for them, since they don't have their own determined dreams with resolute will, that is, not having one's own sense of sovereignty. In addition, as for the solution to this, based upon the contents of "chapter of moral training" in Gyeokmongyogyeol, realization and cultivation of one's own sense of sovereignty are reviewed, and the methods for philosophical thinking of the youth are proposed. It is considered that Confucianism in the Joseon Dynasty will contribute to the philosophical practice through the process of philosophical thinking if the methods for philosophy education are developed and directly applied to the spot of education through the above research. In addition, the youth in our country will realize that training in philosophical thinking is a practical process that can be achieved by themselves in their learning process.

Automated Scoring of Scientific Argumentation Using Expert Morpheme Classification Approaches (전문가의 형태소 분류를 활용한 과학 논증 자동 채점)

  • Lee, Manhyoung;Ryu, Suna
    • Journal of The Korean Association For Science Education
    • /
    • v.40 no.3
    • /
    • pp.321-336
    • /
    • 2020
  • We explore automated scoring models of scientific argumentation. We consider how a new analytical approach using a machine learning technique may enhance the understanding of spoken argumentation in the classroom. We sampled 2,605 utterances that occurred during a high school student's science class on molecular structure and classified the utterances into five argumentative elements. Next, we performed Text Preprocessing for the classified utterances. As machine learning techniques, we applied support vector machines, decision tree, random forest, and artificial neural network. For enhancing the identification of rebuttal elements, we used a heuristic feature-engineering method that applies experts' classification of morphemes of scientific argumentation.

Dwelling Depression Measurement Based on Image Analysis Modeling: Focusing on K-HTP (이미지분석 모델링 기반 고령자 주거우울 측정 연구 -K-HTP를 중심으로-)

  • Lee, Yewon;Park, Chongwook;Woo, Sungju
    • The Journal of the Convergence on Culture Technology
    • /
    • v.4 no.3
    • /
    • pp.1-6
    • /
    • 2018
  • With the increase of the elderly population, demand for improvement of quality of life and measurement of mental state has increased. However, much of the self-reported diagnosis does not reflect cognitive impairment. This study aims to measure the dwelling depression by applying K-HTP and verify the validity. 301 persons over 65 years old who live as single and couple households in Daejeon and surrounding districts were surveyed during 22 January to 2 February, 2018. The correlations between the dwelling depression and K-HTP are clarified and the validity was evaluated. The correlations between the geriatric dwelling depression index(GDDI) and the GDDI based on K-HTP(GDDI-K) are clarified and the accuracy was analyzed. The results showed that the K-HTP can be utilized to measure the dwelling depression. We suggested a new measurement tool and provide further benefits for researches on diagnoses using the projective method.

A Discourse for the Theory of Adaptive Learning Object Design (적응적 학습객체 설계 이론을 위한 개념적 연구)

  • Jo, Il-Hyun
    • Journal of The Korean Association of Information Education
    • /
    • v.9 no.3
    • /
    • pp.483-500
    • /
    • 2005
  • The purpose of the study was to explore the conceptual and theoretical fundamentals of learning object. Learning object, a new paradigm for instructional design in the era of information technology, has attracted much research efforts since it has lots of advantages in terms of production efficiency and use effectiveness. A theory for the systematic design of this new instructional design, however, looks far from mature. Since the birth of the idea of a learning object has been found in the field of computer software design, such as object-oriented software development, learning object does not have enough theoretical underpinnings in terms of learning and instruction. The researcher tried to establish theoretical foundations for this new, alien concept as a learning design theory. Relevant research efforts and discourses have been discussed for this purpose.

  • PDF

A narrative on Badminton Sports club Activity and Adaption of a North Korean refugee child (탈북아동의 배드민턴 스포츠클럽 활동과 적응)

  • Lee, Je-Haeng
    • Journal of the Korea Convergence Society
    • /
    • v.8 no.8
    • /
    • pp.333-343
    • /
    • 2017
  • This study aimed to narrate the feature of a north korean refugee child at an elementary school in south korea through badminton sports cub activity. The narrative was as follows. Dongsik was unhappy at the first time in living here(south korea). After a while he had encountered badminton activity by chance. And he changed his life at the school and he has his own dream now. Teacher choi has remembered the first meeting with Dongsik. Dongsik was a boy who did not talk with anyone. After a long time Dongsik spent his life without uttering a word. Now he has a talk about his dream due to badminton. He dreams of being a badminton player. What is primarily important for north korean refugee children is educational direction. School eduction has to integrate not segregate to participate in all program together. We all have to take both expectation and confidence for north korean refugee children.

Implementation of Academic Journal Map based on Electronic Cultural Atlas using Google Maps (구글맵스 전자문화지도 기반의 학술지 지도 구현)

  • Kang, Ji-Hoon;Moon, Sang-Ho
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.19 no.4
    • /
    • pp.864-870
    • /
    • 2015
  • With the growing interests in ICT convergence, interdisciplinary researches are still actively discussed and studied today such as cultural atlas DB, electronic cultural atlas. Electronic cultural atlas can show various cultural aspects on the map by using point, line, polygon, and so on. Users can obtain the informations connected to the three values through manipulating theme, spatial and time values. In this regard, text information can be visualized as various ways like point, line, and polygon on the map. This can make the user get information effectively by usability, accessibility and immediacy of the system. In this paper, we implemented academic journal map system based on the electronic cultural atlas using Google Maps. This system supports various journals information services based on electronic cultural atlas and can be used efficiently as a tool for academic trends analysis. Also, it provides academic material through visualization of information based on the map. In detail, it can be utilized as the foundations for research of Humanities and Area Studies.

The Implementation of the Digital watermarking for 3D Polygonal Model (3차원 형상 모델의 디지털 워터마킹 구현)

  • Kim, Sun-Hyung;Lee, Sun-Heum;Kim, Gee-Seog;Ahn, Deog-Sang
    • The KIPS Transactions:PartD
    • /
    • v.9D no.5
    • /
    • pp.925-930
    • /
    • 2002
  • This paper discusses techniques for embedding data into 3D polygonal models of geometry. Much researches of Watermarking had been gone as element technology of DRM (digital rights management). But, few research had gone to 3D polygonal model. Most research is limited at text document, 2D image, animation, music etc. RP system is suitable a few production in various goods species, and it is used much in industry to possible reason that produce prototype and find error or incongruent factor at early stage on design in product development childhood. This paper is research about method that insert watermark in STL ( stereolithography) file that have 3D shape model. Proposed algorithm inserts watermark in normal vector region and facet's interior region of 3D shape data. For this reason, 3D shape does not produce some flexure and fulfill invisibility of watermark. Experiment results that insert and extract watermark in normal netter region and facet's Interior region of 3D shape data by proposed algorithm do not influence entirely in 3D shape and show that insertion and extraction of watermark are possible.