• International Journal of Internet, Broadcasting and Communication
• /
• 제15권3호
• /
• pp.58-65
• /
• 2023

Bankruptcy is a significant risk for start-up companies, but with the help of cutting-edge artificial intelligence technology, we can now predict bankruptcy with detailed explanations. In this paper, we implemented the Category Boosting algorithm following data cleaning and editing using OpenRefine. We further explained our model using the Shapash library, incorporating domain knowledge. By leveraging the 5C's credit domain knowledge, financial analysts in banks or investors can utilize the detailed results provided by our model to enhance their decision-making processes, even without extensive knowledge about AI. This empowers investors to identify potential bankruptcy risks in their business models, enabling them to make necessary improvements or reconsider their ventures before proceeding. As a result, our model serves as a "glass-box" model, allowing end-users to understand which specific financial indicators contribute to the prediction of bankruptcy. This transparency enhances trust and provides valuable insights for decision-makers in mitigating bankruptcy risks.

• 스마트미디어저널
• /
• 제13권4호
• /
• pp.9-15
• /
• 2024

Tree-based algorithms have been the dominant methods used build a prediction model for tabular data. This also includes personal credit data. However, they are limited to compatibility with categorical and numerical data only, and also do not capture information of the relationship between other features. In this work, we proposed an ensemble model using the Transformer architecture that includes text features and harness the self-attention mechanism to tackle the feature relationships limitation. We describe a text formatter module, that converts the original tabular data into sentence data that is fed into FinBERT along with other text features. Furthermore, we employed FT-Transformer that train with the original tabular data. We evaluate this multi-modal approach with two popular tree-based algorithms known as, Random Forest and Extreme Gradient Boosting, XGBoost and TabTransformer. Our proposed method shows superior Default Recall, F1 score and AUC results across two public data sets. Our results are significant for financial institutions to reduce the risk of financial loss regarding defaulters.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제15권2호
• /
• pp.135-139
• /
• 2009

본 논문에서는 지역 전문가를 이용한 새로운 앙상블 방법을 제시하고자 한다. 이 앙상블 방법에서는 학습 데이타를 분할하여 속성 공간의 서로 다른 지역을 이용하여 전문가를 학습시킨다. 새로운 데이타를 분류할 때에는 그 데이타가 속한 지역을 담당하는 전문가들로 가중치 투표를 한다. UCI 기계 학습 데이타 저장소에 있는 10개의 데이타를 이용하여 단일 분류기, Bagging, Adaboost와 정확도를 비교하였다. 학습 알고리즘으로는 SVM, Naive Bayes, C4.5를 사용하였다. 그 결과 지역 전문가의 앙상블 학습 방법이 C4.5를 학습 알고리즘으로 사용한 Bagging, Adaboost와는 비슷한 성능을 보였으며 나머지 분류기보다는 좋은 성능을 보였다.

• International Journal of Human Ecology
• /
• 제9권2호
• /
• pp.117-126
• /
• 2008

The study identifies gender stereotypes, examines the level of women's economic empowerment at the household level and explores the influence of factors on it. Data were collected from 159 randomly selected farm women using both qualitative and quantitative survey methods between January and March 2007 from three villages of the Mymensingh District of Bangladesh. Four key informants (2 local leaders and 2 development personnel) were questioned to elicit views in the light of boosting women's empowerment. Five constructs of empowerment covering 30 indicators were aggregated together to develop a cumulative economic empowerment index (CEEI) to obtain multidimensional views of women's empowerment. The findings show that there were some prejudices against women in allocating divisions of labour and access to education, food, property, decision-making and institutions. The distribution of the CEEI demonstrates that the majority of the respondents (86%) had a low to moderate level of empowerment. A multiple regression analysis showed positive significant effects of education, training, media contact and freedom of mobility on women's CEEI, while domestic abuse restrained it. It is concluded that interventions by development agencies in co-ordination with the local community was necessary to attain women's self-reliance in the study area. Development actors can undertake some core strategies to enhance women's level of awarencess, knowledge, skills, and productive resources through providing training, loans, and information. To change traditional beliefs, it is important to create awareness of various gender issues amongst rural people through different methods and media.

• 전력전자학회논문지
• /
• 제16권4호
• /
• pp.326-333
• /
• 2011

본 논문에서는 Plug-in Hybrid Electric Vehicles (PHEVs)의 배터리 충전을 위해 필요한 추가적인 충전기 없이, 구동모터의 인덕턴스와 구동 드라이버인 3상 인버터를 이용하는 배터리 충전 기법들을 소개한다. 모터의 코일을 승압용 에너지 저장장치로 사용하고 인버터 스위치 제어를 통해 부스트 컨버터로 동작되도록 하여 추가적인 충전기를 제거함으로써, 충전장치가 차지하는 부피, 무게 및 단가를 저감할 수 있다. PHEVs의 시스템 구조와 제어기 구성에 따라 분류된 다양한 배터리 충전 기법들을 비교분석하고 시뮬레이션 결과를 통해 검증한다.

• 스마트미디어저널
• /
• 제5권1호
• /
• pp.30-37
• /
• 2016

In recent decades, the study of human brain function has dramatically increased thanks to the advent of Functional Magnetic Resonance Imaging. This is a powerful tool which provides a deep view of the activities of the brain. From fMRI data, the neuroscientists analyze which parts of the brain have responsibility for a particular action and finding the common pattern representing each state involved in these tasks. This is one of the most challenges in neuroscience area because of noisy, sparsity of data as well as the differences of anatomical brain structure of each person. In this paper, we propose the use of appropriate discriminant methods, such as Fisher Discriminant Ratio and hypothesis testing, together with strong boosting ability of Adaboost classifier. We prove that discriminant methods are effective in classifying cognitive states. The experiment results show significant better accuracy than previous works. We also show that it is possible to train a successful classifier without prior anatomical knowledge and use only a small number of features.

• Journal of Information Processing Systems
• /
• 제13권4호
• /
• pp.834-851
• /
• 2017

The amount of sources of information available on the web using ontologies as support continues to increase and is often heterogeneous and distributed. Ontology alignment is the solution to ensure semantic interoperability. In this paper, we describe a new ontology alignment approach, which consists of combining structure-based and reasoning-based approaches in order to discover new semantic correspondences between entities of different ontologies. We used the biblio test of the benchmark series and anatomy series of the Ontology Alignment Evaluation Initiative (OAEI) 2012 evaluation campaign to evaluate the performance of our approach. We compared our approach successively with LogMap and YAM++ systems. We also analyzed the contribution of our method compared to structural and semantic methods. The results obtained show that our performance provides good performance. Indeed, these results are better than those of the LogMap system in terms of precision, recall, and F-measure. Our approach has also been proven to be more relevant than YAM++ for certain types of ontologies and significantly improves the structure-based and reasoningbased methods.

• 산경연구논집
• /
• 제14권5호
• /
• pp.11-18
• /
• 2023

Purpose: The current research based on the comprehensive literature evaluation aims to gain insight into the factors contributing to an Oriental art gallery's success and the strategies used to advertise and sell these works effectively. Understanding and experience in Oriental art are essential for finding solutions to these issues. Research design, data and methodology: The current research conducted the following stages to conduct a thorough literature analysis on the issues that plague Oriental art gallery practitioners and the methods used to increase sales of this kind of art: Finding Valuable Resources and Subjects, Screening and Selection of Articles, Data Extraction and Analysis, Synthesis of Findings. Results: After reviewing the many aspects that affect the success of a gallery specializing in Oriental art, there were four key approaches that have emerged for boosting sales of this kind of artwork. Based on the findings, these approaches are grounded in four areas: consumer preferences; marketing methods; pricing strategies; and art investments. Conclusions: All in all, the current study finally indicates that practitioners should consider cultural background, age, gender, income, and level of education when developing marketing strategies and selecting artwork to exhibit. Target marketing is an effective method for attracting and retaining customers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권6호
• /
• pp.1540-1561
• /
• 2024

This research proposes a novel approach for Tamil Handwritten Character Recognition (THCR) that combines feature selection and ensemble learning techniques. The Tamil script is complex and highly variable, requiring a robust and accurate recognition system. Feature selection is used to reduce dimensionality while preserving discriminative features, improving classification performance and reducing computational complexity. Several feature selection methods are compared, and individual classifiers (support vector machines, neural networks, and decision trees) are evaluated through extensive experiments. Ensemble learning techniques such as bagging, and boosting are employed to leverage the strengths of multiple classifiers and enhance recognition accuracy. The proposed approach is evaluated on the HP Labs Dataset, achieving an impressive 95.56% accuracy using an ensemble learning framework based on support vector machines. The dataset consists of 82,928 samples with 247 distinct classes, contributed by 500 participants from Tamil Nadu. It includes 40,000 characters with 500 user variations. The results surpass or rival existing methods, demonstrating the effectiveness of the approach. The research also offers insights for developing advanced recognition systems for other complex scripts. Future investigations could explore the integration of deep learning techniques and the extension of the proposed approach to other Indic scripts and languages, advancing the field of handwritten character recognition.

• 한국초등과학교육학회지:초등과학교육
• /
• 제25권1호
• /
• pp.96-108
• /
• 2006

The purpose of this study was to investigate how prospective elementary school teachers perceived teacher's inquiryoriented teaching practice, with an emphasis on students' scientific explanations based on scientific evidence. For this study, 94 prospective elementary school teachers were participated. 14 among 94 participants had chances to intensively experience this particular teaching methods for 15 weeks. All of the 94 participants observed the intended science teaching practice for 4th graders in two different elementary schools, which utilized the science talks emphasizing students' scientific explanation activity. For quantitative data analysis, they were asked to provide their reaction to the science teaching methods after their classroom observation. For qualitative data analysis, 5 among the participants, who had relatively long term experience with this teaching practice, were chosen to interview in order to understand their individual reasons of the ways they perceived about the inquiry-oriented teaching methods boosting students' scientific explanation. The results show that the prospective elementary teachers generally thought the emphasis of students' scientific explanation based on scientific evidence could enhance young elementary students' science content understanding, stimulate their curiosity/interests, and further develop their ability to engage actively in scientific discussions. However, some prospective teachers tended to think that the science teaching. methods would not be effective in terms of managing science classes, though. This study concludes that the prospective teachers tended to hold an endemic dilemma. On the one hand, they had their clear preference to the inquiry-oriented teaching practice as the most ideal teaching methods. On the other hand, they also had their persistent hesitance in using these methods due to their fear that elementary students might not adequately grasp the important science content when engaged in scientific discourse through an inquiry-oriented class.