Seonho Kim

Aggregative and stochastic model of main path identification: a case study on graphene

This paper suggests a new method to search main path, as a knowledge trajectory, in the citation network. To enhance the performance and remedy the problems suggested by other researchers for main path analysis (Hummon and Doreian, Social Networks 11(1): 39–63, 1989), we applied two techniques, the aggregative approach and the stochastic approach. The first technique is used to offer improvement of link count methods, such as SPC, SPLC, SPNP, and NPPC, which have a potential problem of making a mistaken picture since they calculate link weights based on a individual topology of a citation link; the other technique, the second-order Markov chains, is used for path dependent search to improve the Hummon and Doreian’s priority first search method. The case study on graphene that tested the performance of our new method showed promising results, assuring us that our new method can be an improved alternative of main path analysis. Our method’s beneficial effects are summed up in eight aspects: (1) path dependent search, (2) basic research search rather than applied research, (3) path merge and split, (4) multiple main paths, (5) backward search for knowledge origin identification, (6) robustness for indiscriminately selected citations, (7) availability in an acyclic network, (8) completely automated search.

A quantitative approach to recommend promising technologies for SME innovation: a case study on knowledge arbitrage from LCD to solar cell

Small and medium-sized enterprises (SMEs) are more important today than in the past, due to their capabilities of creating jobs and boosting the economy. SMEs need continual innovation to survive in a competitive market and to continue growth. But SMEs suffer from the lack of information to generate innovative ideas. The objectives of this study are to suggest a new method to recommend promising technologies to SMEs that need “knowledge arbitrage” and to help SMEs come up with ideas on new R&D. To this end, this study used three analytic techniques: co-word analysis, collaborative filtering, and regression analysis. The suggested method is tested to assure its usefulness by the real case of knowledge arbitrage from LCD to Solar cell. The main contribution of this study is that it is the first to suggest the new method using recommendation algorithm (collaborative filtering) for SMEs’ knowledge arbitrage.

Longitudinal Study-Based Dementia Prediction for Public Health

The issue of public health in Korea has attracted significant attention given the aging of the country’s population, which has created many types of social problems. The approach proposed in this article aims to address dementia, one of the most significant symptoms of aging and a public health care issue in Korea. The Korean National Health Insurance Service Senior Cohort Database contains personal medical data of every citizen in Korea. There are many different medical history patterns between individuals with dementia and normal controls. The approach used in this study involved examination of personal medical history features from personal disease history, sociodemographic data, and personal health examinations to develop a prediction model. The prediction model used a support-vector machine learning technique to perform a 10-fold cross-validation analysis. The experimental results demonstrated promising performance (80.9% F-measure). The proposed approach supported the significant influence of personal medical history features during an optimal observation period. It is anticipated that a biomedical “big data”-based disease prediction model may assist the diagnosis of any disease more correctly.

Building Hierarchical Knowledge Base of Research Interests and Learning Topics for Social Computing Support

본 논문은 연구ㆍ학습 주제 지식베이스를 통한 소셜컴퓨팅 지원에 관한 연구로 두 가지 하부 연구로 구성되었다. 첫 번째 연구는 다양한 학문분야에서 전자 도서관 이용자들의 연구 및 학습 주제를 추출하기 위해 분야별로 분류가 잘 되어 있는 NDLTD Union catalog의 석박사 학위 논문 (Electronic Theses and Dissertations : ETDs)을 분석하여 계층적 지식베이스를 구축하는 연구이다. 석박사 학위 논문 이외에 ACM Transactions 저널의 논문과 컴퓨터 분야 국제 학술대회 웹사이트도 추가로 분석하였는데 이는 컴퓨팅 분야의 보다 세분화된 지식베이스를 얻기 위해서이다. 계층적 지식베이스는 개인화 서비스, 추천시스템, 텍스트 마이닝, 기술기회탐색, 정보 가시화 등의 정보서비스와 소셜컴퓨팅에 유용하게 사용될 수 있다. 본 논문의 두 번째 연구 부분에서는 우리가 만든 계층적 지식기반을 활용하여 4개의 사용자 커뮤니티 마이닝 알고리즘 중에서 우리가 수행중인 소셜 컴퓨팅 연구, 즉 구성원간의 결합도에 기반한 추천시스템,에 최상의 성능을 보이는 그룹핑 알고리즘을 찾는 성능 평가 연구 결과를 제시하였다. 우리는 이 논문을 통해서 우리가 제안하는 연구ㆍ학습 주제 데이터베이스를 사용하는 방법이 기존에 사용자 커뮤니티 마이닝을 위해 사용되던 비용이 많이 필요하고, 느리며, 개인정보 침해의 위험이 있는 인터뷰나 설문에 기반한 방법을 자동화되고, 비용이 적게 들고, 빠르고, 개인정보 침해 위험이 없으며, 반복 수행시에도 일관된 결과를 보여주는 방법으로 대체할 수 있음을 보이고자 한다.

Wave2Vec: Vectorizing Electroencephalography Bio-Signal for Prediction of Brain Disease

Interest in research involving health-medical information analysis based on artificial intelligence, especially for deep learning techniques, has recently been increasing. Most of the research in this field has been focused on searching for new knowledge for predicting and diagnosing disease by revealing the relation between disease and various information features of data. These features are extracted by analyzing various clinical pathology data, such as EHR (electronic health records), and academic literature using the techniques of data analysis, natural language processing, etc. However, still needed are more research and interest in applying the latest advanced artificial intelligence-based data analysis technique to bio-signal data, which are continuous physiological records, such as EEG (electroencephalography) and ECG (electrocardiogram). Unlike the other types of data, applying deep learning to bio-signal data, which is in the form of time series of real numbers, has many issues that need to be resolved in preprocessing, learning, and analysis. Such issues include leaving feature selection, learning parts that are black boxes, difficulties in recognizing and identifying effective features, high computational complexities, etc. In this paper, to solve these issues, we provide an encoding-based Wave2vec time series classifier model, which combines signal-processing and deep learning-based natural language processing techniques. To demonstrate its advantages, we provide the results of three experiments conducted with EEG data of the University of California Irvine, which are a real-world benchmark bio-signal dataset. After converting the bio-signals (in the form of waves), which are a real number time series, into a sequence of symbols or a sequence of wavelet patterns that are converted into symbols, through encoding, the proposed model vectorizes the symbols by learning the sequence using deep learning-based natural language processing. The models of each class can be constructed through learning from the vectorized wavelet patterns and training data. The implemented models can be used for prediction and diagnosis of diseases by classifying the new data. The proposed method enhanced data readability and intuition of feature selection and learning processes by converting the time series of real number data into sequences of symbols. In addition, it facilitates intuitive and easy recognition, and identification of influential patterns. Furthermore, real-time large-capacity data analysis is facilitated, which is essential in the development of real-time analysis diagnosis systems, by drastically reducing the complexity of calculation without deterioration of analysis performance by data simplification through the encoding process.