• Korean Journal of Cognitive Science
• /
• v.26 no.3
• /
• pp.263-278
• /
• 2015

Online Q&A for the National Institute of the Korean Language provides expert's answers for questions about the Korean language, in which many similar questions are repeatedly posted like other Q&A boards. So, if a system automatically finds questions that are similar to a user's question, it can immediately provide users with recommendable answers to their question and prevent experts from wasting time to answer to similar questions repeatedly. In this paper, we set 5 classes of questions based on its topic which are frequently asked, and propose to classify questions to those classes. Our system searches similar questions by combining topic similarity, vector similarity and sequence similarity. Experiment shows that our method improves search correctness with topic classification. In experiment, Mean Reciprocal Rank(MRR) of our system is 0.756, and precision for the first result is 68.31% and precision for top five results is 87.32%.

• Journal of Digital Contents Society
• /
• v.19 no.5
• /
• pp.927-933
• /
• 2018

One of the factors that have the greatest influence on the error of the question answering system that finds and provides answers to natural language questions is the step of searching for documents or passages that contain correct answers. In order to improve the retrieval performance, it is necessary to understand the characteristics of documents and passages containing correct answers. This paper experimentally analyzes how many question words appear in the correct answer documents, how the location of the question word is distributed, and how the topic of the question and the correct answer document are similar using the corpus composed of the question, the documents with correct answer, and the documents without correct answer. This study explains the causes of previous search research results for question answer system and discusses the necessary elements of effective search step.

• Korean Journal of Oriental Medicine
• /
• v.11 no.2
• /
• pp.53-66
• /
• 2005

This study is concerned with the patentability and protection of intellectual property rights in Traditional Korea Medicine Paper. The results analyzed significance of patentability by investigated for many kinds of Traditional Korea Medicine Paper. It provide extension of intellectual property rights protection and further research region of TKM field by analysing information of patentability. Recently, In the protection of intellectual property rights, the importance of traditional knowledge resource in many country is increased. It will predict the number of apply for the patent increased annually This study will be provide judging guideline and strategy of intellectual property rights protection by search to detect all existing similar patents in Patent Office (Korea, Japan, U.S.A. EPO) about Traditional Korea Medicine Paper. As a result, It can not be investigated about 33% because of paper research or theoretical study or question investigation etc. But the case of 'The Korea Association of Herbology' and 'The Korean Oriental Medical Ophthalmology & Otolaryngology & Dematology Society' have about 10% rate. If it will be constructed DB system, they will be protected by national treatment.

• Annual Conference on Human and Language Technology
• /
• 2014.10a
• /
• pp.201-205
• /
• 2014

국립국어원의 온라인 가나다 서비스는 한국어에 대한 다양한 질문과 정확한 답변을 제공한다. 만일 새롭게 등록되는 질문에 대해 유사한 질문을 자동으로 찾을 수 있다면, 질문자는 빠른 시간에 답변을 얻을 수 있고 서비스 관리자는 수동 답변 작성의 부담을 덜 수 있다. 본 논문에서는 국립국어원 질의응답게시판의 특성을 분석하여 질문의 주제를 6가지로 분류하고, 주제 분류 정보와 벡터 유사도, 수열 유사도를 결합하여 유사한 질문을 검색하는 시스템을 제안한다. 평가에서는 본 논문에서 제시한 주제 분류 정보를 활용한 결과 1위 정답 검색 정확률이 향상되는 결과를 얻었다. 최종 실험에서는 MRR이 0.62, 정답이 1위, 5위내에 검색될 확률은 각각 54.2%, 78.2%를 보였다.

• Journal of Radiation Industry
• /
• v.8 no.3
• /
• pp.169-176
• /
• 2014

For designing the model of the web program, the demand survey for the technology and information has been performed for the students of the related departments, industrialists and researchers. And, the survey, such as advantages and disadvantages, for the current situations has been examined through comparison and analysis by the establishment type and operational process for the present operating web programs having the similar functions in Korea. The contents and web program for the technology and information system have been also developed by the question investigation and the expert opinions. This system for RI-Biomics has been developed by focusing the convenience for the information provision and the information search as the first constructing direction. Information has been collected by the operator in our institute and making contract with Global Trend Briefing of KISTI in Korea. The information collection in the web program has been designed as the direction regularly provided with RSS. Information has been then analyzed by constructing the expert pool provided from the advisory committee for the technology and information, and using them. The publicity for this web program has been performed by webzines and then it is noted that the publicity programs such as some events should be regularly developed when expanded and advanced to a community in future.

• Korean Security Journal
• /
• no.62
• /
• pp.135-158
• /
• 2020

In casinos, security personnel who manage the safety of customers and employees play a very important role. In particular, there is a high percentage of female employees in casinos, and because the ratio of female and male employees is similar, the probability of female customers or female employees experiencing accidents may be similar to or higher than that of males. Women's security agents who handle women's case accidents can provide female customers and employees with a security service that only women can do. However, most of the agents doing security work at casinos are male, and the proportion of women is very low. Therefore, this research is about employees who are currently working as women in casinos and conducted qualitative research to find out about various experiences they experienced while working in the casino. A total of five study participants were interviewed three times to analyze and categorize the data collected. The first question is the professor's recommendation, his personal information search and his acquaintance's recommendation. The second question, the factors behind the necessary skills at work, are various athletic skills, good physical conditions and foreign language skills. In the third question, the satisfaction factors of the task are the scarcity value of the work, the satisfaction of the pay, the suitability of the individual and the expectation of the future, and the unsatisfactory factors of the work are the risk of the work, the stress on the customer, the discrimination against the sex, the gaze around, the tiredness of the shift work. In the fourth question, factors on the need for female casino security agents are providing differentiated services to female customers, protecting female employees and providing opportunities for women in related majors. The results of this study were interviewed by an expert of more than 20 years in the casino security business, and female casino security agents said that since it is a necessary requirement, they should seek a direction for development through institutional and cognitive improvement.

• Journal of Intelligence and Information Systems
• /
• v.21 no.2
• /
• pp.131-150
• /
• 2015

Omission of noun phrases for obligatory cases is a common phenomenon in sentences of Korean and Japanese, which is not observed in English. When an argument of a predicate can be filled with a noun phrase co-referential with the title, the argument is more easily omitted in Encyclopedia texts. The omitted noun phrase is called a zero anaphor or zero pronoun. Encyclopedias like Wikipedia are major source for information extraction by intelligent application systems such as information retrieval and question answering systems. However, omission of noun phrases makes the quality of information extraction poor. This paper deals with the problem of developing a system that can restore omitted noun phrases in encyclopedia documents. The problem that our system deals with is almost similar to zero anaphora resolution which is one of the important problems in natural language processing. A noun phrase existing in the text that can be used for restoration is called an antecedent. An antecedent must be co-referential with the zero anaphor. While the candidates for the antecedent are only noun phrases in the same text in case of zero anaphora resolution, the title is also a candidate in our problem. In our system, the first stage is in charge of detecting the zero anaphor. In the second stage, antecedent search is carried out by considering the candidates. If antecedent search fails, an attempt made, in the third stage, to use the title as the antecedent. The main characteristic of our system is to make use of a structural SVM for finding the antecedent. The noun phrases in the text that appear before the position of zero anaphor comprise the search space. The main technique used in the methods proposed in previous research works is to perform binary classification for all the noun phrases in the search space. The noun phrase classified to be an antecedent with highest confidence is selected as the antecedent. However, we propose in this paper that antecedent search is viewed as the problem of assigning the antecedent indicator labels to a sequence of noun phrases. In other words, sequence labeling is employed in antecedent search in the text. We are the first to suggest this idea. To perform sequence labeling, we suggest to use a structural SVM which receives a sequence of noun phrases as input and returns the sequence of labels as output. An output label takes one of two values: one indicating that the corresponding noun phrase is the antecedent and the other indicating that it is not. The structural SVM we used is based on the modified Pegasos algorithm which exploits a subgradient descent methodology used for optimization problems. To train and test our system we selected a set of Wikipedia texts and constructed the annotated corpus in which gold-standard answers are provided such as zero anaphors and their possible antecedents. Training examples are prepared using the annotated corpus and used to train the SVMs and test the system. For zero anaphor detection, sentences are parsed by a syntactic analyzer and subject or object cases omitted are identified. Thus performance of our system is dependent on that of the syntactic analyzer, which is a limitation of our system. When an antecedent is not found in the text, our system tries to use the title to restore the zero anaphor. This is based on binary classification using the regular SVM. The experiment showed that our system's performance is F1 = 68.58%. This means that state-of-the-art system can be developed with our technique. It is expected that future work that enables the system to utilize semantic information can lead to a significant performance improvement.

• Journal of the Korean School Mathematics Society
• /
• v.6 no.1
• /
• pp.135-161
• /
• 2003

This research is how students and teacher apprehend mathematics education, pointing out problem areas as a basis on how to improve students understanding of mathematics through improved guidance by teachers in the future. 1107 high school students and 105 teachers from around Daejeon and Choongnam province were surveyed and the results were as follows. 1. 77 %( 852) of students viewed the "application of problem solving methods" as understanding mathematic problems. 2. Replies to the question on understanding the study of mathematics resulted in 85.7% of teachers saying "it is the understanding of the basic concept to which you solve the problems" 3. For questions relating to the large difference in-class mathematics achievements and mock University entrance exam achievements, students' response that "for in-class tests you only have to learn problems with similar form but the mock tests are not like that" pointed out the problem in the area of mathematics education. 4. For future mathematic education teachers will have to "explain better and more completely the basic principles and concepts before solving problems" , and make an effort to stimulate students by "creating a more fun atmosphere" . There will also be the need to prevent as much as possible, the use of "formula or memory driven problems" and encourage students to initiate problem solving for themselves.; and encourage students to initiate problem solving for themselves.

• Journal of Dental Anesthesia and Pain Medicine
• /
• v.21 no.4
• /
• pp.283-309
• /
• 2021

Achieving profound anesthesia in mandibular molars with irreversible pulpitis is a tedious task. This review aimed at evaluating the success of buccal/lingual infiltrations administered with a primary inferior alveolar nerve block (IANB) injection or as a supplemental injection after the failure of the primary injection in symptomatic and asymptomatic patients with irreversible pulpitis in human mandibular molars. The review question was "What will be the success of primary and supplemental infiltration injection in the endodontic treatment of patients with irreversible pulpitis in human mandibular molars?" We searched electronic databases, including Pubmed, Scopus, and Ebsco host and we did a comprehensive manual search. The review protocol was framed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist. We included clinical studies that evaluated and compared the anesthetic outcomes of primary IANB with primary and/or supplementary infiltration injections. Standard evaluation of the included studies was performed and suitable data and inferences were assessed. Twenty-six studies were included, of which 13 were selected for the meta-analysis. In the forest plot representation of the studies evaluating infiltrations, the combined risk ratio (RR) was 1.88 (95% CI: 1.49, 2.37), in favor of the secondary infiltrations with a statistical heterogeneity of 77%. The forest plot analysis for studies comparing primary IANB + infiltration versus primary IANB alone showed a low heterogeneity (0%). The included studies had similar RRs and the combined RR was 1.84 (95% CI: 1.44, 2.34). These findings suggest that supplemental infiltrations given along with a primary IANB provide a better success rate. L'Abbe plots were generated to measure the statistical heterogeneity among the studies. Trial sequential analysis suggested that the number of patients included in the analysis was adequate. Based on the qualitative and quantitative analyses, we concluded that the infiltration technique, either as a primary injection or as a supplementary injection, given after the failure of primary IANB, increases the overall anesthetic efficacy.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.129-164
• /
• 2023

In a situation where the use and introduction of Large Language Models (LLMs) is expanding due to recent developments in generative AI technology, it is difficult to find actual application cases or implementation methods for the use of internal company data in existing studies. Accordingly, this study presents a method of implementing generative AI services using the LLM application architecture using the most widely used LangChain framework. To this end, we reviewed various ways to overcome the problem of lack of information, focusing on the use of LLM, and presented specific solutions. To this end, we analyze methods of fine-tuning or direct use of document information and look in detail at the main steps of information storage and retrieval methods using the retrieval augmented generation (RAG) model to solve these problems. In particular, similar context recommendation and Question-Answering (QA) systems were utilized as a method to store and search information in a vector store using the RAG model. In addition, the specific operation method, major implementation steps and cases, including implementation source and user interface were presented to enhance understanding of generative AI technology. This has meaning and value in enabling LLM to be actively utilized in implementing services within companies.