• Journal of KIISE
• /
• v.44 no.11
• /
• pp.1138-1148
• /
• 2017

Topic extraction is a technology that automatically extracts a set of topics from a set of documents, and this has been a major research topic in the area of natural language processing. Representative topic extraction methods include Latent Dirichlet Allocation (LDA) and word clustering-based methods. However, there are problems with these methods, such as repeated topics and mixed topics. The problem of repeated topics is one in which a specific topic is extracted as several topics, while the problem of mixed topic is one in which several topics are mixed in a single extracted topic. To solve these problems, this study proposes a method to extract topics using an LDA that is robust against the problem of repeated topic, going through the steps of separating and merging the topics using the similarity between words to correct the extracted topics. As a result of the experiment, the proposed method showed better performance than the conventional LDA method.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.4
• /
• pp.113-127
• /
• 2021

Topic modeling has been receiving much attention in academic disciplines in recent years. Topic modeling is one of the applications in machine learning and natural language processing. It is a statistical modeling procedure to discover topics in the collection of documents. Recently, there have been many attempts to find out topics in diverse fields of academic research. Although the first Department of Industrial Engineering (I.E.) was established in Hanyang university in 1958, Korean Institute of Industrial Engineers (KIIE) which is truly the most academic society was first founded to contribute to research for I.E. and promote industrial techniques in 1974. Korean Society of Industrial and Systems Engineering (KSIE) was established four years later. However, the research topics for KSIE journal have not been deeply examined up until now. Using topic modeling algorithms, we cautiously aim to detect the research topics of KSIE journal for the first half of the society history, from 1978 to 1999. We made use of titles and abstracts in research papers to find out topics in KSIE journal by conducting four algorithms, LSA, HDP, LDA, and LDA Mallet. Topic analysis results obtained by the algorithms were compared. We tried to show the whole procedure of topic analysis in detail for further practical use in future. We employed visualization techniques by using analysis result obtained from LDA. As a result of thorough analysis of topic modeling, eight major research topics were discovered including Production/Logistics/Inventory, Reliability, Quality, Probability/Statistics, Management Engineering/Industry, Engineering Economy, Human Factor/Safety/Computer/Information Technology, and Heuristics/Optimization.

• Journal of the Korean Society for information Management
• /
• v.39 no.4
• /
• pp.23-48
• /
• 2022

The purpose of this study is to investigate the flow of domestic information literacy education research using keyword network analysis and topic modeling and to explore the direction of information literacy education in the future. For this reason, 306 academic papers related to information literacy education published in academic journals of the library and information science field in Korea were chosen. And through the preprocessing process for abstracts of the paper, total keyword appearance frequency, keyword appearance frequency by period, and keyword simultaneous occurrence frequency were analyzed. Subsequently, keyword network analysis analyzed the degree centrality, between centrality, and eigenvector centrality of keywords. Using structural topic modeling analysis, 15 topics -curriculum, information literacy effect, contents of information literacy education, school library education, information media literacy, information literacy ability evaluation index, library anxiety, public library program, health information literacy ability, digital divide, library assisted instruction improvement, research trend, information literacy model, and teacher role-were derived. In addition, the trend of topics by year was analyzed to confirm the change in relative weight by topic. Based on these results, the direction of information literacy education and the suggestions for follow-up research were presented.

• Journal of Internet Computing and Services
• /
• v.22 no.5
• /
• pp.47-55
• /
• 2021

Analysis of research trends in specific subject areas is performed by examining related topics and subject changes by using topic modeling techniques through keyword extraction for most of the literature information (paper, patents, etc.). Unlike existing research methods, this paper extracts topics related to the research topic using the LDA topic modeling technique for the project information of national R&D projects provided by the National Science and Technology Knowledge Information Service (NTIS) in the field of artificial intelligence. By analyzing these topics, this study aims to analyze research topics and investment directions for national R&D projects. NTIS provides a vast amount of national R&D information, from information on tasks carried out through national R&D projects to research results (thesis, patents, etc.) generated through research. In this paper, the search results were confirmed by performing artificial intelligence keywords and related classification searches in NTIS integrated search, and basic data was constructed by downloading the latest three-year project information. Using the LDA topic modeling library provided by Python, related topics and keywords were extracted and analyzed for basic data (research goals, research content, expected effects, keywords, etc.) to derive insights on the direction of research investment.

• Journal of Information Processing Systems
• /
• v.14 no.1
• /
• pp.79-100
• /
• 2018

Twitter is one of the most popular social platforms for online users to share trendy information and views on any event. Twitter reports an event faster than any other medium and contains enormous information and views regarding an event. Consequently, Twitter topic summarization is one of the most convenient ways to get instant gist of any event. However, the information shared on Twitter is often full of nonstandard abbreviations, acronyms, out of vocabulary (OOV) words and with grammatical mistakes which create challenges to find reliable and useful information related to any event. Undoubtedly, Twitter event summarization is a challenging task where traditional text summarization methods do not work well. In last decade, various research works introduced different approaches for automatic Twitter topic summarization. The main aim of this survey work is to make a broad overview of promising summarization approaches on a Twitter topic. We also focus on automatic evaluation of summarization techniques by surveying recent evaluation methodologies. At the end of the survey, we emphasize on both current and future research challenges in this domain through a level of depth analysis of the most recent summarization approaches.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.488-497
• /
• 2004

This paper describes the hybrid document summarization using the indicative summarization and the query-based summarization. The learning models are built from teaming documents in order to extract topic phrases. We use Naive Bayesian, Decision Tree and Supported Vector Machine as the machine learning algorithm. The system extracts topic phrases automatically from new document based on these models and outputs the summary of the document using query-based summarization which considers the extracted topic phrases as queries and calculates the locality-based similarity of each topic phrase. We examine how the topic phrases affect the summarization and how many phrases are proper to summarization. Then, we evaluate the extracted summary by comparing with manual summary, and we also compare our summarization system with summarization mettled from MS-Word.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.1
• /
• pp.13-18
• /
• 2019

In this study, we investigate important keywords and their relationships among the keywords for social issues, and analyze topics to find subjects of the social issues. In particular, we collected twitter data with the keyword 'metoo' which has attracted much attention in these days, and perform keyword analysis and topic modeling. First, we preprocess the twitter data, identified important keywords, and analyzed the relatedness of the keywords. After then, topic modeling is performed to find subjects related to 'metoo'. Our experimental results showed that relatedness of keywords and subjects on social issues in twitter are well identified based on keyword analysis and topic modeling.

• Journal of odor and indoor environment
• /
• v.17 no.4
• /
• pp.322-329
• /
• 2018

The objective of this study is to identify the research trend in the field of indoor environment in Korea. We collected 419 papers published in the Journal of the Korean Society for indoor environment between 2004 and 2018, and attempted to produce datasets using a topic modeling technique, Latent Dirichlet Allocation(LDA). The result of topic modeling showed that 8 topics ("VOCs investigation", "Subway environment", "Building thermal environment", "School health", "Building particulate matter", "Asbestos risk", "Radon risk", "Air cleaner and treatment") could be extracted using Gibbs sampling method. In terms of topic trends, investigation of volatile organic compounds, subway environment, school health, and building particulate matter showed a decreasing tendency, while the building thermal environment, asbestos risk, radon risk, air cleaners, and air treatment showed an increasing tendency. The results of this topic modeling could help us to understand current trends related indoor environment, and provide valuable information in developing future research and policy frameworks.

• The Journal of Economics, Marketing and Management
• /
• v.9 no.1
• /
• pp.1-14
• /
• 2021

Purpose: The article deals with the proposition that consumers' fashion consumption behavior will still follow the consumption behavior of fast fashion, despite recognizing the importance of slow fashion. Research design, data and methodology: The research model to verify this proposition is topic modelling with big data including unstructured textual data. we combined 5,506 news articles posted on Naver news search platform during the 2003-2019 period about fast fashion and slow fashion, high-frequency words have been derived, and topics have been found using LDA model. Based on these, we examined consumers' perception and consumption behavior on slow fashion through the analysis of Topic Network. Results: (1) Looking at the status of annual article collection, consumers' interest in slow fashion mainly began in 2005 and showed a steady increase up to 2019. (2) Term Frequency analysis showed that the keywords for slow fashion are the lowest, with consumers' consumption patterns continuing around 'brand.' (3) Each topic's weight in articles showed that 'social value' - which includes slow fashion - ranked sixth among the 9 topics, low linkage with other topics. (4) Lastly, 'brand' and 'fashion trend' were key topics, and the topic 'social value' accounted for a low proportion. Conclusion: Slow fashion was not a considerable factor of consumption behavior. Consumption patterns in fashion sector are still dominated by general consumption patterns centered on brands and fast fashion.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.2
• /
• pp.130-142
• /
• 2008

Different from other ontology languages, TopicMap is capable of integrating numerous amount of heterogenous information resources using the locational information without any information transformation. Although many conventional editors have been developed for topic maps, they are standalone-type only for writing XTM documents. As a result, these tools request too much time for handling large-scale data and provoke practical problems to integrate with legacy systems which are mostly based on relational database. In this paper, we model a large-scale topic map structure based on XTM 1.0 into RDB structure to minimize the processing time and build up the ontology in legacy systems. We implement a topic map platform called X-TOP that can enhance the efficiency of ontology construction and provide interoperability between XTM documents and database. Moreover, we can use conventional SQL tools and other application development tools for topic map construction in X-TOP. The X-TOP is implemented to have 3-tier architecture to support flexible user interfaces and diverse DBMS. This paper shows the usability of X-TOP by means of the comparison with conventional tools and the application to healthcare cancer ontology management.