• Journal of Intelligence and Information Systems
• /
• v.22 no.3
• /
• pp.129-142
• /
• 2016

Customer product reviews have become one of the important factors for purchase decision makings. Customers believe that reviews written by others who have already had an experience with the product offer more reliable information than that provided by sellers. However, there are too many products and reviews, the advantage of e-commerce can be overwhelmed by increasing search costs. Reading all of the reviews to find out the pros and cons of a certain product can be exhausting. To help users find the most useful information about products without much difficulty, e-commerce companies try to provide various ways for customers to write and rate product reviews. To assist potential customers, online stores have devised various ways to provide useful customer reviews. Different methods have been developed to classify and recommend useful reviews to customers, primarily using feedback provided by customers about the helpfulness of reviews. Most shopping websites provide customer reviews and offer the following information: the average preference of a product, the number of customers who have participated in preference voting, and preference distribution. Most information on the helpfulness of product reviews is collected through a voting system. Amazon.com asks customers whether a review on a certain product is helpful, and it places the most helpful favorable and the most helpful critical review at the top of the list of product reviews. Some companies also predict the usefulness of a review based on certain attributes including length, author(s), and the words used, publishing only reviews that are likely to be useful. Text mining approaches have been used for classifying useful reviews in advance. To apply a text mining approach based on all reviews for a product, we need to build a term-document matrix. We have to extract all words from reviews and build a matrix with the number of occurrences of a term in a review. Since there are many reviews, the size of term-document matrix is so large. It caused difficulties to apply text mining algorithms with the large term-document matrix. Thus, researchers need to delete some terms in terms of sparsity since sparse words have little effects on classifications or predictions. The purpose of this study is to suggest a better way of building term-document matrix by deleting useless terms for review classification. In this study, we propose neutrality index to select words to be deleted. Many words still appear in both classifications - useful and not useful - and these words have little or negative effects on classification performances. Thus, we defined these words as neutral terms and deleted neutral terms which are appeared in both classifications similarly. After deleting sparse words, we selected words to be deleted in terms of neutrality. We tested our approach with Amazon.com's review data from five different product categories: Cellphones & Accessories, Movies & TV program, Automotive, CDs & Vinyl, Clothing, Shoes & Jewelry. We used reviews which got greater than four votes by users and 60% of the ratio of useful votes among total votes is the threshold to classify useful and not-useful reviews. We randomly selected 1,500 useful reviews and 1,500 not-useful reviews for each product category. And then we applied Information Gain and Support Vector Machine algorithms to classify the reviews and compared the classification performances in terms of precision, recall, and F-measure. Though the performances vary according to product categories and data sets, deleting terms with sparsity and neutrality showed the best performances in terms of F-measure for the two classification algorithms. However, deleting terms with sparsity only showed the best performances in terms of Recall for Information Gain and using all terms showed the best performances in terms of precision for SVM. Thus, it needs to be careful for selecting term deleting methods and classification algorithms based on data sets.

• Journal of Intelligence and Information Systems
• /
• v.17 no.1
• /
• pp.111-125
• /
• 2011

Clustering is a process of grouping similar or relevant documents into a cluster and assigning a meaningful concept to the cluster. By this process, clustering facilitates fast and correct search for the relevant documents by narrowing down the range of searching only to the collection of documents belonging to related clusters. For effective clustering, techniques are required for identifying similar documents and grouping them into a cluster, and discovering a concept that is most relevant to the cluster. One of the problems often appearing in this context is the detection of a complex concept that overlaps with several simple concepts at the same hierarchical level. Previous clustering methods were unable to identify and represent a complex concept that belongs to several different clusters at the same level in the concept hierarchy, and also could not validate the semantic hierarchical relationship between a complex concept and each of simple concepts. In order to solve these problems, this paper proposes a new clustering method that identifies and represents complex concepts efficiently. We developed the Hierarchical Overlapping Clustering (HOC) algorithm that modified the traditional Agglomerative Hierarchical Clustering algorithm to allow overlapped clusters at the same level in the concept hierarchy. The HOC algorithm represents the clustering result not by a tree but by a lattice to detect complex concepts. We developed a system that employs the HOC algorithm to carry out the goal of complex concept detection. This system operates in three phases; 1) the preprocessing of documents, 2) the clustering using the HOC algorithm, and 3) the validation of semantic hierarchical relationships among the concepts in the lattice obtained as a result of clustering. The preprocessing phase represents the documents as x-y coordinate values in a 2-dimensional space by considering the weights of terms appearing in the documents. First, it goes through some refinement process by applying stopwords removal and stemming to extract index terms. Then, each index term is assigned a TF-IDF weight value and the x-y coordinate value for each document is determined by combining the TF-IDF values of the terms in it. The clustering phase uses the HOC algorithm in which the similarity between the documents is calculated by applying the Euclidean distance method. Initially, a cluster is generated for each document by grouping those documents that are closest to it. Then, the distance between any two clusters is measured, grouping the closest clusters as a new cluster. This process is repeated until the root cluster is generated. In the validation phase, the feature selection method is applied to validate the appropriateness of the cluster concepts built by the HOC algorithm to see if they have meaningful hierarchical relationships. Feature selection is a method of extracting key features from a document by identifying and assigning weight values to important and representative terms in the document. In order to correctly select key features, a method is needed to determine how each term contributes to the class of the document. Among several methods achieving this goal, this paper adopted the $x^2$�� statistics, which measures the dependency degree of a term t to a class c, and represents the relationship between t and c by a numerical value. To demonstrate the effectiveness of the HOC algorithm, a series of performance evaluation is carried out by using a well-known Reuter-21578 news collection. The result of performance evaluation showed that the HOC algorithm greatly contributes to detecting and producing complex concepts by generating the concept hierarchy in a lattice structure.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.16 no.4
• /
• pp.180-195
• /
• 2011

A balanced trophic model for Sinjido marine ecosystem was constructed using ECOPATH model and data obtained 1994 in the region. The model integrates available information on biomass and food spectrum, and analyses ecosystem properties, dynamics of the main species populations and the key trophic pathways of the system, and then compares these results with those of other marine environments. The model comprises 17 groups of benthic algae, phytoplankton, zooplankton, gastropoda, polychaeta, bivalvia, echinodermata, crustacean, cephalopoda, goby, flatfish, rays and skates, croaker, blenny, conger, flatheads, and detritus. The model shows trophic levels of 1.0~4.0 from primary producers and detritus to top predator as flathead group. The model estimates total biomass(B) of 0.1 $kgWW/m^2$, total net primary production(PP) of 1.6 $kgWW/m^2/yr$, total system throughput(TST) of 3.4 $kgWW/m^2/yr$ and TST's components of consumption 7%, exports 43%, respiratory flows 4% and flows into detritus 46%. The model also calculates PP/TR of 0.012, PP/B of 0.015, omnivory index(OI) of 0.12, Fin's cycling index(FCI) of 0.7%, Fin's mean path length(MPL) of2.11, ascendancy(A) of 4.1 $kgWW/m^2/yr$ bits, development capacity(C) of 8.2 $kgWW/m^2/yr$ bits and A/C of 51%. In particular this study focuses the analysis of mixed trophic impacts and describes the indirect impact of a groupb upon another through mediating one based on 4 types. A large proportion of total export in TST means higher exchange rate in the study region than in semi enclosed basins, which seems by strong tidal currents along the channels between islands, called Sinjido, Choyakdo and Saengildo. Among ecosystem theory and cycling indices, B, TST, PP/TR, FCI, MPL and OI are shown low, indicating the system is not fully mature according to Odum's theory. Additionally, high A/C reveals the maximum capacity of the region is small. To sum up, the study region has high exports of trophic flow and low capacity to develop, and reaches a development stage in the moment. This is a pilot research applied to the Sinjido in terms of trophic flow and food web system such that it may be helpful for comparison and management of the ecosystem in the future.

• The Journal of Korean Society of Virology
• /
• v.27 no.2
• /
• pp.227-237
• /
• 1997

In order to search for anti-Herpes simplex virus (HSV) type-1 agents from Korean medicinal herbs and Korean traditional prescriptions (herb complexes), we selected 80 medicinal herbs and 45 prescriptions, based on a review of the Korean traditional medicine books. Both methanol extracts and boiling-water extracts were tested by means of the MTT assay (tetrazolium based colorimetric assay). Ten of the 125 methanol extracts: CM-11, CM-18, CM-19, CM-21, CM-22, CM-39, MM-3, MM-18, MM-29, MM-73 (see explanation of nomenclature below), showed efficacy against HSV-1. Twelve of the water extracts: CW-2, CW-3-I, CW-3-II, CW-18, MW-3, MW-5 MW-6, MW-12, MW-47, MW-69, MW-73 and MW-79 were active. #3 (individual herb) and #73 (individual herb) were interesting because both water and methanol extracts were active. Especially, #3 is a part of composition of Hong-il-$laksamd{\check{u}}ngbang$ and Hojanghaedokt'ang which have anti-HSV-1 activitives. The SI value of MW-69 and CW-18 was relative high as $10.2{\pm}0.7$ and $11.8{\pm}2.2$. The cytotoxic effect on Vera cells of $Panch'{\check{o}}nch'onch'{\check{o}}ngbang$, Taraxacum platycarpum H. Dahlst. and acycloguanosine was determined by MTT assay. Water extracts of $Panch'{\check{o}}nch'onch'{\check{o}}ngbang$ (prescription) and Taraxacum platycarpum H. Dahlst. showed very weak cytotoxic effects on Vero cells at > $100\;{\mu}g/ml$ but acycloguanosine showed strang cytotoxic effects on Vera cells at > $100\;{\mu}g/ml$. As a result, #3, #73, MW-69 and CW-18 are considered as potentially useful for anti-HSV-1 agent and will be the focus of further research. Abbreviations: CM - methanol extracts of traditional prescriptions; CW - water extracts of traditional prescriptions; MM - methanol extracts of individual herbs; MW - water extracts of individual herbs.

• The Korean Journal of Financial Studies
• /
• v.3 no.1
• /
• pp.233-265
• /
• 1996

We examine the impact of public and private information on price movements using the thirty DJIA stocks and twenty-one NASDAQ stocks. We find that the standard deviation of daily returns on information days (dividend announcement, earnings announcement, insider purchase, or insider sale) is much higher than on no-information days. Both public information matters at the NYSE, probably due to masked identification of insiders. Earnings announcement has the greatest impact for both DJIA and NASDAQ stocks, and there is some evidence of positive impact of insider asle on return volatility of NASDAQ stocks. There has been considerable debate, e.g., French and Roll (1986), over whether market volatility is due to public information or private information-the latter gathered through costly search and only revealed through trading. Public information is composed of (1) marketwide public information such as regularly scheduled federal economic announcements (e.g., employment, GNP, leading indicators) and (2) company-specific public information such as dividend and earnings announcements. Policy makers and corporate insiders have a better access to marketwide private information (e.g., a new monetary policy decision made in the Federal Reserve Board meeting) and company-specific private information, respectively, compated to the general public. Ederington and Lee (1993) show that marketwide public information accounts for most of the observed volatility patterns in interest rate and foreign exchange futures markets. Company-specific public information is explored by Patell and Wolfson (1984) and Jennings and Starks (1985). They show that dividend and earnings announcements induce higher than normal volatility in equity prices. Kyle (1985), Admati and Pfleiderer (1988), Barclay, Litzenberger and Warner (1990), Foster and Viswanathan (1990), Back (1992), and Barclay and Warner (1993) show that the private information help by informed traders and revealed through trading influences market volatility. Cornell and Sirri (1992)' and Meulbroek (1992) investigate the actual insider trading activities in a tender offer case and the prosecuted illegal trading cased, respectively. This paper examines the aggregate and individual impact of marketwide information, company-specific public information, and company-specific private information on equity prices. Specifically, we use the thirty common stocks in the Dow Jones Industrial Average (DJIA) and twenty one National Association of Securities Dealers Automated Quotations (NASDAQ) common stocks to examine how their prices react to information. Marketwide information (public and private) is estimated by the movement in the Standard and Poors (S & P) 500 Index price for the DJIA stocks and the movement in the NASDAQ Composite Index price for the NASDAQ stocks. Divedend and earnings announcements are used as a subset of company-specific public information. The trading activity of corporate insiders (major corporate officers, members of the board of directors, and owners of at least 10 percent of any equity class) with an access to private information can be cannot legally trade on private information. Therefore, most insider transactions are not necessarily based on private information. Nevertheless, we hypothesize that market participants observe how insiders trade in order to infer any information that they cannot possess because insiders tend to buy (sell) when they have good (bad) information about their company. For example, Damodaran and Liu (1993) show that insiders of real estate investment trusts buy (sell) after they receive favorable (unfavorable) appraisal news before the information in these appraisals is released to the public. Price discovery in a competitive multiple-dealership market (NASDAQ) would be different from that in a monopolistic specialist system (NYSE). Consequently, we hypothesize that NASDAQ stocks are affected more by private information (or more precisely, insider trading) than the DJIA stocks. In the next section, we describe our choices of the fifty-one stocks and the public and private information set. We also discuss institutional differences between the NYSE and the NASDAQ market. In Section II, we examine the implications of public and private information for the volatility of daily returns of each stock. In Section III, we turn to the question of the relative importance of individual elements of our information set. Further analysis of the five DJIA stocks and the four NASDAQ stocks that are most sensitive to earnings announcements is given in Section IV, and our results are summarized in Section V.

• Sijohaknonchong
• /
• v.44
• /
• pp.105-138
• /
• 2016

Recently with the research function, "XML Digital collection of Sijo Texts in the Period of Modern Enlightenment" DB data is being provided through the Korean Research Memory (http://www.krm.or.kr) and the foundation for the constructing the contents of Sijo Texts in the Period of Modern Enlightenment has been laid. In this paper, by reviewing the characteristics and problems of Digital collection of Sijo Texts in the Period of Modern Enlightenment and searching for the improvement, I tried to find a way to make it into the content. This database has the primary meaning in the integrating and glancing at the vast amounts of Sijo in the Period of Modern Enlightenment to reaching 12,500 pieces. In addition, it is the first Sijo data base which is provide the variety of search features according to literature, name of poet, title of work, original text, per period, and etc. However, this database has the limits to verifying the overall aspects of the Sijo in the Period of Modern Enlightenment. The title and original text, which is written in the archaic word or Chinese character, could not be searched, because the standard type text of modern language is not formatted. And also the works and the individual Sijo works released after 1945 were missing in the database. It is inconvenient to extract the datum according to the poet, because poets are marked in the various ways such as one's real name, nom de plume and etc. To solve this kind of problems and improve the utilization of the database, I proposed the providing the standard type text of modern language, giving the index terms about content, providing the information on the work format and etc. Furthermore, if the Sijo database in the Period of Modern Enlightenment which is prepared the character of the Sijo Culture Information System could be built, it could be connected with the academic, educational contents. For the specific plan, I suggested as follow, - learning support materials for the Modern history and the national territory recognition on the Modern Age - source materials for studying indigenous animals and plants characters creating the commercial characters - applicability as the Sijo learning tool such as Sijo Game.

• Journal of KIISE:Databases
• /
• v.30 no.4
• /
• pp.381-396
• /
• 2003

Subsequence matching is an operation that finds subsequences whose changing patterns are similar to a given query sequence from time-series databases. This paper points out the performance bottleneck in subsequence matching, and then proposes an effective method that improves the performance of entire subsequence matching significantly by resolving the performance bottleneck. First, we analyze the disk access and CPU processing times required during the index searching and post processing steps through preliminary experiments. Based on their results, we show that the post processing step is the main performance bottleneck in subsequence matching, and them claim that its optimization is a crucial issue overlooked in previous approaches. In order to resolve the performance bottleneck, we propose a simple but quite effective method that processes the post processing step in the optimal way. By rearranging the order of candidate subsequences to be compared with a query sequence, our method completely eliminates the redundancy of disk accesses and CPU processing occurred in the post processing step. We formally prove that our method is optimal and also does not incur any false dismissal. We show the effectiveness of our method by extensive experiments. The results show that our method achieves significant speed-up in the post processing step 3.91 to 9.42 times when using a data set of real-world stock sequences and 4.97 to 5.61 times when using data sets of a large volume of synthetic sequences. Also, the results show that our method reduces the weight of the post processing step in entire subsequence matching from about 90% to less than 70%. This implies that our method successfully resolves th performance bottleneck in subsequence matching. As a result, our method provides excellent performance in entire subsequence matching. The experimental results reveal that it is 3.05 to 5.60 times faster when using a data set of real-world stock sequences and 3.68 to 4.21 times faster when using data sets of a large volume of synthetic sequences compared with the previous one.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.48 no.4
• /
• pp.373-383
• /
• 2022

Inflammation caused by active oxygen and the resulting barrier damage have been consistently pointed out as the cause of wrinkle formation. In this study, effective index ingredient search and efficacy analysis were performed to verify the value of use as a functional cosmetic material related to antioxidant, anti-inflammatory and skin barrier improvement, and anti-aging for extracts of four types of Eleutherococcus divaricatus var. chiisanensis (ED), Eleutherococcus senticosus (EN), Eleutherococcus sessiliflorus (ES), and Eleutherococcus sieboldianus (EI) belonging to the Eleutherococcus genus. To identify the effective index composition, the content of the ingredients was measured by high-performance liquid chromatography. The content of eleutheroside E and chlorogenic acid was the highest in ED among the Eleutherococcus genus. As for anti-oxidant activity, DPPH radical scavenging activity was the highest in ED. In anti-inflammatory effects, ED extracts inhibited nitric oxide generation in inflammatory macrophage cells due to lipopolysaccharide by 40% at 100 ㎍/mL. In the case of IL-6 inhibition, which is known as a pro-inflammatory cytokine, ED showed 41% inhibition at 100 ㎍/mL. In addition, filaggrin and involucrin, which are skin barrier-related factors, were increased by 2.5 times and 1.6 times, respectively, in 100 ㎍/mL of ED extracts, and as for the collagenase, which is a wrinkle-related factor, ED extract showed 29% efficacy at 100 ㎍/mL. Thus, these result suggested that ED extract, among the four Eleutherococcus genus, can be used as a cosmetic ingredient for suppressing inflammation in the skin, reinforcing the skin barrier, and reducing wrinkles.

• The Journal of Internal Korean Medicine
• /
• v.44 no.4
• /
• pp.726-740
• /
• 2023

Objectives: This review investigated the research on warm herbal foot bath therapy (WHFT) for insomnia. Methods: A search was conducted on the China National Knowledge Infrastructure (CNKI) database to collect relevant studies published up to August 29, 2023. Randomized controlled trials (RCTs) comparing WHFT and sleeping pills in patients with insomnia were included. The methodological quality of the included studies was assessed using the Cochrane risk-of-bias assessment tool. The results of the meta-analysis were presented as risk ratios (RRs) or mean differences (MDs) and their 95% confidence intervals (CIs). Results: A total of 11 RCTs were included. WHFT as monotherapy resulted in a significantly higher total effective rate (TER) (RR, 1.25; 95% CI, 1.15 to 1.36; I²=25%) and an improved Pittsburgh Sleep Quality Index (PSQI) global sore (MD, -3.10; 95% CI, -4.24 to -1.95; I²=73%) compared to benzodiazepines. Additionally, WHFT as a combined therapy with benzodiazepines resulted in a significantly higher TER (RR, 1.15; 95% CI, 1.04 to 1.27; I²=0%) and an improved PSQI global score (MD, -2.23; 95% CI, -4.09 to -0.38; I²=80%) compared to benzodiazepines alone. In network analysis visualizing the components of HWFT, four clusters were discovered, and Polygoni Multiflori Ramuls and Ziziphi Spinosae Semen were the key herbs used in WHFT. Overall, the methodological quality of the included studies was poor. Conclusions: There was limited evidence that WHFT as a monotherapy or combined therapy was effective in improving insomnia. The findings can be used as basic data for future WHFT research in South Korea.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.19 no.2
• /
• pp.215-235
• /
• 2024

As artificial intelligence (AI) technology is recognized as a key technology that will determine future national competitiveness, competition for AI technology and industry promotion policies in major countries is intensifying. This study aims to present implications for domestic policy making by analyzing the policies of major countries on the start-up of AI companies, which are the basis of the AI industry ecosystem. The top four countries and the EU for the number of new investment attraction companies in the 2023 AI Index announced by the HAI Research Institute at Stanford University in the United States were selected, The United States enacted the National AI Initiative Act (NAIIA) in 2021. Through this law, The US Government is promoting continued leadership in the United States in AI R&D, developing reliable AI systems in the public and private sectors, building an AI system ecosystem across society, and strengthening DB management and access to AI policies conducted by all federal agencies. In the 14th Five-Year (2021-2025) Plan and 2035 Long-term Goals held in 2021, China has specified AI as the first of the seven strategic high-tech technologies, and is developing policies aimed at becoming the No. 1 AI global powerhouse by 2030. The UK is investing in innovative R&D companies through the 'Future Fund Breakthrough' in 2021, and is expanding related investments by preparing national strategies to leap forward as AI leaders, such as the implementation plan of the national AI strategy in 2022. Israel is supporting technology investment in start-up companies centered on the Innovation Agency, and the Innovation Agency is leading mid- to long-term investments of 2 to 15 years and regulatory reforms for new technologies. The EU is strengthening its digital innovation hub network and creating the InvestEU (European Strategic Investment Fund) and AI investment fund to support the use of AI by SMEs. This study aims to contribute to analyzing the policies of major foreign countries in making AI company start-up policies and providing a basis for Korea's strategy search. The limitations of the study are the limitations of the countries to be analyzed and the failure to attempt comparative analysis of the policy environments of the countries under the same conditions.