• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.77-97
• /
• 2019

The large amount of data that emerges from the initial connection environment of the Fourth Industrial Revolution is a major factor that distinguishes the Fourth Industrial Revolution from the existing production environment. This environment has two-sided features that allow it to produce data while using it. And the data produced so produces another value. Due to the massive scale of data, future information systems need to process more data in terms of quantities than existing information systems. In addition, in terms of quality, only a large amount of data, Ability is required. In a small-scale information system, it is possible for a person to accurately understand the system and obtain the necessary information, but in a variety of complex systems where it is difficult to understand the system accurately, it becomes increasingly difficult to acquire the desired information. In other words, more accurate processing of large amounts of data has become a basic condition for future information systems. This problem related to the efficient performance of the information system can be solved by building a semantic web which enables various information processing by expressing the collected data as an ontology that can be understood by not only people but also computers. For example, as in most other organizations, IT has been introduced in the military, and most of the work has been done through information systems. Currently, most of the work is done through information systems. As existing systems contain increasingly large amounts of data, efforts are needed to make the system easier to use through its data utilization. An ontology-based system has a large data semantic network through connection with other systems, and has a wide range of databases that can be utilized, and has the advantage of searching more precisely and quickly through relationships between predefined concepts. In this paper, we propose a defense ontology as a method for effective data management and decision support. In order to judge the applicability and effectiveness of the actual system, we reconstructed the existing air force munitions situation management system as an ontology based system. It is a system constructed to strengthen management and control of logistics situation of commanders and practitioners by providing real - time information on maintenance and distribution situation as it becomes difficult to use complicated logistics information system with large amount of data. Although it is a method to take pre-specified necessary information from the existing logistics system and display it as a web page, it is also difficult to confirm this system except for a few specified items in advance, and it is also time-consuming to extend the additional function if necessary And it is a system composed of category type without search function. Therefore, it has a disadvantage that it can be easily utilized only when the system is well known as in the existing system. The ontology-based logistics situation management system is designed to provide the intuitive visualization of the complex information of the existing logistics information system through the ontology. In order to construct the logistics situation management system through the ontology, And the useful functions such as performance - based logistics support contract management and component dictionary are further identified and included in the ontology. In order to confirm whether the constructed ontology can be used for decision support, it is necessary to implement a meaningful analysis function such as calculation of the utilization rate of the aircraft, inquiry about performance-based military contract. Especially, in contrast to building ontology database in ontology study in the past, in this study, time series data which change value according to time such as the state of aircraft by date are constructed by ontology, and through the constructed ontology, It is confirmed that it is possible to calculate the utilization rate based on various criteria as well as the computable utilization rate. In addition, the data related to performance-based logistics contracts introduced as a new maintenance method of aircraft and other munitions can be inquired into various contents, and it is easy to calculate performance indexes used in performance-based logistics contract through reasoning and functions. Of course, we propose a new performance index that complements the limitations of the currently applied performance indicators, and calculate it through the ontology, confirming the possibility of using the constructed ontology. Finally, it is possible to calculate the failure rate or reliability of each component, including MTBF data of the selected fault-tolerant item based on the actual part consumption performance. The reliability of the mission and the reliability of the system are calculated. In order to confirm the usability of the constructed ontology-based logistics situation management system, the proposed system through the Technology Acceptance Model (TAM), which is a representative model for measuring the acceptability of the technology, is more useful and convenient than the existing system.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.123-139
• /
• 2019

The Korean NPL market was formed by the government and foreign capital shortly after the 1997 IMF crisis. However, this market is short-lived, as the bad debt has started to increase after the global financial crisis in 2009 due to the real economic recession. NPL has become a major investment in the market in recent years when the domestic capital market's investment capital began to enter the NPL market in earnest. Although the domestic NPL market has received considerable attention due to the overheating of the NPL market in recent years, research on the NPL market has been abrupt since the history of capital market investment in the domestic NPL market is short. In addition, decision-making through more scientific and systematic analysis is required due to the decline in profitability and the price fluctuation due to the fluctuation of the real estate business. In this study, we propose a prediction model that can determine the achievement of the benchmark yield by using the NPL market related data in accordance with the market demand. In order to build the model, we used Korean NPL data from December 2013 to December 2017 for about 4 years. The total number of things data was 2291. As independent variables, only the variables related to the dependent variable were selected for the 11 variables that indicate the characteristics of the real estate. In order to select the variables, one to one t-test and logistic regression stepwise and decision tree were performed. Seven independent variables (purchase year, SPC (Special Purpose Company), municipality, appraisal value, purchase cost, OPB (Outstanding Principle Balance), HP (Holding Period)). The dependent variable is a bivariate variable that indicates whether the benchmark rate is reached. This is because the accuracy of the model predicting the binomial variables is higher than the model predicting the continuous variables, and the accuracy of these models is directly related to the effectiveness of the model. In addition, in the case of a special purpose company, whether or not to purchase the property is the main concern. Therefore, whether or not to achieve a certain level of return is enough to make a decision. For the dependent variable, we constructed and compared the predictive model by calculating the dependent variable by adjusting the numerical value to ascertain whether 12%, which is the standard rate of return used in the industry, is a meaningful reference value. As a result, it was found that the hit ratio average of the predictive model constructed using the dependent variable calculated by the 12% standard rate of return was the best at 64.60%. In order to propose an optimal prediction model based on the determined dependent variables and 7 independent variables, we construct a prediction model by applying the five methodologies of discriminant analysis, logistic regression analysis, decision tree, artificial neural network, and genetic algorithm linear model we tried to compare them. To do this, 10 sets of training data and testing data were extracted using 10 fold validation method. After building the model using this data, the hit ratio of each set was averaged and the performance was compared. As a result, the hit ratio average of prediction models constructed by using discriminant analysis, logistic regression model, decision tree, artificial neural network, and genetic algorithm linear model were 64.40%, 65.12%, 63.54%, 67.40%, and 60.51%, respectively. It was confirmed that the model using the artificial neural network is the best. Through this study, it is proved that it is effective to utilize 7 independent variables and artificial neural network prediction model in the future NPL market. The proposed model predicts that the 12% return of new things will be achieved beforehand, which will help the special purpose companies make investment decisions. Furthermore, we anticipate that the NPL market will be liquidated as the transaction proceeds at an appropriate price.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.141-166
• /
• 2019

Recently, channels like social media and SNS create enormous amount of data. In all kinds of data, portions of unstructured data which represented as text data has increased geometrically. But there are some difficulties to check all text data, so it is important to access those data rapidly and grasp key points of text. Due to needs of efficient understanding, many studies about text summarization for handling and using tremendous amounts of text data have been proposed. Especially, a lot of summarization methods using machine learning and artificial intelligence algorithms have been proposed lately to generate summary objectively and effectively which called "automatic summarization". However almost text summarization methods proposed up to date construct summary focused on frequency of contents in original documents. Those summaries have a limitation for contain small-weight subjects that mentioned less in original text. If summaries include contents with only major subject, bias occurs and it causes loss of information so that it is hard to ascertain every subject documents have. To avoid those bias, it is possible to summarize in point of balance between topics document have so all subject in document can be ascertained, but still unbalance of distribution between those subjects remains. To retain balance of subjects in summary, it is necessary to consider proportion of every subject documents originally have and also allocate the portion of subjects equally so that even sentences of minor subjects can be included in summary sufficiently. In this study, we propose "subject-balanced" text summarization method that procure balance between all subjects and minimize omission of low-frequency subjects. For subject-balanced summary, we use two concept of summary evaluation metrics "completeness" and "succinctness". Completeness is the feature that summary should include contents of original documents fully and succinctness means summary has minimum duplication with contents in itself. Proposed method has 3-phases for summarization. First phase is constructing subject term dictionaries. Topic modeling is used for calculating topic-term weight which indicates degrees that each terms are related to each topic. From derived weight, it is possible to figure out highly related terms for every topic and subjects of documents can be found from various topic composed similar meaning terms. And then, few terms are selected which represent subject well. In this method, it is called "seed terms". However, those terms are too small to explain each subject enough, so sufficient similar terms with seed terms are needed for well-constructed subject dictionary. Word2Vec is used for word expansion, finds similar terms with seed terms. Word vectors are created after Word2Vec modeling, and from those vectors, similarity between all terms can be derived by using cosine-similarity. Higher cosine similarity between two terms calculated, higher relationship between two terms defined. So terms that have high similarity values with seed terms for each subjects are selected and filtering those expanded terms subject dictionary is finally constructed. Next phase is allocating subjects to every sentences which original documents have. To grasp contents of all sentences first, frequency analysis is conducted with specific terms that subject dictionaries compose. TF-IDF weight of each subjects are calculated after frequency analysis, and it is possible to figure out how much sentences are explaining about each subjects. However, TF-IDF weight has limitation that the weight can be increased infinitely, so by normalizing TF-IDF weights for every subject sentences have, all values are changed to 0 to 1 values. Then allocating subject for every sentences with maximum TF-IDF weight between all subjects, sentence group are constructed for each subjects finally. Last phase is summary generation parts. Sen2Vec is used to figure out similarity between subject-sentences, and similarity matrix can be formed. By repetitive sentences selecting, it is possible to generate summary that include contents of original documents fully and minimize duplication in summary itself. For evaluation of proposed method, 50,000 reviews of TripAdvisor are used for constructing subject dictionaries and 23,087 reviews are used for generating summary. Also comparison between proposed method summary and frequency-based summary is performed and as a result, it is verified that summary from proposed method can retain balance of all subject more which documents originally have.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.167-194
• /
• 2019

This research starts from the four basic concepts of incentive incompatibility, limited information, myopia and decision variable which are confronted when making decisions in keyword bidding. In order to make these concept concrete, four framework approaches are designed as follows; Strategic approach for the incentive incompatibility, Statistical approach for the limited information, Alternative optimization for myopia, and New model approach for decision variable. The purpose of this research is to propose the statistical optimization model in constructing the portfolio of Sponsored Search Advertising (SSA) in the Sponsor's perspective through empirical tests which can be used in portfolio decision making. Previous research up to date formulates the CTR estimation model using CPC, Rank, Impression, CVR, etc., individually or collectively as the independent variables. However, many of the variables are not controllable in keyword bidding. Only CPC and Rank can be used as decision variables in the bidding system. Classical SSA model is designed on the basic assumption that the CPC is the decision variable and CTR is the response variable. However, this classical model has so many huddles in the estimation of CTR. The main problem is the uncertainty between CPC and Rank. In keyword bid, CPC is continuously fluctuating even at the same Rank. This uncertainty usually raises questions about the credibility of CTR, along with the practical management problems. Sponsors make decisions in keyword bids under the limited information, and the strategic portfolio approach based on statistical models is necessary. In order to solve the problem in Classical SSA model, the New SSA model frame is designed on the basic assumption that Rank is the decision variable. Rank is proposed as the best decision variable in predicting the CTR in many papers. Further, most of the search engine platforms provide the options and algorithms to make it possible to bid with Rank. Sponsors can participate in the keyword bidding with Rank. Therefore, this paper tries to test the validity of this new SSA model and the applicability to construct the optimal portfolio in keyword bidding. Research process is as follows; In order to perform the optimization analysis in constructing the keyword portfolio under the New SSA model, this study proposes the criteria for categorizing the keywords, selects the representing keywords for each category, shows the non-linearity relationship, screens the scenarios for CTR and CPC estimation, selects the best fit model through Goodness-of-Fit (GOF) test, formulates the optimization models, confirms the Spillover effects, and suggests the modified optimization model reflecting Spillover and some strategic recommendations. Tests of Optimization models using these CTR/CPC estimation models are empirically performed with the objective functions of (1) maximizing CTR (CTR optimization model) and of (2) maximizing expected profit reflecting CVR (namely, CVR optimization model). Both of the CTR and CVR optimization test result show that the suggested SSA model confirms the significant improvements and this model is valid in constructing the keyword portfolio using the CTR/CPC estimation models suggested in this study. However, one critical problem is found in the CVR optimization model. Important keywords are excluded from the keyword portfolio due to the myopia of the immediate low profit at present. In order to solve this problem, Markov Chain analysis is carried out and the concept of Core Transit Keyword (CTK) and Expected Opportunity Profit (EOP) are introduced. The Revised CVR Optimization model is proposed and is tested and shows validity in constructing the portfolio. Strategic guidelines and insights are as follows; Brand keywords are usually dominant in almost every aspects of CTR, CVR, the expected profit, etc. Now, it is found that the Generic keywords are the CTK and have the spillover potentials which might increase consumers awareness and lead them to Brand keyword. That's why the Generic keyword should be focused in the keyword bidding. The contribution of the thesis is to propose the novel SSA model based on Rank as decision variable, to propose to manage the keyword portfolio by categories according to the characteristics of keywords, to propose the statistical modelling and managing based on the Rank in constructing the keyword portfolio, and to perform empirical tests and propose a new strategic guidelines to focus on the CTK and to propose the modified CVR optimization objective function reflecting the spillover effect in stead of the previous expected profit models.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.171-196
• /
• 2018

With the development of the web, two-way communication and evaluation became possible and marketing paradigms shifted. In order to meet the needs of consumers, web design trends are continuously responding to consumer feedback. As the web becomes more and more important, both academics and businesses are studying consumer emotions and satisfaction on the web. However, some consumer characteristics are not well considered. Demographic characteristics such as age and sex have been studied extensively, but few studies consider psychological characteristics such as regulatory focus (i.e., emotional regulation). In this study, we analyze the effect of web style on consumer emotion. Many studies analyze the relationship between the web and regulatory focus, but most concentrate on the purpose of web use, particularly motivation and information search, rather than on web style and design. The web communicates with users through visual elements. Because the human brain is influenced by all five senses, both design factors and emotional responses are important in the web environment. Therefore, in this study, we examine the relationship between consumer emotion and satisfaction and web style and design. Previous studies have considered the effects of web layout, structure, and color on emotions. In this study, however, we excluded these web components, in contrast to earlier studies, and analyzed the relationship between consumer satisfaction and emotional indexes of web-style only. To perform this analysis, we collected consumer surveys presenting 40 web style themes to 204 consumers. Each consumer evaluated four themes. The emotional adjectives evaluated by consumers were composed of 18 contrast pairs, and the upper emotional indexes were extracted through factor analysis. The emotional indexes were 'softness,' 'modernity,' 'clearness,' and 'jam.' Hypotheses were established based on the assumption that emotional indexes have different effects on consumer satisfaction. After the analysis, hypotheses 1, 2, and 3 were accepted and hypothesis 4 was rejected. While hypothesis 4 was rejected, its effect on consumer satisfaction was negative, not positive. This means that emotional indexes such as 'softness,' 'modernity,' and 'clearness' have a positive effect on consumer satisfaction. In other words, consumers prefer emotions that are soft, emotional, natural, rounded, dynamic, modern, elaborate, unique, bright, pure, and clear. 'Jam' has a negative effect on consumer satisfaction. It means, consumer prefer the emotion which is empty, plain, and simple. Regulatory focus shows differences in motivation and propensity in various domains. It is important to consider organizational behavior and decision making according to the regulatory focus tendency, and it affects not only political, cultural, ethical judgments and behavior but also broad psychological problems. Regulatory focus also differs from emotional response. Promotion focus responds more strongly to positive emotional responses. On the other hand, prevention focus has a strong response to negative emotions. Web style is a type of service, and consumer satisfaction is affected not only by cognitive evaluation but also by emotion. This emotional response depends on whether the consumer will benefit or harm himself. Therefore, it is necessary to confirm the difference of the consumer's emotional response according to the regulatory focus which is one of the characteristics and viewpoint of the consumers about the web style. After MMR analysis result, hypothesis 5.3 was accepted, and hypothesis 5.4 was rejected. But hypothesis 5.4 supported in the opposite direction to the hypothesis. After validation, we confirmed the mechanism of emotional response according to the tendency of regulatory focus. Using the results, we developed the structure of web-style recommendation system and recommend methods through regulatory focus. We classified the regulatory focus group in to three categories that promotion, grey, prevention. Then, we suggest web-style recommend method along the group. If we further develop this study, we expect that the existing regulatory focus theory can be extended not only to the motivational part but also to the emotional behavioral response according to the regulatory focus tendency. Moreover, we believe that it is possible to recommend web-style according to regulatory focus and emotional desire which consumers most prefer.

• Journal of Food Hygiene and Safety
• /
• v.34 no.1
• /
• pp.22-29
• /
• 2019

Validamycin A is an aminoglycoside fungicide produced by Streptomyces hygroscopicus that inhibits trehalase. The purpose of this study was to develop a method for detecting validamycin A in agricultural samples to establish MRL values for use in Korea. The validamycin A residues in samples were extracted using methanol/water (50/50, v/v) and purified with a hydrophilic-lipophilic balance (HLB) cartridges. The analyte was quantified and confirmed by liquid chromatograph-tandem mass spectrometer (LC-MS/MS) in positive ion mode using multiple reaction monitoring (MRM). Matrix-matched calibration curves were linear over the calibration ranges (0.005~0.5 ng) into a blank extract with $R^2$ > 0.99. The limits of detection and quantification were 0.005 and 0.01 mg/kg, respectively. For validation validamycin A, recovery studies were carried out three different concentration levels (LOQ, $LOQ{\times}10$, $LOQ{\times}50$, n = 5) with five replicates at each level. The average recovery range was from 72.5~118.3%, with relative standard deviation (RSD) less than 10.3%. All values were consistent with the criteria ranges requested in the Codex guidelines (CAC/GL 40-1993, 2003) and the NIFDS (National Institute of Food and Drug Safety) guideline (2016). Therefore, the proposed analytical method is accurate, effective and sensitive for validamycin A determination in agricultural commodities.

• Journal of Food Hygiene and Safety
• /
• v.34 no.1
• /
• pp.40-51
• /
• 2019

Ginseng has a unique production system that is different from those used for other crops. It is subject to the Ginseng Industry Act., requires a long-term cultivation period of 4-6 years, involves complicated cultivation characteristics whereby ginseng is not produced in a single location, and many ginseng farmers engage in mixed-farming. Therefore, to bring the production of Ginseng in line with GAP standards, it is necessary to better understand the on-site practices of Ginseng farmers according to established control points, and to provide a proper action plan for improving efficiency. Among ginseng farmers in Korea who applied for GAP certification, 77.6% obtained it, which is lower than the 94.1% of farmers who obtained certification for other products. 13.7% of the applicants were judged to be unsuitable during document review due to their use of unregistered pesticides and soil heavy metals. Another 8.7% of applicants failed to obtain certification due to inadequate management results. This is a considerably higher rate of failure than the 5.3% incompatibility of document inspection and 0.6% incompatibility of on-site inspection, which suggests that it is relatively more difficult to obtain GAP certification for ginseng farming than for other crops. Ginseng farmers were given an average of 2.65 points out of 10 essential control points and a total 72 control points, which was slightly lower than the 2.81 points obtained for other crops. In particular, ginseng farmers were given an average of 1.96 points in the evaluation of compliance with the safe use standards for pesticides, which was much lower than the average of 2.95 points for other crops. Therefore, it is necessary to train ginseng farmers to comply with the safe use of pesticides. In the other essential control points, the ginseng farmers were rated at an average of 2.33 points, lower than the 2.58 points given for other crops. Several other areas of compliance in which the ginseng farmers also rated low in comparison to other crops were found. These inclued record keeping over 1 year, record of pesticide use, pesticide storages, posts harvest storage management, hand washing before and after work, hygiene related to work clothing, training of workers safety and hygiene, and written plan of hazard management. Also, among the total 72 control points, there are 12 control points (10 required, 2 recommended) that do not apply to ginseng. Therefore, it is considered inappropriate to conduct an effective evaluation of the ginseng production process based on the existing certification standards. In conclusion, differentiated certification standards are needed to expand GAP certification for ginseng farmers, and it is also necessary to develop programs that can be implemented in a more systematic and field-oriented manner to provide the farmers with proper GAP management education.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.111-136
• /
• 2018

In this paper, we propose a methodology to extract answer information about queries from various types of unstructured documents collected from multi-sources existing on web in order to expand knowledge base. The proposed methodology is divided into the following steps. 1) Collect relevant documents from Wikipedia, Naver encyclopedia, and Naver news sources for "subject-predicate" separated queries and classify the proper documents. 2) Determine whether the sentence is suitable for extracting information and derive the confidence. 3) Based on the predicate feature, extract the information in the proper sentence and derive the overall confidence of the information extraction result. In order to evaluate the performance of the information extraction system, we selected 400 queries from the artificial intelligence speaker of SK-Telecom. Compared with the baseline model, it is confirmed that it shows higher performance index than the existing model. The contribution of this study is that we develop a sequence tagging model based on bi-directional LSTM-CRF using the predicate feature of the query, with this we developed a robust model that can maintain high recall performance even in various types of unstructured documents collected from multiple sources. The problem of information extraction for knowledge base extension should take into account heterogeneous characteristics of source-specific document types. The proposed methodology proved to extract information effectively from various types of unstructured documents compared to the baseline model. There is a limitation in previous research that the performance is poor when extracting information about the document type that is different from the training data. In addition, this study can prevent unnecessary information extraction attempts from the documents that do not include the answer information through the process for predicting the suitability of information extraction of documents and sentences before the information extraction step. It is meaningful that we provided a method that precision performance can be maintained even in actual web environment. The information extraction problem for the knowledge base expansion has the characteristic that it can not guarantee whether the document includes the correct answer because it is aimed at the unstructured document existing in the real web. When the question answering is performed on a real web, previous machine reading comprehension studies has a limitation that it shows a low level of precision because it frequently attempts to extract an answer even in a document in which there is no correct answer. The policy that predicts the suitability of document and sentence information extraction is meaningful in that it contributes to maintaining the performance of information extraction even in real web environment. The limitations of this study and future research directions are as follows. First, it is a problem related to data preprocessing. In this study, the unit of knowledge extraction is classified through the morphological analysis based on the open source Konlpy python package, and the information extraction result can be improperly performed because morphological analysis is not performed properly. To enhance the performance of information extraction results, it is necessary to develop an advanced morpheme analyzer. Second, it is a problem of entity ambiguity. The information extraction system of this study can not distinguish the same name that has different intention. If several people with the same name appear in the news, the system may not extract information about the intended query. In future research, it is necessary to take measures to identify the person with the same name. Third, it is a problem of evaluation query data. In this study, we selected 400 of user queries collected from SK Telecom 's interactive artificial intelligent speaker to evaluate the performance of the information extraction system. n this study, we developed evaluation data set using 800 documents (400 questions * 7 articles per question (1 Wikipedia, 3 Naver encyclopedia, 3 Naver news) by judging whether a correct answer is included or not. To ensure the external validity of the study, it is desirable to use more queries to determine the performance of the system. This is a costly activity that must be done manually. Future research needs to evaluate the system for more queries. It is also necessary to develop a Korean benchmark data set of information extraction system for queries from multi-source web documents to build an environment that can evaluate the results more objectively.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.219-240
• /
• 2018

Sentiment analysis, which is one of the text mining techniques, is a method for extracting subjective content embedded in text documents. Recently, the sentiment analysis methods have been widely used in many fields. As good examples, data-driven surveys are based on analyzing the subjectivity of text data posted by users and market researches are conducted by analyzing users' review posts to quantify users' reputation on a target product. The basic method of sentiment analysis is to use sentiment dictionary (or lexicon), a list of sentiment vocabularies with positive, neutral, or negative semantics. In general, the meaning of many sentiment words is likely to be different across domains. For example, a sentiment word, 'sad' indicates negative meaning in many fields but a movie. In order to perform accurate sentiment analysis, we need to build the sentiment dictionary for a given domain. However, such a method of building the sentiment lexicon is time-consuming and various sentiment vocabularies are not included without the use of general-purpose sentiment lexicon. In order to address this problem, several studies have been carried out to construct the sentiment lexicon suitable for a specific domain based on 'OPEN HANGUL' and 'SentiWordNet', which are general-purpose sentiment lexicons. However, OPEN HANGUL is no longer being serviced and SentiWordNet does not work well because of language difference in the process of converting Korean word into English word. There are restrictions on the use of such general-purpose sentiment lexicons as seed data for building the sentiment lexicon for a specific domain. In this article, we construct 'KNU Korean Sentiment Lexicon (KNU-KSL)', a new general-purpose Korean sentiment dictionary that is more advanced than existing general-purpose lexicons. The proposed dictionary, which is a list of domain-independent sentiment words such as 'thank you', 'worthy', and 'impressed', is built to quickly construct the sentiment dictionary for a target domain. Especially, it constructs sentiment vocabularies by analyzing the glosses contained in Standard Korean Language Dictionary (SKLD) by the following procedures: First, we propose a sentiment classification model based on Bidirectional Long Short-Term Memory (Bi-LSTM). Second, the proposed deep learning model automatically classifies each of glosses to either positive or negative meaning. Third, positive words and phrases are extracted from the glosses classified as positive meaning, while negative words and phrases are extracted from the glosses classified as negative meaning. Our experimental results show that the average accuracy of the proposed sentiment classification model is up to 89.45%. In addition, the sentiment dictionary is more extended using various external sources including SentiWordNet, SenticNet, Emotional Verbs, and Sentiment Lexicon 0603. Furthermore, we add sentiment information about frequently used coined words and emoticons that are used mainly on the Web. The KNU-KSL contains a total of 14,843 sentiment vocabularies, each of which is one of 1-grams, 2-grams, phrases, and sentence patterns. Unlike existing sentiment dictionaries, it is composed of words that are not affected by particular domains. The recent trend on sentiment analysis is to use deep learning technique without sentiment dictionaries. The importance of developing sentiment dictionaries is declined gradually. However, one of recent studies shows that the words in the sentiment dictionary can be used as features of deep learning models, resulting in the sentiment analysis performed with higher accuracy (Teng, Z., 2016). This result indicates that the sentiment dictionary is used not only for sentiment analysis but also as features of deep learning models for improving accuracy. The proposed dictionary can be used as a basic data for constructing the sentiment lexicon of a particular domain and as features of deep learning models. It is also useful to automatically and quickly build large training sets for deep learning models.