• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.109-123
• /
• 2012

Many kinds of the education systems are provided to students. Many kinds of the contents like School subjects, license, job training education and so on are provided through many kinds of the media like text, image, video and so on. Students will apply the knowledge they learnt and will use it when they learn other things. In the existing education system, there have been many situations that the education system isn't really helpful to the students because too hard contents are transferred to them or because too easy contents are transferred to them and they learn the contents they already know again. To solve this phenomenon, a method that transfers the most proper lecture contents to the students is suggested in the thesis. Because the difficulty is relative, the contents A can be easier than the contents B to a group of the students and the contents B can be easier than the contents A to another group of the students. Therefore, it is not easy to measure the difficulty of the lecture contents. A method considering this phenomenon to transfer the proper lecture contents is suggested in the thesis. The whole lecture contents are divided into many lecture modules. The students solve the pattern recognition questions, a kind of the prior test questions, before studying the lecture contents and the system selects and provides the most proper lecture module among many lecture modules to the students according to the score about the questions. When the system selects the lecture module and transfer it to the student, the students' answer and the difficulty of the lecture modules are considered. In the existing education system, 1 kind of the content is transferred to various students. If the same lecture contents is transferred to various students, the contents will not be transferred efficiently. The system selects the proper contents using the students' pattern recognition answers. The pattern recognition question is a kind of the prior test question that is developed on the basis of the lecture module and used to recognize whether the student knows the contents of the lecture module. Because the difficulty of the lecture module reflects the all scores of the students' answers, whenever a student submits the answer, the difficulty is changed. The suggested system measures the relative knowledge of the students using the answers and designates the difficulty. The improvement of the suggested method is only applied when the order of the lecture contents has nothing to do with the progress of the lecture. If the contents of the unit 1 should be studied before studying the contents of the unit 2, the suggested method is not applied. The suggested method is introduced on the basis of the subject "English grammar", subjects that the order is not important, in the thesis. If the suggested method is applied properly to the education environment, the students who don't know enough basic knowledge will learn the basic contents well and prepare the basis to learn the harder lecture contents. The students who already know the lecture contents will not study those again and save more time to learn more various lecture contents. Many improvement effects like these and so on will be provided to the education environment. If the suggested method that is introduced on the basis of the subject "English grammar" is applied to the various education systems like primary education, secondary education, job education and so on, more improvement effects will be provided. The direction to realize these things is suggested in the thesis. The suggested method is realized with the MySQL database and Java, JSP program. It will be very good if the suggested method is researched developmentally and become helpful to the development of the Korea education.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.43-61
• /
• 2019

Development of technologies in artificial intelligence has been rapidly increasing with the Fourth Industrial Revolution, and researches related to AI have been actively conducted in a variety of fields such as autonomous vehicles, natural language processing, and robotics. These researches have been focused on solving cognitive problems such as learning and problem solving related to human intelligence from the 1950s. The field of artificial intelligence has achieved more technological advance than ever, due to recent interest in technology and research on various algorithms. The knowledge-based system is a sub-domain of artificial intelligence, and it aims to enable artificial intelligence agents to make decisions by using machine-readable and processible knowledge constructed from complex and informal human knowledge and rules in various fields. A knowledge base is used to optimize information collection, organization, and retrieval, and recently it is used with statistical artificial intelligence such as machine learning. Recently, the purpose of the knowledge base is to express, publish, and share knowledge on the web by describing and connecting web resources such as pages and data. These knowledge bases are used for intelligent processing in various fields of artificial intelligence such as question answering system of the smart speaker. However, building a useful knowledge base is a time-consuming task and still requires a lot of effort of the experts. In recent years, many kinds of research and technologies of knowledge based artificial intelligence use DBpedia that is one of the biggest knowledge base aiming to extract structured content from the various information of Wikipedia. DBpedia contains various information extracted from Wikipedia such as a title, categories, and links, but the most useful knowledge is from infobox of Wikipedia that presents a summary of some unifying aspect created by users. These knowledge are created by the mapping rule between infobox structures and DBpedia ontology schema defined in DBpedia Extraction Framework. In this way, DBpedia can expect high reliability in terms of accuracy of knowledge by using the method of generating knowledge from semi-structured infobox data created by users. However, since only about 50% of all wiki pages contain infobox in Korean Wikipedia, DBpedia has limitations in term of knowledge scalability. This paper proposes a method to extract knowledge from text documents according to the ontology schema using machine learning. In order to demonstrate the appropriateness of this method, we explain a knowledge extraction model according to the DBpedia ontology schema by learning Wikipedia infoboxes. Our knowledge extraction model consists of three steps, document classification as ontology classes, proper sentence classification to extract triples, and value selection and transformation into RDF triple structure. The structure of Wikipedia infobox are defined as infobox templates that provide standardized information across related articles, and DBpedia ontology schema can be mapped these infobox templates. Based on these mapping relations, we classify the input document according to infobox categories which means ontology classes. After determining the classification of the input document, we classify the appropriate sentence according to attributes belonging to the classification. Finally, we extract knowledge from sentences that are classified as appropriate, and we convert knowledge into a form of triples. In order to train models, we generated training data set from Wikipedia dump using a method to add BIO tags to sentences, so we trained about 200 classes and about 2,500 relations for extracting knowledge. Furthermore, we evaluated comparative experiments of CRF and Bi-LSTM-CRF for the knowledge extraction process. Through this proposed process, it is possible to utilize structured knowledge by extracting knowledge according to the ontology schema from text documents. In addition, this methodology can significantly reduce the effort of the experts to construct instances according to the ontology schema.

• Korean Journal of Heritage: History & Science
• /
• v.54 no.2
• /
• pp.78-97
• /
• 2021

Seojangdaeyajodo is a drawing of military night training on February 12th (lunar leap month), 1795. Focusing on the Seojangdaeyajodo, the characteristics and of the costumes worn by various types of officials were examined. There were 34 officials located near King Jeongjo in and around Seojangdae, with 27 Dangsanggwan and 7 Danghagwan. They wore three types of costumes, including armor, yungbok, and military uniforms. All of the twelve armor wearers and the five officials wearing yungbok were dangsanggwan, and the military uniform wearers included eleven dangsanggwan and six danghagwan. For the shape of the armor, the armor relics of General Yeoban, suitable for riding horses, and the armor painting of Muyedobotongji were referenced, and the composition of the armor was based on practicality. The armor consists of a helmet, a suit of armor, a neck guard, armpit guards, arm guards, and a crotch guard. The color of the armor was red and green, which are the most frequently used colors in Seojangdaeyajodo. The composition of yungbok was jurip, navy cheollik, red gwangdahoe, socks made of leather, and suhwaja. The composition of the military uniform was a lined jeolrip, dongdari, jeonbok, yodae, jeondae, and suhwaja. There were differences in the fabrics used in dangsanggwan and danghagwan military uniforms. Dangsanggwan used fabric with depictions of clouds and jewels, and danghagwan used unpatterned fabric. Moreover, jade, gold, and silver were used for detailed ornamental materials in dangsanggwan. The weapons included bows and a bow case, a sword, a rattan stick, wrist straps, and a ggakji. In the records of the King Jeongjo period, various colored heopsu were mentioned; the colors of the dongdari and jeonbok of dangsanggwan and danghagwan were referenced in various colors. It was presented as an illustration of costumes that could be used to produce objects accurately reflecting the above historical results. The basic principle of the illustration was to present the modeling standards for 3D content production. Samples of form, color, and material of the corresponding times and statuses were presented. The front, the side, and the back of each costume and its accessories were presented, and the colors were presented in RGB and CMYK.

• Journal of Christian Education in Korea
• /
• v.71
• /
• pp.277-303
• /
• 2022

The COVID-19 Pandemic, which began in 2020, has led to many changes in the Korean church. It created a situation in which not only the change and form of worship time, but also the definition, direction, and philosophy of ministry had to be re-established. In the early days of COVID-19 Pandemic, the Korean church recognized this as a crisis, but gradually regarded these as opportunities and tried to produce positive results. The Department of Education has also undergone many changes, especially in its summer ministry, and is expected to have undergone more dramatic changes in form, location and method than in any other church event or service. However, no accurate data on this has been collected. Accordingly, Mirae with Dreams (CEO: Pastor Kim Eun-ho), a corporation established by the Oryun Church for the next generation of ministry, conducted a survey on the summer ministry of the Korean church, which has been registered as a future member with dreams every year since 2020 when the COVID-19 fan dummy began. A similar survey was conducted in 2022 following 2021, and 260 churches responded, and the results are as follows. In 2022, the summer ministry of the Ministry of Education of the Korean Church returned to the form before the COVID-19 Pandemic. Unlike 2021, when many of them were held online, more than 81 percent said they had conducted summer camps offline, and 31 percent also conducted or attended outdoor camps. In terms of the importance of roles, when online was also the main focus, parents and teachers were equally viewed or emphasized, while in this summer's survey, 90 percent of respondents said that the role of teachers in charge or department was important. Summer events were mainly summer Bible schools and retreats, but 25% of all respondents said they conducted missionary work and evangelism at home and abroad. Compared to 2021, participation in summer camps has increased in all departments, including infant and kindergarten, elementary and middle school, and especially in infant and middle school. While preparing for the summer camp, most of the respondents said that the focus was on content and topics, and the main focus was on children's accessibility compared to 2021. As a result of synthesizing the description of the reason for the respondents who could not conduct the summer camp, about 40% said they could not conduct the summer camp due to a lack of volunteers. This is more than 30% who pointed out COVID-19 as the cause, which can be seen as an urgent problem to be solved at the Korean church and denomination level. In addition, this paper also mentioned detailed changes in each question, referring to the changes in summer camps from 2020 to 2022.

• Journal of the Korean Society of Earth Science Education
• /
• v.16 no.2
• /
• pp.247-260
• /
• 2023

In this study, when applying the 'geological structure' content element of high school earth science I developed according to the 2015 curriculum to the STEAM program using a web-based expert system using 3D data of Google Earth and drones, the creative problem-solving ability of high school students, attitudes toward STEAM, and the results of this study are as follows. First, after applying the STEAM program, high school students' creative problem-solving ability showed meaningful results at the p<.001 level. Second, STEAM attitudes showed a significant value at the p<.001 level, confirming that they had a positive impact on high school students' attitudes towards STEAM. It was judged that web-based class activities using Google Earth and drones were useful for integrated thinking such as learners' sense of efficacy and value recognition for usefulness of knowledge. High school students' satisfaction with the STEAM program was 3.251, showing a slightly high average. It was confirmed that web-based class activities such as drones and Google Earth had a positive impact on learners' class satisfaction. However, it was interpreted that the lack of time for class activities limited the ability of the learners to increase their interest in class. The proposal of this research is as follows. First of all, in consideration of the production of presentation materials and practical training in the STEAM program, activities such as block time and advance instruction for class understanding before class are necessary. Secondly, in order to revitalize STEAM education in the high school curriculum, we judge that research on the development of various integrated education programs that can be applied to the high school grade system is necessary.

• The Korean Society of Law and Medicine
• /
• v.24 no.4
• /
• pp.67-101
• /
• 2023

The utilization of generative AI in the medical field is also being rapidly researched. Access to vast data sets reduces the time and energy spent in selecting information. However, as the effort put into content creation decreases, there is a greater likelihood of associated issues arising. For example, with generative AI, users must discern the accuracy of results themselves, as these AIs learn from data within a set period and generate outcomes. While the answers may appear plausible, their sources are often unclear, making it challenging to determine their veracity. Additionally, the possibility of presenting results from a biased or distorted perspective cannot be discounted at present on ethical grounds. Despite these concerns, the field of generative AI is continually advancing, with an increasing number of users leveraging it in various sectors, including biomedical and life sciences. This raises important legal considerations regarding who bears responsibility and to what extent for any damages caused by these high-performance AI algorithms. A general overview of issues with generative AI includes those discussed above, but another perspective arises from its fundamental nature as a large-scale language model ('LLM') AI. There is a civil law concern regarding "the memorization of training data within artificial neural networks and its subsequent reproduction". Medical data, by nature, often reflects personal characteristics of patients, potentially leading to issues such as the regeneration of personal information. The extensive application of generative AI in scenarios beyond traditional AI brings forth the possibility of legal challenges that cannot be ignored. Upon examining the technical characteristics of generative AI and focusing on legal issues, especially concerning the protection of personal information, it's evident that current laws regarding personal information protection, particularly in the context of health and medical data utilization, are inadequate. These laws provide processes for anonymizing and de-identification, specific personal information but fall short when generative AI is applied as software in medical devices. To address the functionalities of generative AI in clinical software, a reevaluation and adjustment of existing laws for the protection of personal information are imperative.

• Journal of Global Scholars of Marketing Science
• /
• v.20 no.1
• /
• pp.49-58
• /
• 2010

Corporate social responsibility has become a very important issue for researchers (Greenfield, 2004; Maignan & Ralston, 2002; McWilliams et al., 2006; Pearce & Doh 2005), and many consider it necessary for businesses to define their role in society and apply social and ethical standards to their businesses (Lichtenstein et al., 2004). As a result, a significant number of retailers have adopted CSR as a strategic tool to promote their businesses. To this end, this study sought to discover U.S. consumers' attitudes and behavior in ethical purchasing and consumption based on their subjective perception and evaluation of a retailer. The objectives of this study include: 1) determine the participants awareness of retailers corporate social responsibility; 2) assess how participants evaluate retailers corporate social responsibility; 3) examine whether participants evaluation process of retailers CSR influence their attitude toward the retailer; and 4) assess if participants attitude toward the retailers CSR influence their purchase behavior. This study does not focus on actual retailers' CSR performance because a consumer's decision making process is based on an individual assessment not an actual fact. This study examines US college students' awareness and evaluations of retailers' corporate social responsibility (CSR). Fifty six college students at a major Southeastern university participated in the study. The age of the participants ranged from 18 to 26 years old. Content analysis was conducted with open coding and focused coding. Over 100 single-spaced pages of written responses were collected and analyzed. Two steps of coding (i.e., open coding and focused coding) were conducted (Esterberg, 2002). Coding results and analytic memos were used to understand participants' awareness of CSR and their ethical purchasing behavior supported through the selection and inclusion of direct quotes that were extracted from the written responses. Names used here are pseudonyms to protect confidentiality of participants. Participants were asked to write about retailers, their aware-ness of CSR issues, and to evaluate a retailer's CSR performance. A majority (n = 28) of respondents indicated their awareness of CSR but have not felt the need to act on this issue. Few (n=8) indicated that they are aware of this issue but not greatly concerned. Findings suggest that when college students evaluate retailers' CSR performance, they use three dimensions of CSR: employee support, community support, and environmental support. Employee treatment and support were found as an important criterion in evaluation of retailers' CSR. Respondents indicated that their good experience with a retailer as an employee made them have a positive perception and attitude toward the retailer. Regarding employee support four themes emerged: employee rewards and incentives based on performance, working environment, employee education and training program, and employee and family discounts. Well organized rewards and incentives were mentioned as an important attribute. The factors related to the working environment included: how well retailers follow the rules related to working hours, lunch time and breaks was also one of the most mentioned attributes. Regarding community support, three themes emerged: contributing a percentage of sales to the local community, financial contribution to charity organizations, and events for community support. Regarding environments, two themes emerged: recycling and selling organic or green products. It was mentioned in the responses that retailers are trying to do what they can to be environmentally friendly. One respondent mentioned that the company is creating stores that have an environmentally friendly design. Information about what the company does to help the environment can easily be found on the company’s website as well. Respondents have also noticed that the stores are starting to offer products that are organic and environmentally friendly. A retailer was also mentioned by a respondent in this category in reference to how the company uses eco-friendly cups and how they are helping to rebuild homes in New Orleans. The respondents noticed that a retailer offers reusable bags for their consumers to purchase. One respondent stated that a retailer uses its products to help the environment, through offering organic cotton. After thorough analysis of responses, we found that a participant's evaluation of a retailers' CSR influenced their attitudes towards retailers. However, there was a significant gap between attitudes and purchasing behavior. Although the participants had positive attitudes toward retailers CSR, the lack of funds and time influenced their purchase behavior. Overall, half (n=28) of the respondents mentioned that CSR performance affects their purchasing decisions making when shopping. Findings from this study provide support for retailers to consider their corporate social responsibility when developing their image with the consumer. This study implied that consumers evaluate retailers based on employee, community and environmental support. The evaluation, attitude and purchase behavior of consumers seem to be intertwined. That is, evaluation is based on the knowledge the consumer has of the retailers CSR. That knowledge may influence their attitude toward the retailer and thus influence their purchase behavior. Participants also indicated that having CSR makes them think highly of the retailer, but it does not influence their purchase behavior. Price and convenience seem to surpass the importance of CSR among the participants. Implications, recommendations for future research, and limitations of the study are also discussed.

• The Korean Nurse
• /
• v.28 no.3
• /
• pp.52-67
• /
• 1989

True nursing care means total nursing care which includes physical, emotional and spiritual care. The modern nursing care has tendency to focus toward physical care and needs attention toward emotional and spiritual care. The total nursing care is mandatory for patients with terminal cancer and for this purpose, hospice care became emerged. Hospice case originated from the place or shelter for the travellers to Jerusalem in medieval stage. However, the meaning of modem hospice care became changed to total nursing care for dying patients. Modern hospice care has been developed in England, and spreaded to U.S.A. and Canada for the patients with terminal cancer. Nowaday, it became a part of nursing care and the concept of hospice care extended to the palliative care of the cancer patients. Recently, it was introduced to Korea and received attention as model of total nursing care. This study was attempted to assess the efficacy of hospice care. The purpose of this study was to prove a difference in terms of physical, emotional a d spiritual aspect between the group who received hospice care and who didn't receive hospice care. The subject for this study were 113 patients with advanced cancer who were hospitalized in the S different hospitals. 67 patients received hospice care in 4 different hospitals, and 46 patients didn't receive hospice care in another 4 different hospitals. The method of this study was the questionaire which was made through the descriptive study. The descriptive study was made by individual contact with 102 patients cf advanced cancer for 9 months period. The measurement tool for questionaire was made by author through the descriptive study, and included the personal religious orientation obtained from chung(originated R. Fleck) and 5 emotional stages before dying from Kubler Ross. The content ol questionaire consisted in 67 items which included 11 for general characteristics, 10 for related condition with cancer, 13 for wishes far physical therapy, 13 for emotional reactions and 20 for personal religious orientation. Data for this study was collected from Aug. 25 to Oct. 6 by author and 4 other nurse's who received education and training by author for the collection of data. The collected data were ana lysed using descriptive statistics, $X^2-test$, t-test and pearson correlation coefficient. Results of the study were as follows: "H.C Group" means the group of patient with cancer who received hospice care. "Non H.C Group" means the group of patient with cancer who did not receive hospice care. 1. There is a difference between H.C Group and Non H.C Group in term of the number of physical symptoms, subjective degree of pain sensation and pain control, subjective beliefs in physical cure, emotional reaction, help of present emotional and spiritual care from other personal, needs of emotional and spiritual care in future, selection of treatment method by patients and personal religious orientation. 2. The comparison of H.C Group and Non H.C Group 1) There is no difference in wishes for physical therapy between two groups(p=.522). Among Non H.C Group, a group, who didn't receive traditional therapy and herb medicine was higher than a group who received these in degree of belief that the traditional therapy and herb medicine can cure their disease, and this result was higher in comparison to H.C Group(p=.025, p=.050). 2) Non H.C Group was higher than H.C Group in degree of emotional reaction(p=.050). H.C Group was higher than Non H.C Group in denial and acceptant stage among 5 different emotional stages before dying described by Kubler Ross, especially among the patient who had disease more than 13 months(p=.0069, p=.0198). 3) Non H.C Group was higher than H. C Group in demanding more emotional and spiritual care to doctor, nurse, family and pastor(p=. 010). 4) Non H.C Group was higher than H.C Group in demanding more emotional and spiritual care to each individual of doctor, nurse and family (p=.0110, p=.0029, P=. 0053). 5) H.C Group was higher th2.n Non H.C Group in degree of intrinsic behavior orientation and intrinsic belief orientation of personal religious orientation(p=.034, p=.026). 6) In H.C Group and Non H.C Group, the degree of emotional demanding of christians was significantly higher than non christians to doctor, nurse, family and pastor(p=. 000, p=.035). 7) In H.C Group there were significant positive correlations as following; (1) Between the degree of emotional demandings to doctor, nurse, family & pastor and: the degree of intrinsic behavior orientation in personal religious orientation(r=. 5512, p=.000). (2) Between the degree of emotional demandings to doctor, nurse. family & pastor and the degree of intrinsic belief orientation in personal religious orientation(r=.4795, p=.000). (3) Between the degree of intrinsic behavior orientation and the degree of intrinsic: belief orientation in personal religious orientation(r=.8986, p=.000). (4) Between the degree of extrinsic religious orientation and the degree of consensus religious orientation in personal religious orientation (r=. 2640, p=.015). In H.C. Group there were significant negative correlations as following; (1) Between the degree of intrinsic behavior orientation and extrinsic religious orientation in personal religious orientation (r=-.4218, p=.000). (2) Between the degree or intrinsic behavior orientation and consensus religious orientation in personal religious orientation(r=-. 4597, p=.000). (3) Between the degree of intrinsic belief orientations and the degree of extrinsic religious orientation in personal religious orientation(r=-.4388, p=.000). (4) Between the degree of intrinsic belief orientation and the degree of consensus religious orientation in personal religious orientation(r=-. 5424, p=.000). 8) In Non H.C Group there were significant positive correlation as following; (1) Between the degree of emotional demandings to doctor, nurse, family & pastor and the degree of intrinsic behavior orientation in personal religious orientation(r= .3566, p=.007). (2) Between the degree of emotional demandings to doctor, nurse, family & pastor and the degree of intrinsic belief orientation in personal religious orientation(r=.3430, p=.010). (3) Between the degree of intrinsic behavior orientation and the degree of intrinsic belief orientation in personal religious orientation(r=.9723, p=.000). In Non H.C Group there were significant negative correlation as following; (1) Between the degree of emotional demandings to doctor, nurse, family & pastor and the degree of extrinsic religious orientation in personal religious orientation(r= -.2862, p=.027). (2) Between the degree of intrinsic behavior orientation and the degree of extrinsic religious orientation in personal religious orientation(r=-. 5083, p=.000). (3) Between the degree of intrinsic belief orientation and the degree of extrinsic religious orientation in personal religious orientation(r=-. 5013, p=.000). In conclusion above datas suggest that hospice care provide effective total nursing care for the patients with terminal cancer, and hospice care is mandatory in all medical institutions.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.141-154
• /
• 2019

Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.

• 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.