• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.23-45
• /
• 2020

Big data is creating in a wide variety of fields such as medical care, manufacturing, logistics, sales site, SNS, and the dataset characteristics are also diverse. In order to secure the competitiveness of companies, it is necessary to improve decision-making capacity using a classification algorithm. However, most of them do not have sufficient knowledge on what kind of classification algorithm is appropriate for a specific problem area. In other words, determining which classification algorithm is appropriate depending on the characteristics of the dataset was has been a task that required expertise and effort. This is because the relationship between the characteristics of datasets (called meta-features) and the performance of classification algorithms has not been fully understood. Moreover, there has been little research on meta-features reflecting the characteristics of multi-class. Therefore, the purpose of this study is to empirically analyze whether meta-features of multi-class datasets have a significant effect on the performance of classification algorithms. In this study, meta-features of multi-class datasets were identified into two factors, (the data structure and the data complexity,) and seven representative meta-features were selected. Among those, we included the Herfindahl-Hirschman Index (HHI), originally a market concentration measurement index, in the meta-features to replace IR(Imbalanced Ratio). Also, we developed a new index called Reverse ReLU Silhouette Score into the meta-feature set. Among the UCI Machine Learning Repository data, six representative datasets (Balance Scale, PageBlocks, Car Evaluation, User Knowledge-Modeling, Wine Quality(red), Contraceptive Method Choice) were selected. The class of each dataset was classified by using the classification algorithms (KNN, Logistic Regression, Nave Bayes, Random Forest, and SVM) selected in the study. For each dataset, we applied 10-fold cross validation method. 10% to 100% oversampling method is applied for each fold and meta-features of the dataset is measured. The meta-features selected are HHI, Number of Classes, Number of Features, Entropy, Reverse ReLU Silhouette Score, Nonlinearity of Linear Classifier, Hub Score. F1-score was selected as the dependent variable. As a result, the results of this study showed that the six meta-features including Reverse ReLU Silhouette Score and HHI proposed in this study have a significant effect on the classification performance. (1) The meta-features HHI proposed in this study was significant in the classification performance. (2) The number of variables has a significant effect on the classification performance, unlike the number of classes, but it has a positive effect. (3) The number of classes has a negative effect on the performance of classification. (4) Entropy has a significant effect on the performance of classification. (5) The Reverse ReLU Silhouette Score also significantly affects the classification performance at a significant level of 0.01. (6) The nonlinearity of linear classifiers has a significant negative effect on classification performance. In addition, the results of the analysis by the classification algorithms were also consistent. In the regression analysis by classification algorithm, Naïve Bayes algorithm does not have a significant effect on the number of variables unlike other classification algorithms. This study has two theoretical contributions: (1) two new meta-features (HHI, Reverse ReLU Silhouette score) was proved to be significant. (2) The effects of data characteristics on the performance of classification were investigated using meta-features. The practical contribution points (1) can be utilized in the development of classification algorithm recommendation system according to the characteristics of datasets. (2) Many data scientists are often testing by adjusting the parameters of the algorithm to find the optimal algorithm for the situation because the characteristics of the data are different. In this process, excessive waste of resources occurs due to hardware, cost, time, and manpower. This study is expected to be useful for machine learning, data mining researchers, practitioners, and machine learning-based system developers. The composition of this study consists of introduction, related research, research model, experiment, conclusion and discussion.

• Asia pacific journal of information systems
• /
• v.21 no.4
• /
• pp.45-62
• /
• 2011

The notion of information systems (IS) continuance has recently emerged as one of the most important research issues in the field of IS. A great deal of research has been conducted thus far on the basis of theories adapted from various disciplines including consumer behaviors and social psychology, in addition to theories regarding information technology (IT) acceptance. This previous body of knowledge provides a robust research framework that can already account for the determination of IS continuance; however, this research points to other, thus-far-unelucidated determinant factors such as habit, which were not included in traditional IT acceptance frameworks, and also re-emphasizes the importance of emotion-related constructs such as satisfaction in addition to conscious intention with rational beliefs such as usefulness. Experiences should also be considered one of the most important factors determining the characteristics of information system (IS) continuance and the features distinct from those determining IS acceptance, because more experienced users may have more opportunities for IS use, which would allow them more frequent use than would be available to less experienced or non-experienced users. Interestingly, experience has dual features that may contradictorily influence IS use. On one hand, attitudes predicated on direct experience have been shown to predict behavior better than attitudes from indirect experience or without experience; as more information is available, direct experience may render IS use a more salient behavior, and may also make IS use more accessible via memory. Therefore, experience may serve to intensify the relationship between IS use and conscious intention with evaluations, On the other hand, experience may culminate in the formation of habits: greater experience may also imply more frequent performance of the behavior, which may lead to the formation of habits, Hence, like experience, users' activation of an IS may be more dependent on habit-that is, unconscious automatic use without deliberation regarding the IS-and less dependent on conscious intentions, Furthermore, experiences can provide basic information necessary for satisfaction with the use of a specific IS, thus spurring the formation of both conscious intentions and unconscious habits, Whereas IT adoption Is a one-time decision, IS continuance may be a series of users' decisions and evaluations based on satisfaction with IS use. Moreover. habits also cannot be formed without satisfaction, even when a behavior is carried out repeatedly. Thus, experiences also play a critical role in satisfaction, as satisfaction is the consequence of direct experiences of actual behaviors. In particular, emotional experiences such as enjoyment can become as influential on IS use as are utilitarian experiences such as usefulness; this is especially true in light of the modern increase in membership-based hedonic systems - including online games, web-based social network services (SNS), blogs, and portals-all of which attempt to provide users with self-fulfilling value. Therefore, in order to understand more clearly the role of experiences in IS continuance, analysis must be conducted under a research framework that includes intentions, habits, and satisfaction, as experience may not only have duration-based moderating effects on the relationship between both intention and habit and the activation of IS use, but may also have content-based positive effects on satisfaction. This is consistent with the basic assumptions regarding the determining factors in IS continuance as suggested by Oritz de Guinea and Markus: consciousness, emotion, and habit. The principal objective of this study was to explore and assess the effects of experiences in IS continuance, with special consideration given to conscious intentions and unconscious habits, as well as satisfaction. IN service of this goal, along with a review of the relevant literature regarding the effects of experiences and habit on continuous IS use, this study suggested a research model that represents the roles of experience: its moderating role in the relationships of IS continuance with both conscious intention and unconscious habit, and its antecedent role in the development of satisfaction. For the validation of this research model. Korean university student users of 'Cyworld', one of the most influential social network services in South Korea, were surveyed, and the data were analyzed via partial least square (PLS) analysis to assess the implications of this study. In result most hypotheses in our research model were statistically supported with the exception of one. Although one hypothesis was not supported, the study's findings provide us with some important implications. First the role of experience in IS continuance differs from its role in IS acceptance. Second, the use of IS was explained by the dynamic balance between habit and intention. Third, the importance of satisfaction was confirmed from the perspective of IS continuance with experience.

• The Journal of Society for e-Business Studies
• /
• v.19 no.4
• /
• pp.231-249
• /
• 2014

We define $21^{st}$ century as an amalgamation of globalization and localization, or Glocalization. Additionally, due to the increasing supply of smart phones and wide usage of social networking services, the ability to utilize such global and regional information has increased a coperation's competitiveness in its market, and even the business models have evolved from the conventional "production and distribution" to E-commerce, through which either a direct or a non-direct transaction is possible. My hypothesis is that the ability to adapt to this trend is possible through transfer of learning, and consequently, this will have an impact on company's performance. Thus, this thesis analyzes the mid- to the long-term impact of such ability and environmental factors on the performance and technology innovation of Korean companies in China. Ultimately, this study intends to engender a basic foundation for a corporation's management strategy in China. Finally this research focuses on those Korean companies in China only and on the proof of influential factors' impact on technological innovation and technological innovation's impact on those corporations' future performances. Section I is an abstract and section II, the case examines the uniqueness and current status of Korean companies in China identifies the concept and the definition of influential factors such as education-orientation, technological innovation, and performance, and then scrutinizes each factors through a closer look at their past researches. Section III explains the thesis model, the survey's method and target, the thesis, variable factors, the content, and the method of analysis. In section IV, the thesis is proved based on the outcome of the survey. The result in Section V highlights the high comprehension of technological innovation: both education-orientation and technological innovation prove to have a positive (+) correlation with the performance. The vision on education orientation proves to have a positive (+) influence on technological innovation. The vision on education-orientation and technological innovation prove to have a positive (+) influence individually on company's performance.

• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.143-159
• /
• 2015

Predicting IT trends has been a long and important subject for information systems research. IT trend prediction makes it possible to acknowledge emerging eras of innovation and allocate budgets to prepare against rapidly changing technological trends. Towards the end of each year, various domestic and global organizations predict and announce IT trends for the following year. For example, Gartner Predicts 10 top IT trend during the next year, and these predictions affect IT and industry leaders and organization's basic assumptions about technology and the future of IT, but the accuracy of these reports are difficult to verify. Social media data can be useful tool to verify the accuracy. As social media services have gained in popularity, it is used in a variety of ways, from posting about personal daily life to keeping up to date with news and trends. In the recent years, rates of social media activity in Korea have reached unprecedented levels. Hundreds of millions of users now participate in online social networks and communicate with colleague and friends their opinions and thoughts. In particular, Twitter is currently the major micro blog service, it has an important function named 'tweets' which is to report their current thoughts and actions, comments on news and engage in discussions. For an analysis on IT trends, we chose Tweet data because not only it produces massive unstructured textual data in real time but also it serves as an influential channel for opinion leading on technology. Previous studies found that the tweet data provides useful information and detects the trend of society effectively, these studies also identifies that Twitter can track the issue faster than the other media, newspapers. Therefore, this study investigates how frequently the predicted IT trends for the following year announced by public organizations are mentioned on social network services like Twitter. IT trend predictions for 2013, announced near the end of 2012 from two domestic organizations, the National IT Industry Promotion Agency (NIPA) and the National Information Society Agency (NIA), were used as a basis for this research. The present study analyzes the Twitter data generated from Seoul (Korea) compared with the predictions of the two organizations to analyze the differences. Thus, Twitter data analysis requires various natural language processing techniques, including the removal of stop words, and noun extraction for processing various unrefined forms of unstructured data. To overcome these challenges, we used SAS IRS (Information Retrieval Studio) developed by SAS to capture the trend in real-time processing big stream datasets of Twitter. The system offers a framework for crawling, normalizing, analyzing, indexing and searching tweet data. As a result, we have crawled the entire Twitter sphere in Seoul area and obtained 21,589 tweets in 2013 to review how frequently the IT trend topics announced by the two organizations were mentioned by the people in Seoul. The results shows that most IT trend predicted by NIPA and NIA were all frequently mentioned in Twitter except some topics such as 'new types of security threat', 'green IT', 'next generation semiconductor' since these topics non generalized compound words so they can be mentioned in Twitter with other words. To answer whether the IT trend tweets from Korea is related to the following year's IT trends in real world, we compared Twitter's trending topics with those in Nara Market, Korea's online e-Procurement system which is a nationwide web-based procurement system, dealing with whole procurement process of all public organizations in Korea. The correlation analysis show that Tweet frequencies on IT trending topics predicted by NIPA and NIA are significantly correlated with frequencies on IT topics mentioned in project announcements by Nara market in 2012 and 2013. The main contribution of our research can be found in the following aspects: i) the IT topic predictions announced by NIPA and NIA can provide an effective guideline to IT professionals and researchers in Korea who are looking for verified IT topic trends in the following topic, ii) researchers can use Twitter to get some useful ideas to detect and predict dynamic trends of technological and social issues.

• Information Systems Review
• /
• v.22 no.2
• /
• pp.137-166
• /
• 2020

While South Koreans overseas travelling rate has been increased every year, domestic travelling rate has been at a standstill for several years. The purpose of this study is to analyze domestic traveling styles of Koreans according to their generations in order to provide generation-specific traveling services. For this purpose, we categorized the survey respondents into four different generations, which are Millennium (age 19~34), X generation (35~54), Baby Boomer (55~64) and senior by following the criterions of the Korea National Tourism Organization. After then, we analyze factors related to travel preparation process, the actual traveling activities and satisfaction after the travel. In this study, 16,713 data collected by the Ministry of Culture, Sports and Tourism are used. The results of this study show that Korean people tends to acquire domestic traveling information from their own or acquaintances past experiences. Also, they do not prefer the organized trip for domestic travels, thus do not buy package products a lot. In addition, natural scenery, rich in cultural heritage, and convenient accommodation are the most important determinant factors affecting the overall travel satisfaction of level for all generations. The traveling characteristics for each generation are as follows. Millennium get traveling information from the internet a lot, and more specifically, they refer portal sites and social network services (SNS) in many cases. Also, they tend to travel in summer peak season to popular destinations and pursues active traveling experiences. Generation X has similar traveling patterns with Millennium, however they major transportation method is using their own car. Also, transportation convenience and satisfactory leisure activity are important factors affecting the overall satisfaction level to Generation X. On the other hand, Baby boomer generation has a greater emphasis on appreciation of nature, visiting famous restaurants, and relaxation, rather than actively participating experiencing programs. They travel evenly in summer and spring/fall season to many different areas instead of focusing on popular tourist spots. In addition, shopping and eating delicious food are the important factors affecting the overall satisfaction level for them. Lastly, Senior generation has similar characteristics with Baby boomer in many ways, however, they travel a lot on the same day using public transportations or car rental service. They prefer spring and autumn trips rather than summer peak season, and tend to buy packaged travel products a lot compared with other generations. If these different traveling characteristics of each generation are considered for organizing and customizing tourism services, it is expected that domestic tourism satisfaction level will be ultimately increased.

• Journal of Intelligence and Information Systems
• /
• v.19 no.1
• /
• pp.57-77
• /
• 2013

Social media has become one of the most popular media in web and mobile application. In 2011, social networks and blogs are still the top destination of online users, according to a study from Nielsen Company. In their studies, nearly 4 in 5active users visit social network and blog. Social Networks and Blogs sites rule Americans' Internet time, accounting to 23 percent of time spent online. Facebook is the main social network that the U.S internet users spend time more than the other social network services such as Yahoo, Google, AOL Media Network, Twitter, Linked In and so on. In recent trend, most of the companies promote their products in the Facebook by creating the "Facebook Page" that refers to specific product. The "Like" option allows user to subscribed and received updates their interested on from the page. The film makers which produce a lot of films around the world also take part to market and promote their films by exploiting the advantages of using the "Facebook Page". In addition, a great number of streaming service providers allows users to subscribe their service to watch and enjoy movies and TV program. They can instantly watch movies and TV program over the internet to PCs, Macs and TVs. Netflix alone as the world's leading subscription service have more than 30 million streaming members in the United States, Latin America, the United Kingdom and the Nordics. As the matter of facts, a million of movies and TV program with different of genres are offered to the subscriber. In contrast, users need spend a lot time to find the right movies which are related to their interest genre. Recent years there are many researchers who have been propose a method to improve prediction the rating or preference that would give the most related items such as books, music or movies to the garget user or the group of users that have the same interest in the particular items. One of the most popular methods to build recommendation system is traditional Collaborative Filtering (CF). The method compute the similarity of the target user and other users, which then are cluster in the same interest on items according which items that users have been rated. The method then predicts other items from the same group of users to recommend to a group of users. Moreover, There are many items that need to study for suggesting to users such as books, music, movies, news, videos and so on. However, in this paper we only focus on movie as item to recommend to users. In addition, there are many challenges for CF task. Firstly, the "sparsity problem"; it occurs when user information preference is not enough. The recommendation accuracies result is lower compared to the neighbor who composed with a large amount of ratings. The second problem is "cold-start problem"; it occurs whenever new users or items are added into the system, which each has norating or a few rating. For instance, no personalized predictions can be made for a new user without any ratings on the record. In this research we propose a clustering method according to the users' genre interest extracted from social network service (SNS) and user's movies rating information system to solve the "cold-start problem." Our proposed method will clusters the target user together with the other users by combining the user genre interest and the rating information. It is important to realize a huge amount of interesting and useful user's information from Facebook Graph, we can extract information from the "Facebook Page" which "Like" by them. Moreover, we use the Internet Movie Database(IMDb) as the main dataset. The IMDbis online databases that consist of a large amount of information related to movies, TV programs and including actors. This dataset not only used to provide movie information in our Movie Rating Systems, but also as resources to provide movie genre information which extracted from the "Facebook Page". Formerly, the user must login with their Facebook account to login to the Movie Rating System, at the same time our system will collect the genre interest from the "Facebook Page". We conduct many experiments with other methods to see how our method performs and we also compare to the other methods. First, we compared our proposed method in the case of the normal recommendation to see how our system improves the recommendation result. Then we experiment method in case of cold-start problem. Our experiment show that our method is outperform than the other methods. In these two cases of our experimentation, we see that our proposed method produces better result in case both cases.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.149-169
• /
• 2018

Korea is currently accumulating a large amount of data in public institutions based on the public data open policy and the "Government 3.0". Especially, a lot of data is accumulated in the tourism field. However, the academic discussions utilizing the tourism data are still limited. Moreover, the openness of the data of restaurants, hotels, and online tourism information, and how to use SNS Big Data in tourism are still limited. Therefore, utilization through tourism big data analysis is still low. In this paper, we tried to analyze influencing factors on foreign tourists' satisfaction in Korea through numerical data using data mining technique and R programming technique. In this study, we tried to find ways to revitalize the tourism industry by analyzing about 36,000 big data of the "Survey on the actual situation of foreign tourists from 2013 to 2015" surveyed by the Korea Culture & Tourism Research Institute. To do this, we analyzed the factors that have high influence on the 'Satisfaction', 'Revisit intention', and 'Recommendation' variables of foreign tourists. Furthermore, we analyzed the practical influences of the variables that are mentioned above. As a procedure of this study, we first integrated survey data of foreign tourists conducted by Korea Culture & Tourism Research Institute, which is stored in the tourist information system from 2013 to 2015, and eliminate unnecessary variables that are inconsistent with the research purpose among the integrated data. Some variables were modified to improve the accuracy of the analysis. And we analyzed the factors affecting the dependent variables by using data-mining methods: decision tree(C5.0, CART, CHAID, QUEST), artificial neural network, and logistic regression analysis of SPSS IBM Modeler 16.0. The seven variables that have the greatest effect on each dependent variable were derived. As a result of data analysis, it was found that seven major variables influencing 'overall satisfaction' were sightseeing spot attraction, food satisfaction, accommodation satisfaction, traffic satisfaction, guide service satisfaction, number of visiting places, and country. Variables that had a great influence appeared food satisfaction and sightseeing spot attraction. The seven variables that had the greatest influence on 'revisit intention' were the country, travel motivation, activity, food satisfaction, best activity, guide service satisfaction and sightseeing spot attraction. The most influential variables were food satisfaction and travel motivation for Korean style. Lastly, the seven variables that have the greatest influence on the 'recommendation intention' were the country, sightseeing spot attraction, number of visiting places, food satisfaction, activity, tour guide service satisfaction and cost. And then the variables that had the greatest influence were the country, sightseeing spot attraction, and food satisfaction. In addition, in order to grasp the influence of each independent variables more deeply, we used R programming to identify the influence of independent variables. As a result, it was found that the food satisfaction and sightseeing spot attraction were higher than other variables in overall satisfaction and had a greater effect than other influential variables. Revisit intention had a higher ${\beta}$ value in the travel motive as the purpose of Korean Wave than other variables. It will be necessary to have a policy that will lead to a substantial revisit of tourists by enhancing tourist attractions for the purpose of Korean Wave. Lastly, the recommendation had the same result of satisfaction as the sightseeing spot attraction and food satisfaction have higher ${\beta}$ value than other variables. From this analysis, we found that 'food satisfaction' and 'sightseeing spot attraction' variables were the common factors to influence three dependent variables that are mentioned above('Overall satisfaction', 'Revisit intention' and 'Recommendation'), and that those factors affected the satisfaction of travel in Korea significantly. The purpose of this study is to examine how to activate foreign tourists in Korea through big data analysis. It is expected to be used as basic data for analyzing tourism data and establishing effective tourism policy. It is expected to be used as a material to establish an activation plan that can contribute to tourism development in Korea in the future.

• Journal of Distribution Research
• /
• v.16 no.1
• /
• pp.141-177
• /
• 2011

I. Research Objectives: Most of the previous studies on diffusion have concentrated on efficacy of WOM communication with the use of variables at individual level (Iacobucci 1996; Midgley et al. 1992). However, there is a paucity of studies which investigated network's structural properties as antecedents of WOM from the perspective of consumers' sociocultural propensities. Against this research backbone, this study attempted to link the network's structural properties and consumer' WOM behavior on cross-national basis. The major research objective of this study was to examine the relationship between network properties and WOM by comparing Korean and Chinese consumers. Specific objectives of this research are threefold; firstly, it sought to examine whether network properties (i.e., tie strength, centrality, range) affect WOM (WOM intention and quality of WOM). Secondly, it aimed to explore the moderating effects of cutural orientation (uncertainty avoidance and individuality) on the relationship between network properties and WOM. Thirdly, it substantiates the role of innovativeness as antecedents to both network properties and WOM. II. Research Hypotheses: Based on the above research objectives, the study put forth the following research hypotheses to validate. ${\cdot}$ H 1-1 : The Strength of tie between two counterparts within network will positively influence WOM effectivenes ${\cdot}$ H 1-2 : The network centrality will positively influence the WOM effectiveness ${\cdot}$ H 1-3 : The network range will positively influence the WOM effectiveness ${\cdot}$ H 2-1 : The consumer's uncertainty avoidance tendency will moderate the relationship between network properties and WOM effectiveness ${\cdot}$ H 2-2 : The consumer's individualism tendency will moderate the relationship between network properties and WOM effectiveness ${\cdot}$ H 3-1 : The consumer's innovativeness will positively influence the social network properties ${\cdot}$ H 3-2 : The consumer's innovativeness will positively influence WOM effectiveness III. Methodology: Through a pilot study and back-translation, two versions of questionnaire were prepared, one in Korean and the other in Chinese. The chinese data were collected from the chinese students enrolled in language schools in Suwon city in Korea, while Korean data were collected from students taking classes in a major university in Seoul. A total of 277 questionnaire were used for analysis of Korean data and 212 for Chinese data. The reason why Chinese students living in Korea rather than in China were selected was based on two factors: one was to neutralize the differences (ie, retail channel availability) that may arise from living in separate countries and the second was to minimize the difference in communication venues such as internet accessibility and cell phone usability. SPSS 12.0 and AMOS 7.0 were used for analysis. IV. Results: Prior to hypothesis verification, mean differences between the two countries in terms of major constructs were performed with the following result; As for network properties (tie strength, centrality and range), Koreans showed higher scores in all three constructs. For cultural orientation traits, Koreans scored higher only on uncertainty avoidance trait than Chinese. As a result of verifying the first research objective, confirming the relationship between network properties and WOM effectiveness, on Korean side, tie strength(Beta=.116; t=1.785) and centrality (Beta=.499; t=6.776) significantly influenced on WOM intention, and similar finding was obtained for Chinese side, with tie strength (Beta=.246; t=3.544) and centrality (Beta=.247; t=3.538) being significant. However, with regard to WOM argument quality, Korean data yielded only centrality (Beta=.82; t=7.600) having a significant impact on WOM, whereas China showed both tie strength(Beat=.142; t=2.052) and centrality(Beta=.348; t=5.031) being influential. To answer for the second research objective addressing the moderating role of cultural orientation, moderated regression anaylsis was performed and the result showed that uncertainty avoidance moderated between network range and WOM intention for both Korea and China, But for Korea, the uncertainty avoidance moderated between tie strength and WOM quality, while for China it moderated between network range and WOM intention. And innovativeness moderated between tie strength and WOM intention for Korea but it moderated between network range and WOM intention for China. As a result of analysing for third research objective, we found that for Korea, innovativeness positively influenced centrality only (Beta=.546; t=10.808), while for China it influenced both tie strength (Beta=.203; t=2.998) and centrality(Beta=.518; t=8.782). But for both countries alike, the innovativeness influenced positively on WOM (WOM intention and WOM quality). V. Implications: The study yields the two practical implications. Firstly, the result suggests that companies targeting multinational customers need to identify segments which are susceptible to the positive WOM and WOM information based on individual traits such as uncertainty avoidance and individualism and based on that, develop marketing communication strategy. Secondly, the companies need to divide the market on Roger's five innovation stages and based on this information, enforce marketing strategy which utilizes social networking tools such as public media and WOM. For instance, innovator and early adopters, if provided with new product information, will be able to capitalize upon the network advantages and thus add informational value to network operations using SNS or corporate blog.

• 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.