• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.139-156
• /
• 2021

The biggest reason for using a deep learning model in image classification is that it is possible to consider the relationship between each region by extracting each region's features from the overall information of the image. However, the CNN model may not be suitable for emotional image data without the image's regional features. To solve the difficulty of classifying emotion images, many researchers each year propose a CNN-based architecture suitable for emotion images. Studies on the relationship between color and human emotion were also conducted, and results were derived that different emotions are induced according to color. In studies using deep learning, there have been studies that apply color information to image subtraction classification. The case where the image's color information is additionally used than the case where the classification model is trained with only the image improves the accuracy of classifying image emotions. This study proposes two ways to increase the accuracy by incorporating the result value after the model classifies an image's emotion. Both methods improve accuracy by modifying the result value based on statistics using the color of the picture. When performing the test by finding the two-color combinations most distributed for all training data, the two-color combinations most distributed for each test data image were found. The result values were corrected according to the color combination distribution. This method weights the result value obtained after the model classifies an image's emotion by creating an expression based on the log function and the exponential function. Emotion6, classified into six emotions, and Artphoto classified into eight categories were used for the image data. Densenet169, Mnasnet, Resnet101, Resnet152, and Vgg19 architectures were used for the CNN model, and the performance evaluation was compared before and after applying the two-stage learning to the CNN model. Inspired by color psychology, which deals with the relationship between colors and emotions, when creating a model that classifies an image's sentiment, we studied how to improve accuracy by modifying the result values based on color. Sixteen colors were used: red, orange, yellow, green, blue, indigo, purple, turquoise, pink, magenta, brown, gray, silver, gold, white, and black. It has meaning. Using Scikit-learn's Clustering, the seven colors that are primarily distributed in the image are checked. Then, the RGB coordinate values of the colors from the image are compared with the RGB coordinate values of the 16 colors presented in the above data. That is, it was converted to the closest color. Suppose three or more color combinations are selected. In that case, too many color combinations occur, resulting in a problem in which the distribution is scattered, so a situation fewer influences the result value. Therefore, to solve this problem, two-color combinations were found and weighted to the model. Before training, the most distributed color combinations were found for all training data images. The distribution of color combinations for each class was stored in a Python dictionary format to be used during testing. During the test, the two-color combinations that are most distributed for each test data image are found. After that, we checked how the color combinations were distributed in the training data and corrected the result. We devised several equations to weight the result value from the model based on the extracted color as described above. The data set was randomly divided by 80:20, and the model was verified using 20% of the data as a test set. After splitting the remaining 80% of the data into five divisions to perform 5-fold cross-validation, the model was trained five times using different verification datasets. Finally, the performance was checked using the test dataset that was previously separated. Adam was used as the activation function, and the learning rate was set to 0.01. The training was performed as much as 20 epochs, and if the validation loss value did not decrease during five epochs of learning, the experiment was stopped. Early tapping was set to load the model with the best validation loss value. The classification accuracy was better when the extracted information using color properties was used together than the case using only the CNN architecture.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.231-252
• /
• 2021

Artificial intelligence (AI) is a key technology that will change the future the most. It affects the industry as a whole and daily life in various ways. As data availability increases, artificial intelligence finds an optimal solution and infers/predicts through self-learning. Research and investment related to automation that discovers and solves problems on its own are ongoing continuously. Automation of artificial intelligence has benefits such as cost reduction, minimization of human intervention and the difference of human capability. However, there are side effects, such as limiting the artificial intelligence's autonomy and erroneous results due to algorithmic bias. In the labor market, it raises the fear of job replacement. Prior studies on the utilization of artificial intelligence have shown that individuals do not necessarily use the information (or advice) it provides. Algorithm error is more sensitive than human error; so, people avoid algorithms after seeing errors, which is called "algorithm aversion." Recently, artificial intelligence has begun to be understood from the perspective of the augmentation of human intelligence. We have started to be interested in Human-AI collaboration rather than AI alone without human. A study of 1500 companies in various industries found that human-AI collaboration outperformed AI alone. In the medicine area, pathologist-deep learning collaboration dropped the pathologist cancer diagnosis error rate by 85%. Leading AI companies, such as IBM and Microsoft, are starting to adopt the direction of AI as augmented intelligence. Human-AI collaboration is emphasized in the decision-making process, because artificial intelligence is superior in analysis ability based on information. Intuition is a unique human capability so that human-AI collaboration can make optimal decisions. In an environment where change is getting faster and uncertainty increases, the need for artificial intelligence in decision-making will increase. In addition, active discussions are expected on approaches that utilize artificial intelligence for rational decision-making. This study investigates the impact of artificial intelligence on decision-making focuses on human-AI collaboration and the interaction between the decision maker personal traits and advisor type. The advisors were classified into three types: human, artificial intelligence, and human-AI collaboration. We investigated perceived usefulness of advice and the utilization of advice in decision making and whether the decision-maker's personal traits are influencing factors. Three hundred and eleven adult male and female experimenters conducted a task that predicts the age of faces in photos and the results showed that the advisor type does not directly affect the utilization of advice. The decision-maker utilizes it only when they believed advice can improve prediction performance. In the case of human-AI collaboration, decision-makers higher evaluated the perceived usefulness of advice, regardless of the decision maker's personal traits and the advice was more actively utilized. If the type of advisor was artificial intelligence alone, decision-makers who scored high in conscientiousness, high in extroversion, or low in neuroticism, high evaluated the perceived usefulness of the advice so they utilized advice actively. This study has academic significance in that it focuses on human-AI collaboration that the recent growing interest in artificial intelligence roles. It has expanded the relevant research area by considering the role of artificial intelligence as an advisor of decision-making and judgment research, and in aspects of practical significance, suggested views that companies should consider in order to enhance AI capability. To improve the effectiveness of AI-based systems, companies not only must introduce high-performance systems, but also need employees who properly understand digital information presented by AI, and can add non-digital information to make decisions. Moreover, to increase utilization in AI-based systems, task-oriented competencies, such as analytical skills and information technology capabilities, are important. in addition, it is expected that greater performance will be achieved if employee's personal traits are considered.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.111-126
• /
• 2020

Most industries have recently become aware of the importance of customer lifetime value as they are exposed to a competitive environment. As a result, preventing customers from churn is becoming a more important business issue than securing new customers. This is because maintaining churn customers is far more economical than securing new customers, and in fact, the acquisition cost of new customers is known to be five to six times higher than the maintenance cost of churn customers. Also, Companies that effectively prevent customer churn and improve customer retention rates are known to have a positive effect on not only increasing the company's profitability but also improving its brand image by improving customer satisfaction. Predicting customer churn, which had been conducted as a sub-research area for CRM, has recently become more important as a big data-based performance marketing theme due to the development of business machine learning technology. Until now, research on customer churn prediction has been carried out actively in such sectors as the mobile telecommunication industry, the financial industry, the distribution industry, and the game industry, which are highly competitive and urgent to manage churn. In addition, These churn prediction studies were focused on improving the performance of the churn prediction model itself, such as simply comparing the performance of various models, exploring features that are effective in forecasting departures, or developing new ensemble techniques, and were limited in terms of practical utilization because most studies considered the entire customer group as a group and developed a predictive model. As such, the main purpose of the existing related research was to improve the performance of the predictive model itself, and there was a relatively lack of research to improve the overall customer churn prediction process. In fact, customers in the business have different behavior characteristics due to heterogeneous transaction patterns, and the resulting churn rate is different, so it is unreasonable to assume the entire customer as a single customer group. Therefore, it is desirable to segment customers according to customer classification criteria, such as loyalty, and to operate an appropriate churn prediction model individually, in order to carry out effective customer churn predictions in heterogeneous industries. Of course, in some studies, there are studies in which customers are subdivided using clustering techniques and applied a churn prediction model for individual customer groups. Although this process of predicting churn can produce better predictions than a single predict model for the entire customer population, there is still room for improvement in that clustering is a mechanical, exploratory grouping technique that calculates distances based on inputs and does not reflect the strategic intent of an entity such as loyalties. This study proposes a segment-based customer departure prediction process (CCP/2DL: Customer Churn Prediction based on Two-Dimensional Loyalty segmentation) based on two-dimensional customer loyalty, assuming that successful customer churn management can be better done through improvements in the overall process than through the performance of the model itself. CCP/2DL is a series of churn prediction processes that segment two-way, quantitative and qualitative loyalty-based customer, conduct secondary grouping of customer segments according to churn patterns, and then independently apply heterogeneous churn prediction models for each churn pattern group. Performance comparisons were performed with the most commonly applied the General churn prediction process and the Clustering-based churn prediction process to assess the relative excellence of the proposed churn prediction process. The General churn prediction process used in this study refers to the process of predicting a single group of customers simply intended to be predicted as a machine learning model, using the most commonly used churn predicting method. And the Clustering-based churn prediction process is a method of first using clustering techniques to segment customers and implement a churn prediction model for each individual group. In cooperation with a global NGO, the proposed CCP/2DL performance showed better performance than other methodologies for predicting churn. This churn prediction process is not only effective in predicting churn, but can also be a strategic basis for obtaining a variety of customer observations and carrying out other related performance marketing activities.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.149-171
• /
• 2020

The period at university is to make decision about getting an actual job. As our society develops rapidly and highly, jobs are diversified, subdivided, and specialized, and students' job preparation period is also getting longer and longer. This study analyzed the log data of college students to see how the various activities that college students experience inside and outside of school might have influences on employment. For this experiment, students' various activities were systematically classified, recorded as an activity data and were divided into six core competencies (Job reinforcement competency, Leadership & teamwork competency, Globalization competency, Organizational commitment competency, Job exploration competency, and Autonomous implementation competency). The effect of the six competency levels on the employment status (employed group, unemployed group) was analyzed. As a result of the analysis, it was confirmed that the difference in level between the employed group and the unemployed group was significant for all of the six competencies, so it was possible to infer that the activities at the school are significant for employment. Next, in order to analyze the impact of the six competencies on the qualitative performance of employment, we had ANOVA analysis after dividing the each competency level into 2 groups (low and high group), and creating 6 groups by the range of first annual salary. Students with high levels of globalization capability, job search capability, and autonomous implementation capability were also found to belong to a higher annual salary group. The theoretical contributions of this study are as follows. First, it connects the competencies that can be extracted from the school experience with the competencies in the Human Resource Management field and adds job search competencies and autonomous implementation competencies which are required for university students to have their own successful career & life. Second, we have conducted this analysis with the competency data measured form actual activity and result data collected from the interview and research. Third, it analyzed not only quantitative performance (employment rate) but also qualitative performance (annual salary level). The practical use of this study is as follows. First, it can be a guide when establishing career development plans for college students. It is necessary to prepare for a job that can express one's strengths based on an analysis of the world of work and job, rather than having a no-strategy, unbalanced, or accumulating excessive specifications competition. Second, the person in charge of experience design for college students, at an organizations such as schools, businesses, local governments, and governments, can refer to the six competencies suggested in this study to for the user-useful experiences design that may motivate more participation. By doing so, one event may bring mutual benefits for both event designers and students. Third, in the era of digital transformation, the government's policy manager who envisions the balanced development of the country can make a policy in the direction of achieving the curiosity and energy of college students together with the balanced development of the country. A lot of manpower is required to start up novel platform services that have not existed before or to digitize existing analog products, services and corporate culture. The activities of current digital-generation-college-students are not only catalysts in all industries, but also for very benefit and necessary for college students by themselves for their own successful career development.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.1-25
• /
• 2020

Sentiment Analysis (SA) is a Natural Language Processing (NLP) task that analyzes the sentiments consumers or the public feel about an arbitrary object from written texts. Furthermore, Aspect-Based Sentiment Analysis (ABSA) is a fine-grained analysis of the sentiments towards each aspect of an object. Since having a more practical value in terms of business, ABSA is drawing attention from both academic and industrial organizations. When there is a review that says "The restaurant is expensive but the food is really fantastic", for example, the general SA evaluates the overall sentiment towards the 'restaurant' as 'positive', while ABSA identifies the restaurant's aspect 'price' as 'negative' and 'food' aspect as 'positive'. Thus, ABSA enables a more specific and effective marketing strategy. In order to perform ABSA, it is necessary to identify what are the aspect terms or aspect categories included in the text, and judge the sentiments towards them. Accordingly, there exist four main areas in ABSA; aspect term extraction, aspect category detection, Aspect Term Sentiment Classification (ATSC), and Aspect Category Sentiment Classification (ACSC). It is usually conducted by extracting aspect terms and then performing ATSC to analyze sentiments for the given aspect terms, or by extracting aspect categories and then performing ACSC to analyze sentiments for the given aspect category. Here, an aspect category is expressed in one or more aspect terms, or indirectly inferred by other words. In the preceding example sentence, 'price' and 'food' are both aspect categories, and the aspect category 'food' is expressed by the aspect term 'food' included in the review. If the review sentence includes 'pasta', 'steak', or 'grilled chicken special', these can all be aspect terms for the aspect category 'food'. As such, an aspect category referred to by one or more specific aspect terms is called an explicit aspect. On the other hand, the aspect category like 'price', which does not have any specific aspect terms but can be indirectly guessed with an emotional word 'expensive,' is called an implicit aspect. So far, the 'aspect category' has been used to avoid confusion about 'aspect term'. From now on, we will consider 'aspect category' and 'aspect' as the same concept and use the word 'aspect' more for convenience. And one thing to note is that ATSC analyzes the sentiment towards given aspect terms, so it deals only with explicit aspects, and ACSC treats not only explicit aspects but also implicit aspects. This study seeks to find answers to the following issues ignored in the previous studies when applying the BERT pre-trained language model to ACSC and derives superior ACSC models. First, is it more effective to reflect the output vector of tokens for aspect categories than to use only the final output vector of [CLS] token as a classification vector? Second, is there any performance difference between QA (Question Answering) and NLI (Natural Language Inference) types in the sentence-pair configuration of input data? Third, is there any performance difference according to the order of sentence including aspect category in the QA or NLI type sentence-pair configuration of input data? To achieve these research objectives, we implemented 12 ACSC models and conducted experiments on 4 English benchmark datasets. As a result, ACSC models that provide performance beyond the existing studies without expanding the training dataset were derived. In addition, it was found that it is more effective to reflect the output vector of the aspect category token than to use only the output vector for the [CLS] token as a classification vector. It was also found that QA type input generally provides better performance than NLI, and the order of the sentence with the aspect category in QA type is irrelevant with performance. There may be some differences depending on the characteristics of the dataset, but when using NLI type sentence-pair input, placing the sentence containing the aspect category second seems to provide better performance. The new methodology for designing the ACSC model used in this study could be similarly applied to other studies such as ATSC.

• Journal of Intelligence and Information Systems
• /
• v.27 no.2
• /
• pp.55-78
• /
• 2021

Tosolve complex and diverse social problems and ensure the quality of life of individuals, social robots that can interact with humans are attracting attention. In the past, robots were recognized as beings that provide labor force as they put into industrial sites on behalf of humans. However, the concept of today's robot has been extended to social robots that coexist with humans and enable social interaction with the advent of Smart technology, which is considered an important driver in most industries. Specifically, there are service robots that respond to customers, the robots that have the purpose of edutainment, and the emotionalrobots that can interact with humans intimately. However, popularization of robots is not felt despite the current information environment in the modern ICT service environment and the 4th industrial revolution. Considering social interaction with users which is an important function of social robots, not only the technology of the robots but also other factors should be considered. The design elements of the robot are more important than other factors tomake consumers purchase essentially a social robot. In fact, existing studies on social robots are at the level of proposing "robot development methodology" or testing the effects provided by social robots to users in pieces. On the other hand, consumer emotions felt from the robot's appearance has an important influence in the process of forming user's perception, reasoning, evaluation and expectation. Furthermore, it can affect attitude toward robots and good feeling and performance reasoning, etc. Therefore, this study aims to verify the effect of appearance of social robot and consumer emotions on consumer's attitude toward social robot. At this time, a social robot design evaluation model is constructed by combining heterogeneous data from different sources. Specifically, the three quantitative indicator data for the appearance of social robots from the ABOT Database is included in the model. The consumer emotions of social robot design has been collected through (1) the existing design evaluation literature and (2) online buzzsuch as product reviews and blogs, (3) qualitative interviews for social robot design. Later, we collected the score of consumer emotions and attitudes toward various social robots through a large-scale consumer survey. First, we have derived the six major dimensions of consumer emotions for 23 pieces of detailed emotions through dimension reduction methodology. Then, statistical analysis was performed to verify the effect of derived consumer emotionson attitude toward social robots. Finally, the moderated regression analysis was performed to verify the effect of quantitatively collected indicators of social robot appearance on the relationship between consumer emotions and attitudes toward social robots. Interestingly, several significant moderation effects were identified, these effects are visualized with two-way interaction effect to interpret them from multidisciplinary perspectives. This study has theoretical contributions from the perspective of empirically verifying all stages from technical properties to consumer's emotion and attitudes toward social robots by linking the data from heterogeneous sources. It has practical significance that the result helps to develop the design guidelines based on consumer emotions in the design stage of social robot development.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.1-21
• /
• 2021

As the global business environment changes, uncertainties in technology development and market needs increase, and competition among companies intensifies, interests and demands for R&D activities of individual companies are increasing. In order to cope with these environmental changes, R&D companies are strengthening R&D investment as one of the means to enhance the qualitative competitiveness of R&D while paying more attention to facility investment. As a result, facilities or R&D investment elements are inevitably a burden for R&D companies to bear future uncertainties. It is true that the management strategy of increasing investment in R&D as a means of enhancing R&D capability is highly uncertain in terms of corporate performance. In this study, the structural factors that influence the R&D capabilities of companies are explored in terms of technology management capabilities, R&D capabilities, and corporate classification attributes by utilizing data mining techniques, and the characteristics these individual factors present according to the level of R&D capabilities are analyzed. This study also showed cluster analysis and experimental results based on evidence data for all domestic R&D companies, and is expected to provide important implications for corporate management strategies to enhance R&D capabilities of individual companies. For each of the three viewpoints, detailed evaluation indexes were composed of 7, 2, and 4, respectively, to quantitatively measure individual levels in the corresponding area. In the case of technology management capability and R&D capability, the sub-item evaluation indexes that are being used by current domestic technology evaluation agencies were referenced, and the final detailed evaluation index was newly constructed in consideration of whether data could be obtained quantitatively. In the case of corporate classification attributes, the most basic corporate classification profile information is considered. In particular, in order to grasp the homogeneity of the R&D competency level, a comprehensive score for each company was given using detailed evaluation indicators of technology management capability and R&D capability, and the competency level was classified into five grades and compared with the cluster analysis results. In order to give the meaning according to the comparative evaluation between the analyzed cluster and the competency level grade, the clusters with high and low trends in R&D competency level were searched for each cluster. Afterwards, characteristics according to detailed evaluation indicators were analyzed in the cluster. Through this method of conducting research, two groups with high R&D competency and one with low level of R&D competency were analyzed, and the remaining two clusters were similar with almost high incidence. As a result, in this study, individual characteristics according to detailed evaluation indexes were analyzed for two clusters with high competency level and one cluster with low competency level. The implications of the results of this study are that the faster the replacement cycle of professional managers who can effectively respond to changes in technology and market demand, the more likely they will contribute to enhancing R&D capabilities. In the case of a private company, it is necessary to increase the intensity of input of R&D capabilities by enhancing the sense of belonging of R&D personnel to the company through conversion to a corporate company, and to provide the accuracy of responsibility and authority through the organization of the team unit. Since the number of technical commercialization achievements and technology certifications are occurring both in the case of contributing to capacity improvement and in case of not, it was confirmed that there is a limit in reviewing it as an important factor for enhancing R&D capacity from the perspective of management. Lastly, the experience of utility model filing was identified as a factor that has an important influence on R&D capability, and it was confirmed the need to provide motivation to encourage utility model filings in order to enhance R&D capability. As such, the results of this study are expected to provide important implications for corporate management strategies to enhance individual companies' R&D capabilities.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.73-95
• /
• 2021

This study uses Node2vec graph embedding method and Light GBM link prediction to explore undeveloped export candidate countries in Korea's food and beverage industry. Node2vec is the method that improves the limit of the structural equivalence representation of the network, which is known to be relatively weak compared to the existing link prediction method based on the number of common neighbors of the network. Therefore, the method is known to show excellent performance in both community detection and structural equivalence of the network. The vector value obtained by embedding the network in this way operates under the condition of a constant length from an arbitrarily designated starting point node. Therefore, it has the advantage that it is easy to apply the sequence of nodes as an input value to the model for downstream tasks such as Logistic Regression, Support Vector Machine, and Random Forest. Based on these features of the Node2vec graph embedding method, this study applied the above method to the international trade information of the Korean food and beverage industry. Through this, we intend to contribute to creating the effect of extensive margin diversification in Korea in the global value chain relationship of the industry. The optimal predictive model derived from the results of this study recorded a precision of 0.95 and a recall of 0.79, and an F1 score of 0.86, showing excellent performance. This performance was shown to be superior to that of the binary classifier based on Logistic Regression set as the baseline model. In the baseline model, a precision of 0.95 and a recall of 0.73 were recorded, and an F1 score of 0.83 was recorded. In addition, the light GBM-based optimal prediction model derived from this study showed superior performance than the link prediction model of previous studies, which is set as a benchmarking model in this study. The predictive model of the previous study recorded only a recall rate of 0.75, but the proposed model of this study showed better performance which recall rate is 0.79. The difference in the performance of the prediction results between benchmarking model and this study model is due to the model learning strategy. In this study, groups were classified by the trade value scale, and prediction models were trained differently for these groups. Specific methods are (1) a method of randomly masking and learning a model for all trades without setting specific conditions for trade value, (2) arbitrarily masking a part of the trades with an average trade value or higher and using the model method, and (3) a method of arbitrarily masking some of the trades with the top 25% or higher trade value and learning the model. As a result of the experiment, it was confirmed that the performance of the model trained by randomly masking some of the trades with the above-average trade value in this method was the best and appeared stably. It was found that most of the results of potential export candidates for Korea derived through the above model appeared appropriate through additional investigation. Combining the above, this study could suggest the practical utility of the link prediction method applying Node2vec and Light GBM. In addition, useful implications could be derived for weight update strategies that can perform better link prediction while training the model. On the other hand, this study also has policy utility because it is applied to trade transactions that have not been performed much in the research related to link prediction based on graph embedding. The results of this study support a rapid response to changes in the global value chain such as the recent US-China trade conflict or Japan's export regulations, and I think that it has sufficient usefulness as a tool for policy decision-making.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.1-22
• /
• 2021

Aspect Based Sentiment Analysis (ABSA), which analyzes sentiment based on aspects that appear in the text, is drawing attention because it can be used in various business industries. ABSA is a study that analyzes sentiment by aspects for multiple aspects that a text has. It is being studied in various forms depending on the purpose, such as analyzing all targets or just aspects and sentiments. Here, the aspect refers to the property of a target, and the target refers to the text that causes the sentiment. For example, for restaurant reviews, you could set the aspect into food taste, food price, quality of service, mood of the restaurant, etc. Also, if there is a review that says, "The pasta was delicious, but the salad was not," the words "steak" and "salad," which are directly mentioned in the sentence, become the "target." So far, in ABSA, most studies have analyzed sentiment only based on aspects or targets. However, even with the same aspects or targets, sentiment analysis may be inaccurate. Instances would be when aspects or sentiment are divided or when sentiment exists without a target. For example, sentences like, "Pizza and the salad were good, but the steak was disappointing." Although the aspect of this sentence is limited to "food," conflicting sentiments coexist. In addition, in the case of sentences such as "Shrimp was delicious, but the price was extravagant," although the target here is "shrimp," there are opposite sentiments coexisting that are dependent on the aspect. Finally, in sentences like "The food arrived too late and is cold now." there is no target (NULL), but it transmits a negative sentiment toward the aspect "service." Like this, failure to consider both aspects and targets - when sentiment or aspect is divided or when sentiment exists without a target - creates a dual dependency problem. To address this problem, this research analyzes sentiment by considering both aspects and targets (Target-Aspect-Sentiment Detection, hereby TASD). This study detected the limitations of existing research in the field of TASD: local contexts are not fully captured, and the number of epochs and batch size dramatically lowers the F1-score. The current model excels in spotting overall context and relations between each word. However, it struggles with phrases in the local context and is relatively slow when learning. Therefore, this study tries to improve the model's performance. To achieve the objective of this research, we additionally used auxiliary loss in aspect-sentiment classification by constructing CNN(Convolutional Neural Network) layers parallel to existing models. If existing models have analyzed aspect-sentiment through BERT encoding, Pooler, and Linear layers, this research added CNN layer-adaptive average pooling to existing models, and learning was progressed by adding additional loss values for aspect-sentiment to existing loss. In other words, when learning, the auxiliary loss, computed through CNN layers, allowed the local context to be captured more fitted. After learning, the model is designed to do aspect-sentiment analysis through the existing method. To evaluate the performance of this model, two datasets, SemEval-2015 task 12 and SemEval-2016 task 5, were used and the f1-score increased compared to the existing models. When the batch was 8 and epoch was 5, the difference was largest between the F1-score of existing models and this study with 29 and 45, respectively. Even when batch and epoch were adjusted, the F1-scores were higher than the existing models. It can be said that even when the batch and epoch numbers were small, they can be learned effectively compared to the existing models. Therefore, it can be useful in situations where resources are limited. Through this study, aspect-based sentiments can be more accurately analyzed. Through various uses in business, such as development or establishing marketing strategies, both consumers and sellers will be able to make efficient decisions. In addition, it is believed that the model can be fully learned and utilized by small businesses, those that do not have much data, given that they use a pre-training model and recorded a relatively high F1-score even with limited resources.