• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.101-124
• /
• 2018

Recently, most of the technologies have been developed in various forms through the advancement of single technology or interaction with other technologies. Particularly, these technologies have the characteristic of the convergence caused by the interaction between two or more techniques. In addition, efforts in responding to technological changes by advance are continuously increasing through forecasting promising convergence technologies that will emerge in the near future. According to this phenomenon, many researchers are attempting to perform various analyses about forecasting promising convergence technologies. A convergence technology has characteristics of various technologies according to the principle of generation. Therefore, forecasting promising convergence technologies is much more difficult than forecasting general technologies with high growth potential. Nevertheless, some achievements have been confirmed in an attempt to forecasting promising technologies using big data analysis and social network analysis. Studies of convergence technology through data analysis are actively conducted with the theme of discovering new convergence technologies and analyzing their trends. According that, information about new convergence technologies is being provided more abundantly than in the past. However, existing methods in analyzing convergence technology have some limitations. Firstly, most studies deal with convergence technology analyze data through predefined technology classifications. The technologies appearing recently tend to have characteristics of convergence and thus consist of technologies from various fields. In other words, the new convergence technologies may not belong to the defined classification. Therefore, the existing method does not properly reflect the dynamic change of the convergence phenomenon. Secondly, in order to forecast the promising convergence technologies, most of the existing analysis method use the general purpose indicators in process. This method does not fully utilize the specificity of convergence phenomenon. The new convergence technology is highly dependent on the existing technology, which is the origin of that technology. Based on that, it can grow into the independent field or disappear rapidly, according to the change of the dependent technology. In the existing analysis, the potential growth of convergence technology is judged through the traditional indicators designed from the general purpose. However, these indicators do not reflect the principle of convergence. In other words, these indicators do not reflect the characteristics of convergence technology, which brings the meaning of new technologies emerge through two or more mature technologies and grown technologies affect the creation of another technology. Thirdly, previous studies do not provide objective methods for evaluating the accuracy of models in forecasting promising convergence technologies. In the studies of convergence technology, the subject of forecasting promising technologies was relatively insufficient due to the complexity of the field. Therefore, it is difficult to find a method to evaluate the accuracy of the model that forecasting promising convergence technologies. In order to activate the field of forecasting promising convergence technology, it is important to establish a method for objectively verifying and evaluating the accuracy of the model proposed by each study. To overcome these limitations, we propose a new method for analysis of convergence technologies. First of all, through topic modeling, we derive a new technology classification in terms of text content. It reflects the dynamic change of the actual technology market, not the existing fixed classification standard. In addition, we identify the influence relationships between technologies through the topic correspondence weights of each document, and structuralize them into a network. In addition, we devise a centrality indicator (PGC, potential growth centrality) to forecast the future growth of technology by utilizing the centrality information of each technology. It reflects the convergence characteristics of each technology, according to technology maturity and interdependence between technologies. Along with this, we propose a method to evaluate the accuracy of forecasting model by measuring the growth rate of promising technology. It is based on the variation of potential growth centrality by period. In this paper, we conduct experiments with 13,477 patent documents dealing with technical contents to evaluate the performance and practical applicability of the proposed method. As a result, it is confirmed that the forecast model based on a centrality indicator of the proposed method has a maximum forecast accuracy of about 2.88 times higher than the accuracy of the forecast model based on the currently used network indicators.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.119-133
• /
• 2020

Fama asserted that in an efficient market, we can't make a trading rule that consistently outperforms the average stock market returns. This study aims to suggest a machine learning algorithm to improve the trading performance of an intraday short volatility strategy applying asymmetric volatility spillover effect, and analyze its trading performance improvement. Generally stock market volatility has a negative relation with stock market return and the Korean stock market volatility is influenced by the US stock market volatility. This volatility spillover effect is asymmetric. The asymmetric volatility spillover effect refers to the phenomenon that the US stock market volatility up and down differently influence the next day's volatility of the Korean stock market. We collected the S&P 500 index, VIX, KOSPI 200 index, and V-KOSPI 200 from 2008 to 2018. We found the negative relation between the S&P 500 and VIX, and the KOSPI 200 and V-KOSPI 200. We also documented the strong volatility spillover effect from the VIX to the V-KOSPI 200. Interestingly, the asymmetric volatility spillover was also found. Whereas the VIX up is fully reflected in the opening volatility of the V-KOSPI 200, the VIX down influences partially in the opening volatility and its influence lasts to the Korean market close. If the stock market is efficient, there is no reason why there exists the asymmetric volatility spillover effect. It is a counter example of the efficient market hypothesis. To utilize this type of anomalous volatility spillover pattern, we analyzed the intraday volatility selling strategy. This strategy sells short the Korean volatility market in the morning after the US stock market volatility closes down and takes no position in the volatility market after the VIX closes up. It produced profit every year between 2008 and 2018 and the percent profitable is 68%. The trading performance showed the higher average annual return of 129% relative to the benchmark average annual return of 33%. The maximum draw down, MDD, is -41%, which is lower than that of benchmark -101%. The Sharpe ratio 0.32 of SVS strategy is much greater than the Sharpe ratio 0.08 of the Benchmark strategy. The Sharpe ratio simultaneously considers return and risk and is calculated as return divided by risk. Therefore, high Sharpe ratio means high performance when comparing different strategies with different risk and return structure. Real world trading gives rise to the trading costs including brokerage cost and slippage cost. When the trading cost is considered, the performance difference between 76% and -10% average annual returns becomes clear. To improve the performance of the suggested volatility trading strategy, we used the well-known SVM algorithm. Input variables include the VIX close to close return at day t-1, the VIX open to close return at day t-1, the VK open return at day t, and output is the up and down classification of the VK open to close return at day t. The training period is from 2008 to 2014 and the testing period is from 2015 to 2018. The kernel functions are linear function, radial basis function, and polynomial function. We suggested the modified-short volatility strategy that sells the VK in the morning when the SVM output is Down and takes no position when the SVM output is Up. The trading performance was remarkably improved. The 5-year testing period trading results of the m-SVS strategy showed very high profit and low risk relative to the benchmark SVS strategy. The annual return of the m-SVS strategy is 123% and it is higher than that of SVS strategy. The risk factor, MDD, was also significantly improved from -41% to -29%.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.59-83
• /
• 2018

With the increasing importance of sentiment analysis to grasp the needs of customers and the public, various types of deep learning models have been actively applied to English texts. In the sentiment analysis of English texts by deep learning, natural language sentences included in training and test datasets are usually converted into sequences of word vectors before being entered into the deep learning models. In this case, word vectors generally refer to vector representations of words obtained through splitting a sentence by space characters. There are several ways to derive word vectors, one of which is Word2Vec used for producing the 300 dimensional Google word vectors from about 100 billion words of Google News data. They have been widely used in the studies of sentiment analysis of reviews from various fields such as restaurants, movies, laptops, cameras, etc. Unlike English, morpheme plays an essential role in sentiment analysis and sentence structure analysis in Korean, which is a typical agglutinative language with developed postpositions and endings. A morpheme can be defined as the smallest meaningful unit of a language, and a word consists of one or more morphemes. For example, for a word '예쁘고', the morphemes are '예쁘(= adjective)' and '고(=connective ending)'. Reflecting the significance of Korean morphemes, it seems reasonable to adopt the morphemes as a basic unit in Korean sentiment analysis. Therefore, in this study, we use 'morpheme vector' as an input to a deep learning model rather than 'word vector' which is mainly used in English text. The morpheme vector refers to a vector representation for the morpheme and can be derived by applying an existent word vector derivation mechanism to the sentences divided into constituent morphemes. By the way, here come some questions as follows. What is the desirable range of POS(Part-Of-Speech) tags when deriving morpheme vectors for improving the classification accuracy of a deep learning model? Is it proper to apply a typical word vector model which primarily relies on the form of words to Korean with a high homonym ratio? Will the text preprocessing such as correcting spelling or spacing errors affect the classification accuracy, especially when drawing morpheme vectors from Korean product reviews with a lot of grammatical mistakes and variations? We seek to find empirical answers to these fundamental issues, which may be encountered first when applying various deep learning models to Korean texts. As a starting point, we summarized these issues as three central research questions as follows. First, which is better effective, to use morpheme vectors from grammatically correct texts of other domain than the analysis target, or to use morpheme vectors from considerably ungrammatical texts of the same domain, as the initial input of a deep learning model? Second, what is an appropriate morpheme vector derivation method for Korean regarding the range of POS tags, homonym, text preprocessing, minimum frequency? Third, can we get a satisfactory level of classification accuracy when applying deep learning to Korean sentiment analysis? As an approach to these research questions, we generate various types of morpheme vectors reflecting the research questions and then compare the classification accuracy through a non-static CNN(Convolutional Neural Network) model taking in the morpheme vectors. As for training and test datasets, Naver Shopping's 17,260 cosmetics product reviews are used. To derive morpheme vectors, we use data from the same domain as the target one and data from other domain; Naver shopping's about 2 million cosmetics product reviews and 520,000 Naver News data arguably corresponding to Google's News data. The six primary sets of morpheme vectors constructed in this study differ in terms of the following three criteria. First, they come from two types of data source; Naver news of high grammatical correctness and Naver shopping's cosmetics product reviews of low grammatical correctness. Second, they are distinguished in the degree of data preprocessing, namely, only splitting sentences or up to additional spelling and spacing corrections after sentence separation. Third, they vary concerning the form of input fed into a word vector model; whether the morphemes themselves are entered into a word vector model or with their POS tags attached. The morpheme vectors further vary depending on the consideration range of POS tags, the minimum frequency of morphemes included, and the random initialization range. All morpheme vectors are derived through CBOW(Continuous Bag-Of-Words) model with the context window 5 and the vector dimension 300. It seems that utilizing the same domain text even with a lower degree of grammatical correctness, performing spelling and spacing corrections as well as sentence splitting, and incorporating morphemes of any POS tags including incomprehensible category lead to the better classification accuracy. The POS tag attachment, which is devised for the high proportion of homonyms in Korean, and the minimum frequency standard for the morpheme to be included seem not to have any definite influence on the classification accuracy.

• Asia pacific journal of information systems
• /
• v.21 no.4
• /
• pp.63-104
• /
• 2011

The study examines the relationships among employee's goal orientation, IT personnel competency, personal effectiveness. The goal orientation includes learning goal orientation, performance approach goal orientation, and performance avoid goal orientation. Personal effectiveness consists of personal work satisfaction and personal work performance. In general, IT personnel competency refers to IT expert's skills, expertise, and knowledge required to perform IT activities in organizations. However, due to the advent of the internet and the generalization of IT, IT personnel competency turns out to be an important competency of technological experts as well as employees in organizations. While the competency of IT itself is important, the appropriate harmony between IT personnel's business capability and technological capability enhances the value of human resources and thus provides organizations with sustainable competitive advantages. The rapid pace of organization change places increased pressure on employees to continually update their skills and adapt their behavior to new organizational realities. This challenge raises a number of important questions concerning organizational behavior? Why do some employees display remarkable flexibility in their behavioral responses to changes in the organization, whereas others firmly resist change or experience great stress when faced with the need to alter behavior? Why do some employees continually strive to improve themselves over their life span, whereas others are content to forge through life using the same basic knowledge and skills? Why do some employees throw themselves enthusiastically into challenging tasks, whereas others avoid challenging tasks? The goal orientation proposed by organizational psychology provides at least a partial answer to these questions. Goal orientations refer to stable personally characteristics fostered by "self-theories" about the nature and development of attributes (such as intelligence, personality, abilities, and skills) people have. Self-theories are one's beliefs and goal orientations are achievement motivation revealed in seeking goals in accordance with one's beliefs. The goal orientations include learning goal orientation, performance approach goal orientation, and performance avoid goal orientation. Specifically, a learning goal orientation refers to a preference to develop the self by acquiring new skills, mastering new situations, and improving one's competence. A performance approach goal orientation refers to a preference to demonstrate and validate the adequacy of one's competence by seeking favorable judgments and avoiding negative judgments. A performance avoid goal orientation refers to a preference to avoid the disproving of one's competence and to avoid negative judgements about it, while focusing on performance. And the study also examines the moderating role of work career of employees to investigate the difference in the relationship between IT personnel competency and personal effectiveness. The study analyzes the collected data using PASW 18.0 and and PLS(Partial Least Square). The study also uses PLS bootstrapping algorithm (sample size: 500) to test research hypotheses. The result shows that the influences of both a learning goal orientation (${\beta}$ = 0.301, t = 3.822, P < 0.000) and a performance approach goal orientation (${\beta}$ = 0.224, t = 2.710, P < 0.01) on IT personnel competency are positively significant, while the influence of a performance avoid goal orientation(${\beta}$ = -0.142, t = 2.398, p < 0.05) on IT personnel competency is negatively significant. The result indicates that employees differ in their psychological and behavioral responses according to the goal orientation of employees. The result also shows that the impact of a IT personnel competency on both personal work satisfaction(${\beta}$ = 0.395, t = 4.897, P < 0.000) and personal work performance(${\beta}$ = 0.575, t = 12.800, P < 0.000) is positively significant. And the impact of personal work satisfaction(${\beta}$ = 0.148, t = 2.432, p < 0.05) on personal work performance is positively significant. Finally, the impacts of control variables (gender, age, type of industry, position, work career) on the relationships between IT personnel competency and personal effectiveness(personal work satisfaction work performance) are partly significant. In addition, the study uses PLS algorithm to find out a GoF(global criterion of goodness of fit) of the exploratory research model which includes a mediating variable, IT personnel competency. The result of analysis shows that the value of GoF is 0.45 above GoFlarge(0.36). Therefore, the research model turns out be good. In addition, the study performs a Sobel Test to find out the statistical significance of the mediating variable, IT personnel competency, which is already turned out to have the mediating effect in the research model using PLS. The result of a Sobel Test shows that the values of Z are all significant statistically (above 1.96 and below -1.96) and indicates that IT personnel competency plays a mediating role in the research model. At the present day, most employees are universally afraid of organizational changes and resistant to them in organizations in which the acceptance and learning of a new information technology or information system is particularly required. The problem is due' to increasing a feeling of uneasiness and uncertainty in improving past practices in accordance with new organizational changes. It is not always possible for employees with positive attitudes to perform their works suitable to organizational goals. Therefore, organizations need to identify what kinds of goal-oriented minds employees have, motivate them to do self-directed learning, and provide them with organizational environment to enhance positive aspects in their works. Thus, the study provides researchers and practitioners with a matter of primary interest in goal orientation and IT personnel competency, of which they have been unaware until very recently. Some academic and practical implications and limitations arisen in the course of the research, and suggestions for future research directions are also discussed.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.21-37
• /
• 2013

Recently, as the advent of the knowledge-based society, there are more people getting interested in the intellectual property. Especially, the ICT companies leading the high-tech industry are working hard to strive for systematic management of intellectual property. As we know, the patent information represents the intellectual capital of the company. Also now the quantitative analysis on the continuously accumulated patent information becomes possible. The analysis at various levels becomes also possible by utilizing the patent information, ranging from the patent level to the enterprise level, industrial level and country level. Through the patent information, we can identify the technology status and analyze the impact of the performance. We are also able to find out the flow of the knowledge through the network analysis. By that, we can not only identify the changes in technology, but also predict the direction of the future research. In the field using the network analysis there are two important analyses which utilize the patent citation information; citation indicator analysis utilizing the frequency of the citation and network analysis based on the citation relationships. Furthermore, this study analyzes whether there are any impacts between the size of the company and patent citation relationships. 74 S&P 500 registered companies that provide IT and communication services are selected for this study. In order to determine the relationship of patent citation between the companies, the patent citation in 2009 and 2010 is collected and sociomatrices which show the patent citation relationship between the companies are created. In addition, the companies' total assets are collected as an index of company size. The distance between companies is defined as the absolute value of the difference between the total assets. And simple differences are considered to be described as the hierarchy of the company. The QAP Correlation analysis and MR-QAP analysis is carried out by using the distance and hierarchy between companies, and also the sociomatrices that shows the patent citation in 2009 and 2010. Through the result of QAP Correlation analysis, the patent citation relationship between companies in the 2009's company's patent citation network and the 2010's company's patent citation network shows the highest correlation. In addition, positive correlation is shown in the patent citation relationships between companies and the distance between companies. This is because the patent citation relationship is increased when there is a difference of size between companies. Not only that, negative correlation is found through the analysis using the patent citation relationship between companies and the hierarchy between companies. Relatively it is indicated that there is a high evaluation about the patent of the higher tier companies influenced toward the lower tier companies. MR-QAP analysis is carried out as follow. The sociomatrix that is generated by using the year 2010 patent citation relationship is used as the dependent variable. Additionally the 2009's company's patent citation network and the distance and hierarchy networks between the companies are used as the independent variables. This study performed MR-QAP analysis to find the main factors influencing the patent citation relationship between the companies in 2010. The analysis results show that all independent variables have positively influenced the 2010's patent citation relationship between the companies. In particular, the 2009's patent citation relationship between the companies has the most significant impact on the 2010's, which means that there is consecutiveness regarding the patent citation relationships. Through the result of QAP correlation analysis and MR-QAP analysis, the patent citation relationship between companies is affected by the size of the companies. But the most significant impact is the patent citation relationships that had been done in the past. The reason why we need to maintain the patent citation relationship between companies is it might be important in the use of strategic aspect of the companies to look into relationships to share intellectual property between each other, also seen as an important auxiliary of the partner companies to cooperate with.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.55-79
• /
• 2013

In this paper, we propose a hybrid genetic algorithm (HGA) approach to effectively solve the reverse logistics network with centralized centers (RLNCC). For the proposed HGA approach, genetic algorithm (GA) is used as a main algorithm. For implementing GA, a new bit-string representation scheme using 0 and 1 values is suggested, which can easily make initial population of GA. As genetic operators, the elitist strategy in enlarged sampling space developed by Gen and Chang (1997), a new two-point crossover operator, and a new random mutation operator are used for selection, crossover and mutation, respectively. For hybrid concept of GA, an iterative hill climbing method (IHCM) developed by Michalewicz (1994) is inserted into HGA search loop. The IHCM is one of local search techniques and precisely explores the space converged by GA search. The RLNCC is composed of collection centers, remanufacturing centers, redistribution centers, and secondary markets in reverse logistics networks. Of the centers and secondary markets, only one collection center, remanufacturing center, redistribution center, and secondary market should be opened in reverse logistics networks. Some assumptions are considered for effectively implementing the RLNCC The RLNCC is represented by a mixed integer programming (MIP) model using indexes, parameters and decision variables. The objective function of the MIP model is to minimize the total cost which is consisted of transportation cost, fixed cost, and handling cost. The transportation cost is obtained by transporting the returned products between each centers and secondary markets. The fixed cost is calculated by opening or closing decision at each center and secondary markets. That is, if there are three collection centers (the opening costs of collection center 1 2, and 3 are 10.5, 12.1, 8.9, respectively), and the collection center 1 is opened and the remainders are all closed, then the fixed cost is 10.5. The handling cost means the cost of treating the products returned from customers at each center and secondary markets which are opened at each RLNCC stage. The RLNCC is solved by the proposed HGA approach. In numerical experiment, the proposed HGA and a conventional competing approach is compared with each other using various measures of performance. For the conventional competing approach, the GA approach by Yun (2013) is used. The GA approach has not any local search technique such as the IHCM proposed the HGA approach. As measures of performance, CPU time, optimal solution, and optimal setting are used. Two types of the RLNCC with different numbers of customers, collection centers, remanufacturing centers, redistribution centers and secondary markets are presented for comparing the performances of the HGA and GA approaches. The MIP models using the two types of the RLNCC are programmed by Visual Basic Version 6.0, and the computer implementing environment is the IBM compatible PC with 3.06Ghz CPU speed and 1GB RAM on Windows XP. The parameters used in the HGA and GA approaches are that the total number of generations is 10,000, population size 20, crossover rate 0.5, mutation rate 0.1, and the search range for the IHCM is 2.0. Total 20 iterations are made for eliminating the randomness of the searches of the HGA and GA approaches. With performance comparisons, network representations by opening/closing decision, and convergence processes using two types of the RLNCCs, the experimental result shows that the HGA has significantly better performance in terms of the optimal solution than the GA, though the GA is slightly quicker than the HGA in terms of the CPU time. Finally, it has been proved that the proposed HGA approach is more efficient than conventional GA approach in two types of the RLNCC since the former has a GA search process as well as a local search process for additional search scheme, while the latter has a GA search process alone. For a future study, much more large-sized RLNCCs will be tested for robustness of our approach.

• Journal of Intelligence and Information Systems
• /
• v.23 no.3
• /
• pp.29-44
• /
• 2017

In Internet commerce, consumers are heavily influenced by product reviews written by other users who have already purchased the product. However, as the product reviews accumulate, it takes a lot of time and effort for consumers to individually check the massive number of product reviews. Moreover, product reviews that are written carelessly actually inconvenience consumers. Thus many online vendors provide mechanisms to identify reviews that customers perceive as most helpful (Cao et al. 2011; Mudambi and Schuff 2010). For example, some online retailers, such as Amazon.com and TripAdvisor, allow users to rate the helpfulness of each review, and use this feedback information to rank and re-order them. However, many reviews have only a few feedbacks or no feedback at all, thus making it hard to identify their helpfulness. Also, it takes time to accumulate feedbacks, thus the newly authored reviews do not have enough ones. For example, only 20% of the reviews in Amazon Review Dataset (Mcauley and Leskovec, 2013) have more than 5 reviews (Yan et al, 2014). The purpose of this study is to analyze the factors affecting the usefulness of online product reviews and to derive a forecasting model that selectively provides product reviews that can be helpful to consumers. In order to do this, we extracted the various linguistic, psychological, and perceptual elements included in product reviews by using text-mining techniques and identifying the determinants among these elements that affect the usability of product reviews. In particular, considering that the characteristics of the product reviews and determinants of usability for apparel products (which are experiential products) and electronic products (which are search goods) can differ, the characteristics of the product reviews were compared within each product group and the determinants were established for each. This study used 7,498 apparel product reviews and 106,962 electronic product reviews from Amazon.com. In order to understand a review text, we first extract linguistic and psychological characteristics from review texts such as a word count, the level of emotional tone and analytical thinking embedded in review text using widely adopted text analysis software LIWC (Linguistic Inquiry and Word Count). After then, we explore the descriptive statistics of review text for each category and statistically compare their differences using t-test. Lastly, we regression analysis using the data mining software RapidMiner to find out determinant factors. As a result of comparing and analyzing product review characteristics of electronic products and apparel products, it was found that reviewers used more words as well as longer sentences when writing product reviews for electronic products. As for the content characteristics of the product reviews, it was found that these reviews included many analytic words, carried more clout, and related to the cognitive processes (CogProc) more so than the apparel product reviews, in addition to including many words expressing negative emotions (NegEmo). On the other hand, the apparel product reviews included more personal, authentic, positive emotions (PosEmo) and perceptual processes (Percept) compared to the electronic product reviews. Next, we analyzed the determinants toward the usefulness of the product reviews between the two product groups. As a result, it was found that product reviews with high product ratings from reviewers in both product groups that were perceived as being useful contained a larger number of total words, many expressions involving perceptual processes, and fewer negative emotions. In addition, apparel product reviews with a large number of comparative expressions, a low expertise index, and concise content with fewer words in each sentence were perceived to be useful. In the case of electronic product reviews, those that were analytical with a high expertise index, along with containing many authentic expressions, cognitive processes, and positive emotions (PosEmo) were perceived to be useful. These findings are expected to help consumers effectively identify useful product reviews in the future.

• Journal of Intelligence and Information Systems
• /
• v.20 no.1
• /
• pp.163-176
• /
• 2014

Social media is becoming the platform for users to communicate their activities, status, emotions, and experiences to other people. In recent years, microblogs, such as Twitter, have gained in popularity because of its ease of use, speed, and reach. Compared to a conventional web blog, a microblog lowers users' efforts and investment for content generation by recommending shorter posts. There has been a lot research into capturing the social phenomena and analyzing the chatter of microblogs. However, measuring television ratings has been given little attention so far. Currently, the most common method to measure TV ratings uses an electronic metering device installed in a small number of sampled households. Microblogs allow users to post short messages, share daily updates, and conveniently keep in touch. In a similar way, microblog users are interacting with each other while watching television or movies, or visiting a new place. In order to measure TV ratings, some features are significant during certain hours of the day, or days of the week, whereas these same features are meaningless during other time periods. Thus, the importance of features can change during the day, and a model capturing the time sensitive relevance is required to estimate TV ratings. Therefore, modeling time-related characteristics of features should be a key when measuring the TV ratings through microblogs. We show that capturing time-dependency of features in measuring TV ratings is vitally necessary for improving their accuracy. To explore the relationship between the content of microblogs and TV ratings, we collected Twitter data using the Get Search component of the Twitter REST API from January 2013 to October 2013. There are about 300 thousand posts in our data set for the experiment. After excluding data such as adverting or promoted tweets, we selected 149 thousand tweets for analysis. The number of tweets reaches its maximum level on the broadcasting day and increases rapidly around the broadcasting time. This result is stems from the characteristics of the public channel, which broadcasts the program at the predetermined time. From our analysis, we find that count-based features such as the number of tweets or retweets have a low correlation with TV ratings. This result implies that a simple tweet rate does not reflect the satisfaction or response to the TV programs. Content-based features extracted from the content of tweets have a relatively high correlation with TV ratings. Further, some emoticons or newly coined words that are not tagged in the morpheme extraction process have a strong relationship with TV ratings. We find that there is a time-dependency in the correlation of features between the before and after broadcasting time. Since the TV program is broadcast at the predetermined time regularly, users post tweets expressing their expectation for the program or disappointment over not being able to watch the program. The highly correlated features before the broadcast are different from the features after broadcasting. This result explains that the relevance of words with TV programs can change according to the time of the tweets. Among the 336 words that fulfill the minimum requirements for candidate features, 145 words have the highest correlation before the broadcasting time, whereas 68 words reach the highest correlation after broadcasting. Interestingly, some words that express the impossibility of watching the program show a high relevance, despite containing a negative meaning. Understanding the time-dependency of features can be helpful in improving the accuracy of TV ratings measurement. This research contributes a basis to estimate the response to or satisfaction with the broadcasted programs using the time dependency of words in Twitter chatter. More research is needed to refine the methodology for predicting or measuring TV ratings.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.1-17
• /
• 2015

In this paper, we report what we have observed with regards to a prediction model for the military based on enlisted men's internal(cumulative records) and external data(SNS data). This work is significant in the military's efforts to supervise them. In spite of their effort, many commanders have failed to prevent accidents by their subordinates. One of the important duties of officers' work is to take care of their subordinates in prevention unexpected accidents. However, it is hard to prevent accidents so we must attempt to determine a proper method. Our motivation for presenting this paper is to mate it possible to predict accidents using enlisted men's internal and external data. The biggest issue facing the military is the occurrence of accidents by enlisted men related to maladjustment and the relaxation of military discipline. The core method of preventing accidents by soldiers is to identify problems and manage them quickly. Commanders predict accidents by interviewing their soldiers and observing their surroundings. It requires considerable time and effort and results in a significant difference depending on the capabilities of the commanders. In this paper, we seek to predict accidents with objective data which can easily be obtained. Recently, records of enlisted men as well as SNS communication between commanders and soldiers, make it possible to predict and prevent accidents. This paper concerns the application of data mining to identify their interests, predict accidents and make use of internal and external data (SNS). We propose both a topic analysis and decision tree method. The study is conducted in two steps. First, topic analysis is conducted through the SNS of enlisted men. Second, the decision tree method is used to analyze the internal data with the results of the first analysis. The dependent variable for these analysis is the presence of any accidents. In order to analyze their SNS, we require tools such as text mining and topic analysis. We used SAS Enterprise Miner 12.1, which provides a text miner module. Our approach for finding their interests is composed of three main phases; collecting, topic analysis, and converting topic analysis results into points for using independent variables. In the first phase, we collect enlisted men's SNS data by commender's ID. After gathering unstructured SNS data, the topic analysis phase extracts issues from them. For simplicity, 5 topics(vacation, friends, stress, training, and sports) are extracted from 20,000 articles. In the third phase, using these 5 topics, we quantify them as personal points. After quantifying their topic, we include these results in independent variables which are composed of 15 internal data sets. Then, we make two decision trees. The first tree is composed of their internal data only. The second tree is composed of their external data(SNS) as well as their internal data. After that, we compare the results of misclassification from SAS E-miner. The first model's misclassification is 12.1%. On the other hand, second model's misclassification is 7.8%. This method predicts accidents with an accuracy of approximately 92%. The gap of the two models is 4.3%. Finally, we test if the difference between them is meaningful or not, using the McNemar test. The result of test is considered relevant.(p-value : 0.0003) This study has two limitations. First, the results of the experiments cannot be generalized, mainly because the experiment is limited to a small number of enlisted men's data. Additionally, various independent variables used in the decision tree model are used as categorical variables instead of continuous variables. So it suffers a loss of information. In spite of extensive efforts to provide prediction models for the military, commanders' predictions are accurate only when they have sufficient data about their subordinates. Our proposed methodology can provide support to decision-making in the military. This study is expected to contribute to the prevention of accidents in the military based on scientific analysis of enlisted men and proper management of them.

• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.