• Journal of KIISE:Information Networking
• /
• v.32 no.3
• /
• pp.279-290
• /
• 2005

Recently, as the serious damage caused by DDoS attacks increases, the rapid detection and the proper response mechanisms are urgent. However, existing security mechanisms do not effectively defend against these attacks, or the defense capability of some mechanisms is only limited to specific DDoS attacks. In this paper, we propose a detection architecture against DDoS attack using data mining technology that can classify the latest types of DDoS attack, and can detect the modification of existing attacks as well as the novel attacks. This architecture consists of a Misuse Detection Module modeling to classify the existing attacks, and an Anomaly Detection Module modeling to detect the novel attacks. And it utilizes the off-line generated models in order to detect the DDoS attack using the real-time traffic. We gathered the NetFlow data generated at an access router of our network in order to model the real network traffic and test it. The NetFlow provides the useful flow-based statistical information without tremendous preprocessing. Also, we mounted the well-known DDoS attack tools to gather the attack traffic. And then, our experimental results show that our approach can provide the outstanding performance against existing attacks, and provide the possibility of detection against the novel attack.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.6
• /
• pp.152-161
• /
• 2023

In general, large-capacity hydrogen storage vessels, typically in the form of vertical cylindrical vessels, are constructed using steel materials. These vessels are anchored to foundation slabs that are specially designed to suit the environmental conditions. This anchoring method involves pre-installed anchors on top of the concrete foundation slab. However, it's important to note that such a design can result in concentrated stresses at the anchoring points when external forces, such as seismic events, are at play. This may lead to potential structural damage due to anchor and concrete damage. For this reason, in this study, it selected an vertical hydrogen storage vessel based on site observations and created a 3D finite element model. Artificial seismic motions made following the procedures specified in ICC-ES AC 156, as well as domestic recorded earthquakes with a magnitude greater than 5.0, were applied to analyze the structural behavior and performance of the target structures. Conducting experiments on a structure built to actual scale would be ideal, but due to practical constraints, it proved challenging to execute. Therefore, it opted for an analytical approach to assess the safety of the target structure. Regarding the structural response characteristics, the acceleration induced by seismic motion was observed to amplify by approximately ten times compared to the input seismic motions. Additionally, there was a tendency for a decrease in amplification as the response acceleration was transmitted to the point where the centre of gravity is located. For the vulnerable components, specifically the sub-system (support columns and anchorages), the stress levels were found to satisfy the allowable stress criteria. However, the concrete's tensile strength exhibited only about a 5% margin of safety compared to the allowable stress. This indicates the need for mitigation strategies in addressing these concerns. Based on the research findings presented in this paper, it is anticipated that predictable load information for the design of storage vessels required for future shaking table tests will be provided.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.39-55
• /
• 2019

Recently banks and large financial institutions have introduced lots of Robo-Advisor products. Robo-Advisor is a Robot to produce the optimal asset allocation portfolio for investors by using the financial engineering algorithms without any human intervention. Since the first introduction in Wall Street in 2008, the market size has grown to 60 billion dollars and is expected to expand to 2,000 billion dollars by 2020. Since Robo-Advisor algorithms suggest asset allocation output to investors, mathematical or statistical asset allocation strategies are applied. Mean variance optimization model developed by Markowitz is the typical asset allocation model. The model is a simple but quite intuitive portfolio strategy. For example, assets are allocated in order to minimize the risk on the portfolio while maximizing the expected return on the portfolio using optimization techniques. Despite its theoretical background, both academics and practitioners find that the standard mean variance optimization portfolio is very sensitive to the expected returns calculated by past price data. Corner solutions are often found to be allocated only to a few assets. The Black-Litterman Optimization model overcomes these problems by choosing a neutral Capital Asset Pricing Model equilibrium point. Implied equilibrium returns of each asset are derived from equilibrium market portfolio through reverse optimization. The Black-Litterman model uses a Bayesian approach to combine the subjective views on the price forecast of one or more assets with implied equilibrium returns, resulting a new estimates of risk and expected returns. These new estimates can produce optimal portfolio by the well-known Markowitz mean-variance optimization algorithm. If the investor does not have any views on his asset classes, the Black-Litterman optimization model produce the same portfolio as the market portfolio. What if the subjective views are incorrect? A survey on reports of stocks performance recommended by securities analysts show very poor results. Therefore the incorrect views combined with implied equilibrium returns may produce very poor portfolio output to the Black-Litterman model users. This paper suggests an objective investor views model based on Support Vector Machines(SVM), which have showed good performance results in stock price forecasting. SVM is a discriminative classifier defined by a separating hyper plane. The linear, radial basis and polynomial kernel functions are used to learn the hyper planes. Input variables for the SVM are returns, standard deviations, Stochastics %K and price parity degree for each asset class. SVM output returns expected stock price movements and their probabilities, which are used as input variables in the intelligent views model. The stock price movements are categorized by three phases; down, neutral and up. The expected stock returns make P matrix and their probability results are used in Q matrix. Implied equilibrium returns vector is combined with the intelligent views matrix, resulting the Black-Litterman optimal portfolio. For comparisons, Markowitz mean-variance optimization model and risk parity model are used. The value weighted market portfolio and equal weighted market portfolio are used as benchmark indexes. We collect the 8 KOSPI 200 sector indexes from January 2008 to December 2018 including 132 monthly index values. Training period is from 2008 to 2015 and testing period is from 2016 to 2018. Our suggested intelligent view model combined with implied equilibrium returns produced the optimal Black-Litterman portfolio. The out of sample period portfolio showed better performance compared with the well-known Markowitz mean-variance optimization portfolio, risk parity portfolio and market portfolio. The total return from 3 year-period Black-Litterman portfolio records 6.4%, which is the highest value. The maximum draw down is -20.8%, which is also the lowest value. Sharpe Ratio shows the highest value, 0.17. It measures the return to risk ratio. Overall, our suggested view model shows the possibility of replacing subjective analysts's views with objective view model for practitioners to apply the Robo-Advisor asset allocation algorithms in the real trading fields.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.63-83
• /
• 2019

Investors prefer to look for trading points based on the graph shown in the chart rather than complex analysis, such as corporate intrinsic value analysis and technical auxiliary index analysis. However, the pattern analysis technique is difficult and computerized less than the needs of users. In recent years, there have been many cases of studying stock price patterns using various machine learning techniques including neural networks in the field of artificial intelligence(AI). In particular, the development of IT technology has made it easier to analyze a huge number of chart data to find patterns that can predict stock prices. Although short-term forecasting power of prices has increased in terms of performance so far, long-term forecasting power is limited and is used in short-term trading rather than long-term investment. Other studies have focused on mechanically and accurately identifying patterns that were not recognized by past technology, but it can be vulnerable in practical areas because it is a separate matter whether the patterns found are suitable for trading. When they find a meaningful pattern, they find a point that matches the pattern. They then measure their performance after n days, assuming that they have bought at that point in time. Since this approach is to calculate virtual revenues, there can be many disparities with reality. The existing research method tries to find a pattern with stock price prediction power, but this study proposes to define the patterns first and to trade when the pattern with high success probability appears. The M & W wave pattern published by Merrill(1980) is simple because we can distinguish it by five turning points. Despite the report that some patterns have price predictability, there were no performance reports used in the actual market. The simplicity of a pattern consisting of five turning points has the advantage of reducing the cost of increasing pattern recognition accuracy. In this study, 16 patterns of up conversion and 16 patterns of down conversion are reclassified into ten groups so that they can be easily implemented by the system. Only one pattern with high success rate per group is selected for trading. Patterns that had a high probability of success in the past are likely to succeed in the future. So we trade when such a pattern occurs. It is a real situation because it is measured assuming that both the buy and sell have been executed. We tested three ways to calculate the turning point. The first method, the minimum change rate zig-zag method, removes price movements below a certain percentage and calculates the vertex. In the second method, high-low line zig-zag, the high price that meets the n-day high price line is calculated at the peak price, and the low price that meets the n-day low price line is calculated at the valley price. In the third method, the swing wave method, the high price in the center higher than n high prices on the left and right is calculated as the peak price. If the central low price is lower than the n low price on the left and right, it is calculated as valley price. The swing wave method was superior to the other methods in the test results. It is interpreted that the transaction after checking the completion of the pattern is more effective than the transaction in the unfinished state of the pattern. Genetic algorithms(GA) were the most suitable solution, although it was virtually impossible to find patterns with high success rates because the number of cases was too large in this simulation. We also performed the simulation using the Walk-forward Analysis(WFA) method, which tests the test section and the application section separately. So we were able to respond appropriately to market changes. In this study, we optimize the stock portfolio because there is a risk of over-optimized if we implement the variable optimality for each individual stock. Therefore, we selected the number of constituent stocks as 20 to increase the effect of diversified investment while avoiding optimization. We tested the KOSPI market by dividing it into six categories. In the results, the portfolio of small cap stock was the most successful and the high vol stock portfolio was the second best. This shows that patterns need to have some price volatility in order for patterns to be shaped, but volatility is not the best.

• Journal of Distribution Research
• /
• v.15 no.1
• /
• pp.49-71
• /
• 2010

Franchise businesses in South Korea have contributed to economic growth and job creation, and its growth potential remains very high. However, despite such virtues, domestic franchise businesses face many problems such as the instability of franchisor's business structure and weak financial conditions. To solve these problems, the government enacted legislation and strengthened franchise related laws. However, the strengthening of laws regulating franchisors had many side effects that interrupted the development of the franchise business. For example, legal regulations regarding franchisors have had the effect of suppressing the franchisor's leadership activities (e.g. activities such as the ability to advocate the franchisor's policies and strategies to the franchisees, in order to facilitate change and innovation). One of the main goals of the franchise business is to build cooperation between the franchisor and the franchisee for their combined success. However, franchisees can refuse to follow the franchisor's strategies because of the current state of franchise-related law and government policy. The purpose of this study to explore the effects of franchisor's leadership style on franchisee's commitment in a franchise system. We classified leadership styles according to the path-goal theory (House & Mitchell, 1974), and it was hypothesized and tested that the four leadership styles proposed by the path-goal theory (i.e. directive, supportive, participative and achievement-oriented leadership) have different effects on franchisee's commitment. Another purpose of this study to explore the how the level of franchisee's self-efficacy influences both the franchisor's leadership style and franchisee's commitment in a franchise system. Results of the present study are expected to provide important theoretical and practical implications as to the role of franchisor's leadership style, as restricted by government regulations and the franchisee's self-efficacy, which could be needed to improve the quality of the long-term relationship between the franchisor and franchisee. Quoted by Northouse(2007), one problem regarding the investigation of leadership is that there are almost as many different definitions of leadership as there are people who have tried to define it. But despite the multitude of ways in which leadership has been conceptualized, the following components can be identified as central to the phenomenon: (a) leadership is a process, (b) leadership involves influence, (c) leadership occurs in a group context, and (d) leadership involves goal attainment. Based on these components, in this study leadership is defined as a process whereby franchisor's influences a group of franchisee' to achieve a common goal. Focusing on this definition, the path-goal theory is about how leaders motivate subordinates to accomplish designated goals. Drawing heavily from research on what motivates employees, path-goal theory first appeared in the leadership literature in the early 1970s in the works of Evans (1970), House (1971), House and Dessler (1974), and House and Mitchell (1974). The stated goal of this leadership theory is to enhance employee performance and employee satisfaction by focusing on employee motivation. In brief, path-goal theory is designed to explain how leaders can help subordinates along the path to their goals by selecting specific behaviors that are best suited to subordinates' needs and to the situation in which subordinates are working (Northouse, 2007). House & Mitchell(1974) predicted that although many different leadership behaviors could have been selected to be a part of path-goal theory, this approach has so far examined directive, supportive, participative, and achievement-oriented leadership behaviors. And they suggested that leaders may exhibit any or all of these four styles with various subordinates and in different situations. However, due to restrictive government regulations, franchisors are not in a position to change their leadership style to suit their circumstances. In addition, quoted by Northouse(2007), ssubordinate characteristics determine how a leader's behavior is interpreted by subordinates in a given work context. Many researchers have focused on subordinates' needs for affiliation, preferences for structure, desires for control, and self-perceived level of task ability. In this study, we have focused on the self-perceived level of task ability, namely, the franchisee's self-efficacy. According to Bandura (1977), self-efficacy is chiefly defined as the personal attitude of one's ability to accomplish concrete tasks. Therefore, it is not an indicator of one's actual abilities, but an opinion of the extent of how one can use that ability. Thus, the judgment of maintain franchisee's commitment depends on the situation (e.g., government regulation and policy and leadership style of franchisor) and how it affects one's ability to mobilize resources to deal with the task, so even if people possess the same ability, there may be differences in self-efficacy. Figure 1 illustrates the model investigated in this study. In this model, it was hypothesized that leadership styles would affect the franchisee's commitment, and self-efficacy would moderate the relationship between leadership style and franchisee's commitment. Theoretically, quoted by Northouse(2007), the path-goal approach suggests that leaders need to choose a leadership style that best fits the needs of subordinates and the work they are doing. According to House & Mitchell (1974), the theory predicts that a directive style of leadership is best in situations in which subordinates are dogmatic and authoritarian, the task demands are ambiguous, and the organizational rule and procedures are unclear. In these situations, franchisor's directive leadership complements the work by providing guidance and psychological structure for franchisees. For work that is structured, unsatisfying, or frustrating, path-goal theory suggests that leaders should use a supportive style. Franchisor's Supportive leadership offers a sense of human touch for franchisees engaged in mundane, mechanized activity. Franchisor's participative leadership is considered best when a task is ambiguous because participation gives greater clarity to how certain paths lead to certain goals; it helps subordinates learn what actions leads to what outcome. Furthermore, House & Mitchell(1974) predicts that achievement-oriented leadership is most effective in settings in which subordinates are required to perform ambiguous tasks. Marsh and O'Neill (1984) tested the idea that organizational members' anger and decline in performance is caused by deficiencies in their level of effort and found that self-efficacy promotes accomplishment, decreases stress and negative consequences like depression and emotional instability. Based on the extant empirical findings and theoretical reasoning, we posit positive and strong relationships between the franchisor's leadership styles and the franchisee's commitment. Furthermore, the level of franchisee's self-efficacy was thought to maintain their commitment. The questionnaires sent to participants consisted of the following measures; leadership style was assessed using a 20 item 7-point likert scale developed by Indvik (1985), self-efficacy was assessed using a 24 item 6-point likert scale developed by Bandura (1977), and commitment was assessed using a 6 item 5-point likert scale developed by Morgan & Hunt (1994). Questionnaires were distributed to Korean optical franchisees in Seoul. It took about 20 days to complete the data collection. A total number of 140 questionnaires were returned and complete data were available from 137 respondents. Results of multiple regression analyses testing the relationships between the each of the four styles of leadership shown by the franchisor as independent variables and franchisee's commitment as the dependent variable showed that the relationship between supportive leadership style and commitment ($\beta$=.13, p<.001),and the relationship between participative leadership style and commitment ($\beta$=.07, p<.001)were significant. However, when participants divided into high and low self-efficacy groups, results of multiple regression analyses showed that only the relationship between achievement-oriented leadership style and commitment ($\beta$=.14, p<.001) was significant in the high self-efficacy group. In the low self-efficacy group, the relationship between supportive leadership style and commitment ($\beta$=.17, p<.001),and the relationship between participative leadership style and commitment ($\beta$=.10, p<.001) were significant. The study focused on the franchisee's self-efficacy in order to explore the possibility that regulation, originally intended to protect the franchisee, may not be the most effective method to maintain the relationships in a franchise business. The key results of the data analysis regarding the moderating role of self-efficacy between leadership behavior style as proposed by path-goal and commitment theory were as follows. First, this study proposed that franchisor should apply the appropriate type of leadership behavior to strengthen the franchisees commitment because the results demonstrated that supportive and participative leadership styles by the franchisors have a positive influence on the franchisee's level of commitment. Second, it is desirable for franchisor to validate the franchisee's efforts, since the franchisee's characteristics such as self-efficacy had a substantial, positive effect on the franchisee's commitment as well as being a meaningful moderator between leadership and commitment. Third, the results as a whole imply that the government should provide institutional support, namely to put the franchisor in a position to clearly identify the characteristics of their franchisees and provide reasonable means to administer the franchisees to achieve the company's goal.

• Journal of Intelligence and Information Systems
• /
• v.18 no.4
• /
• pp.117-127
• /
• 2012

Due to the recent expansion of the Web 2.0 -based services, along with the widespread of smartphones, online social network services are being popularized among users. Online social network services are the online community services which enable users to communicate each other, share information and expand human relationships. In the social network services, each relation between users is represented by a graph consisting of nodes and links. As the users of online social network services are increasing rapidly, the SNS are actively utilized in enterprise marketing, analysis of social phenomenon and so on. Social Network Analysis (SNA) is the systematic way to analyze social relationships among the members of the social network using the network theory. In general social network theory consists of nodes and arcs, and it is often depicted in a social network diagram. In a social network diagram, nodes represent individual actors within the network and arcs represent relationships between the nodes. With SNA, we can measure relationships among the people such as degree of intimacy, intensity of connection and classification of the groups. Ever since Social Networking Services (SNS) have drawn increasing attention from millions of users, numerous researches have made to analyze their user relationships and messages. There are typical representative SNA methods: degree centrality, betweenness centrality and closeness centrality. In the degree of centrality analysis, the shortest path between nodes is not considered. However, it is used as a crucial factor in betweenness centrality, closeness centrality and other SNA methods. In previous researches in SNA, the computation time was not too expensive since the size of social network was small. Unfortunately, most SNA methods require significant time to process relevant data, and it makes difficult to apply the ever increasing SNS data in social network studies. For instance, if the number of nodes in online social network is n, the maximum number of link in social network is n(n-1)/2. It means that it is too expensive to analyze the social network, for example, if the number of nodes is 10,000 the number of links is 49,995,000. Therefore, we propose a heuristic-based method for finding the shortest path among users in the SNS user graph. Through the shortest path finding method, we will show how efficient our proposed approach may be by conducting betweenness centrality analysis and closeness centrality analysis, both of which are widely used in social network studies. Moreover, we devised an enhanced method with addition of best-first-search method and preprocessing step for the reduction of computation time and rapid search of the shortest paths in a huge size of online social network. Best-first-search method finds the shortest path heuristically, which generalizes human experiences. As large number of links is shared by only a few nodes in online social networks, most nods have relatively few connections. As a result, a node with multiple connections functions as a hub node. When searching for a particular node, looking for users with numerous links instead of searching all users indiscriminately has a better chance of finding the desired node more quickly. In this paper, we employ the degree of user node vn as heuristic evaluation function in a graph G = (N, E), where N is a set of vertices, and E is a set of links between two different nodes. As the heuristic evaluation function is used, the worst case could happen when the target node is situated in the bottom of skewed tree. In order to remove such a target node, the preprocessing step is conducted. Next, we find the shortest path between two nodes in social network efficiently and then analyze the social network. For the verification of the proposed method, we crawled 160,000 people from online and then constructed social network. Then we compared with previous methods, which are best-first-search and breath-first-search, in time for searching and analyzing. The suggested method takes 240 seconds to search nodes where breath-first-search based method takes 1,781 seconds (7.4 times faster). Moreover, for social network analysis, the suggested method is 6.8 times and 1.8 times faster than betweenness centrality analysis and closeness centrality analysis, respectively. The proposed method in this paper shows the possibility to analyze a large size of social network with the better performance in time. As a result, our method would improve the efficiency of social network analysis, making it particularly useful in studying social trends or phenomena.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.63-82
• /
• 2016

Individuals gather information online to resolve problems in their daily lives and make various decisions about the purchase of products or services. With the revolutionary development of information technology, Web 2.0 has allowed more people to easily generate and use online reviews such that the volume of information is rapidly increasing, and the usefulness and significance of analyzing the unstructured data have also increased. This paper presents an analysis on the lexical features of expert product reviews to determine their influence on consumers' purchasing decisions. The focus was on how unstructured data can be organized and used in diverse contexts through text mining. In addition, diverse lexical features of expert reviews of contents provided by a third-party review site were extracted and defined. Expert reviews are defined as evaluations by people who have expert knowledge about specific products or services in newspapers or magazines; this type of review is also called a critic review. Consumers who purchased products before the widespread use of the Internet were able to access expert reviews through newspapers or magazines; thus, they were not able to access many of them. Recently, however, major media also now provide online services so that people can more easily and affordably access expert reviews compared to the past. The reason why diverse reviews from experts in several fields are important is that there is an information asymmetry where some information is not shared among consumers and sellers. The information asymmetry can be resolved with information provided by third parties with expertise to consumers. Then, consumers can read expert reviews and make purchasing decisions by considering the abundant information on products or services. Therefore, expert reviews play an important role in consumers' purchasing decisions and the performance of companies across diverse industries. If the influence of qualitative data such as reviews or assessment after the purchase of products can be separately identified from the quantitative data resources, such as the actual quality of products or price, it is possible to identify which aspects of product reviews hamper or promote product sales. Previous studies have focused on the characteristics of the experts themselves, such as the expertise and credibility of sources regarding expert reviews; however, these studies did not suggest the influence of the linguistic features of experts' product reviews on consumers' overall evaluation. However, this study focused on experts' recommendations and evaluations to reveal the lexical features of expert reviews and whether such features influence consumers' overall evaluations and purchasing decisions. Real expert product reviews were analyzed based on the suggested methodology, and five lexical features of expert reviews were ultimately determined. Specifically, the "review depth" (i.e., degree of detail of the expert's product analysis), and "lack of assurance" (i.e., degree of confidence that the expert has in the evaluation) have statistically significant effects on consumers' product evaluations. In contrast, the "positive polarity" (i.e., the degree of positivity of an expert's evaluations) has an insignificant effect, while the "negative polarity" (i.e., the degree of negativity of an expert's evaluations) has a significant negative effect on consumers' product evaluations. Finally, the "social orientation" (i.e., the degree of how many social expressions experts include in their reviews) does not have a significant effect on consumers' product evaluations. In summary, the lexical properties of the product reviews were defined according to each relevant factor. Then, the influence of each linguistic factor of expert reviews on the consumers' final evaluations was tested. In addition, a test was performed on whether each linguistic factor influencing consumers' product evaluations differs depending on the lexical features. The results of these analyses should provide guidelines on how individuals process massive volumes of unstructured data depending on lexical features in various contexts and how companies can use this mechanism from their perspective. This paper provides several theoretical and practical contributions, such as the proposal of a new methodology and its application to real data.

• Journal of Society of Preventive Korean Medicine
• /
• v.5 no.2
• /
• pp.8-28
• /
• 2001

This investigation grasps the level and relevant elements of performance of health promotional activities for the elderly in Korea. It provides fundamental data on health promoting projects targeting the elderly population from farming villages. Hence, this study gropes for an effective approach and measures of health promoting programs. The program needs to be developed with a focus on elderly people from farming villages. In addition, it was carried out in order to provide basic data for development of health projects for local communities. Data gathering was based on survey data targeting patients from the free clinic service. Service was rendered for the residents of farming villages, and conducted at the Offices of CheonBuk Province from October 2000 to December 2000. Analytical results were used to examine the health promotional method for the elderly in the aspect of Oriental Medicine. SPSS 9.0 version as well as T-test and ANOVA were used for survey data analysis. Piersons correlation coefficient was utilized for the relationship for each area, obtaining the following analytical results. 1. The average score for the activities of health promotion was 2.28. Looking at each subcategory, stress management was the highest at 3.65; interpersonal relationship, 3.00; nutrition, 2.55; health responsibility, 2.15; self-realization, 2.03; and exercise was the lowest at 1.89. 2. With respect to lifestyle of the health promotion secondary to general features of elderly people from farming villages, the level of activities of health promoting lifestyle was shown to be higher for males than that of females. Self-realization area was high among males in detailed particulars while the level of execution was high as age decreases in the stress area. 3. Regarding health promoting life style secondary to socioeconomic characteristics, the level of execution was higher for the individuals with a higher level of education and further utilization of spare time. With respect to occupation, the level was highest for people from the fishery. The level decreased in the order of other occupations such as trade, unemployed and agriculture, which was shown to be the lowest. In detailed particulars, it revealed that higher the individuals educational level, the higher the self-realization and stress management areas. The level of interpersonal relationship was the highest among people with little or no education. With respect to self-realization area, the level was highest among the cases where one paid living expenses along with their children. The lowest level of living expenses was seen in the cases where an individual pays for living expenses by himself/herself. There were significant results in all areas except for nutrition areas depending on occupation. The fishery was shown to be the highest. The level of activities was higher as one utilizes more spare time in all areas except for the area of interpersonal relationship.