• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.101-116
• /
• 2015

Recently, due to the development of smart devices and social media, vast amounts of information with the various forms were accumulated. Particularly, considerable research efforts are being directed towards analyzing unstructured big data to resolve various social problems. Accordingly, focus of data-driven decision-making is being moved from structured data analysis to unstructured one. Also, in the field of recommendation system, which is the typical area of data-driven decision-making, the need of using unstructured data has been steadily increased to improve system performance. Approaches to improve the performance of recommendation systems can be found in two aspects- improving algorithms and acquiring useful data with high quality. Traditionally, most efforts to improve the performance of recommendation system were made by the former approach, while the latter approach has not attracted much attention relatively. In this sense, efforts to utilize unstructured data from variable sources are very timely and necessary. Particularly, as the interests of users are directly connected with their needs, identifying the interests of the user through unstructured big data analysis can be a crew for improving performance of recommendation systems. In this sense, this study proposes the methodology of improving recommendation system by measuring interests of the user. Specially, this study proposes the method to quantify interests of the user by analyzing user's internet usage patterns, and to predict user's repurchase based upon the discovered preferences. There are two important modules in this study. The first module predicts repurchase probability of each category through analyzing users' purchase history. We include the first module to our research scope for comparing the accuracy of traditional purchase-based prediction model to our new model presented in the second module. This procedure extracts purchase history of users. The core part of our methodology is in the second module. This module extracts users' interests by analyzing news articles the users have read. The second module constructs a correspondence matrix between topics and news articles by performing topic modeling on real world news articles. And then, the module analyzes users' news access patterns and then constructs a correspondence matrix between articles and users. After that, by merging the results of the previous processes in the second module, we can obtain a correspondence matrix between users and topics. This matrix describes users' interests in a structured manner. Finally, by using the matrix, the second module builds a model for predicting repurchase probability of each category. In this paper, we also provide experimental results of our performance evaluation. The outline of data used our experiments is as follows. We acquired web transaction data of 5,000 panels from a company that is specialized to analyzing ranks of internet sites. At first we extracted 15,000 URLs of news articles published from July 2012 to June 2013 from the original data and we crawled main contents of the news articles. After that we selected 2,615 users who have read at least one of the extracted news articles. Among the 2,615 users, we discovered that the number of target users who purchase at least one items from our target shopping mall 'G' is 359. In the experiments, we analyzed purchase history and news access records of the 359 internet users. From the performance evaluation, we found that our prediction model using both users' interests and purchase history outperforms a prediction model using only users' purchase history from a view point of misclassification ratio. In detail, our model outperformed the traditional one in appliance, beauty, computer, culture, digital, fashion, and sports categories when artificial neural network based models were used. Similarly, our model outperformed the traditional one in beauty, computer, digital, fashion, food, and furniture categories when decision tree based models were used although the improvement is very small.

• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.47-60
• /
• 2012

Video data comes in the form of the unstructured and the complex structure. As the importance of efficient management and retrieval for video data increases, studies on the video parsing based on the visual features contained in the video contents are researched to reconstruct video data as the meaningful structure. The early studies on video parsing are focused on splitting video data into shots, but detecting the shot boundary defined with the physical boundary does not cosider the semantic association of video data. Recently, studies on structuralizing video shots having the semantic association to the video scene defined with the semantic boundary by utilizing clustering methods are actively progressed. Previous studies on detecting the video scene try to detect video scenes by utilizing clustering algorithms based on the similarity measure between video shots mainly depended on color features. However, the correct identification of a video shot or scene and the detection of the gradual transitions such as dissolve, fade and wipe are difficult because color features of video data contain a noise and are abruptly changed due to the intervention of an unexpected object. In this paper, to solve these problems, we propose the Scene Detector by using Color histogram, corner Edge and Object color histogram (SDCEO) that clusters similar shots organizing same event based on visual features including the color histogram, the corner edge and the object color histogram to detect video scenes. The SDCEO is worthy of notice in a sense that it uses the edge feature with the color feature, and as a result, it effectively detects the gradual transitions as well as the abrupt transitions. The SDCEO consists of the Shot Bound Identifier and the Video Scene Detector. The Shot Bound Identifier is comprised of the Color Histogram Analysis step and the Corner Edge Analysis step. In the Color Histogram Analysis step, SDCEO uses the color histogram feature to organizing shot boundaries. The color histogram, recording the percentage of each quantized color among all pixels in a frame, are chosen for their good performance, as also reported in other work of content-based image and video analysis. To organize shot boundaries, SDCEO joins associated sequential frames into shot boundaries by measuring the similarity of the color histogram between frames. In the Corner Edge Analysis step, SDCEO identifies the final shot boundaries by using the corner edge feature. SDCEO detect associated shot boundaries comparing the corner edge feature between the last frame of previous shot boundary and the first frame of next shot boundary. In the Key-frame Extraction step, SDCEO compares each frame with all frames and measures the similarity by using histogram euclidean distance, and then select the frame the most similar with all frames contained in same shot boundary as the key-frame. Video Scene Detector clusters associated shots organizing same event by utilizing the hierarchical agglomerative clustering method based on the visual features including the color histogram and the object color histogram. After detecting video scenes, SDCEO organizes final video scene by repetitive clustering until the simiarity distance between shot boundaries less than the threshold h. In this paper, we construct the prototype of SDCEO and experiments are carried out with the baseline data that are manually constructed, and the experimental results that the precision of shot boundary detection is 93.3% and the precision of video scene detection is 83.3% are satisfactory.

• Journal of Intelligence and Information Systems
• /
• v.16 no.3
• /
• pp.55-75
• /
• 2010

Sequential pattern mining aims to discover interesting sequential patterns in a sequence database, and it is one of the essential data mining tasks widely used in various application fields such as Web access pattern analysis, customer purchase pattern analysis, and DNA sequence analysis. In general sequential pattern mining, only the generation order of data element in a sequence is considered, so that it can easily find simple sequential patterns, but has a limit to find more interesting sequential patterns being widely used in real world applications. One of the essential research topics to compensate the limit is a topic of weighted sequential pattern mining. In weighted sequential pattern mining, not only the generation order of data element but also its weight is considered to get more interesting sequential patterns. In recent, data has been increasingly taking the form of continuous data streams rather than finite stored data sets in various application fields, the database research community has begun focusing its attention on processing over data streams. The data stream is a massive unbounded sequence of data elements continuously generated at a rapid rate. In data stream processing, each data element should be examined at most once to analyze the data stream, and the memory usage for data stream analysis should be restricted finitely although new data elements are continuously generated in a data stream. Moreover, newly generated data elements should be processed as fast as possible to produce the up-to-date analysis result of a data stream, so that it can be instantly utilized upon request. To satisfy these requirements, data stream processing sacrifices the correctness of its analysis result by allowing some error. Considering the changes in the form of data generated in real world application fields, many researches have been actively performed to find various kinds of knowledge embedded in data streams. They mainly focus on efficient mining of frequent itemsets and sequential patterns over data streams, which have been proven to be useful in conventional data mining for a finite data set. In addition, mining algorithms have also been proposed to efficiently reflect the changes of data streams over time into their mining results. However, they have been targeting on finding naively interesting patterns such as frequent patterns and simple sequential patterns, which are found intuitively, taking no interest in mining novel interesting patterns that express the characteristics of target data streams better. Therefore, it can be a valuable research topic in the field of mining data streams to define novel interesting patterns and develop a mining method finding the novel patterns, which will be effectively used to analyze recent data streams. This paper proposes a gap-based weighting approach for a sequential pattern and amining method of weighted sequential patterns over sequence data streams via the weighting approach. A gap-based weight of a sequential pattern can be computed from the gaps of data elements in the sequential pattern without any pre-defined weight information. That is, in the approach, the gaps of data elements in each sequential pattern as well as their generation orders are used to get the weight of the sequential pattern, therefore it can help to get more interesting and useful sequential patterns. Recently most of computer application fields generate data as a form of data streams rather than a finite data set. Considering the change of data, the proposed method is mainly focus on sequence data streams.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.19-41
• /
• 2019

According to the rapidly increasing demand for text data analysis, research and investment in text mining are being actively conducted not only in academia but also in various industries. Text mining is generally conducted in two steps. In the first step, the text of the collected document is tokenized and structured to convert the original document into a computer-readable form. In the second step, tasks such as document classification, clustering, and topic modeling are conducted according to the purpose of analysis. Until recently, text mining-related studies have been focused on the application of the second steps, such as document classification, clustering, and topic modeling. However, with the discovery that the text structuring process substantially influences the quality of the analysis results, various embedding methods have actively been studied to improve the quality of analysis results by preserving the meaning of words and documents in the process of representing text data as vectors. Unlike structured data, which can be directly applied to a variety of operations and traditional analysis techniques, Unstructured text should be preceded by a structuring task that transforms the original document into a form that the computer can understand before analysis. It is called "Embedding" that arbitrary objects are mapped to a specific dimension space while maintaining algebraic properties for structuring the text data. Recently, attempts have been made to embed not only words but also sentences, paragraphs, and entire documents in various aspects. Particularly, with the demand for analysis of document embedding increases rapidly, many algorithms have been developed to support it. Among them, doc2Vec which extends word2Vec and embeds each document into one vector is most widely used. However, the traditional document embedding method represented by doc2Vec generates a vector for each document using the whole corpus included in the document. This causes a limit that the document vector is affected by not only core words but also miscellaneous words. Additionally, the traditional document embedding schemes usually map each document into a single corresponding vector. Therefore, it is difficult to represent a complex document with multiple subjects into a single vector accurately using the traditional approach. In this paper, we propose a new multi-vector document embedding method to overcome these limitations of the traditional document embedding methods. This study targets documents that explicitly separate body content and keywords. In the case of a document without keywords, this method can be applied after extract keywords through various analysis methods. However, since this is not the core subject of the proposed method, we introduce the process of applying the proposed method to documents that predefine keywords in the text. The proposed method consists of (1) Parsing, (2) Word Embedding, (3) Keyword Vector Extraction, (4) Keyword Clustering, and (5) Multiple-Vector Generation. The specific process is as follows. all text in a document is tokenized and each token is represented as a vector having N-dimensional real value through word embedding. After that, to overcome the limitations of the traditional document embedding method that is affected by not only the core word but also the miscellaneous words, vectors corresponding to the keywords of each document are extracted and make up sets of keyword vector for each document. Next, clustering is conducted on a set of keywords for each document to identify multiple subjects included in the document. Finally, a Multi-vector is generated from vectors of keywords constituting each cluster. The experiments for 3.147 academic papers revealed that the single vector-based traditional approach cannot properly map complex documents because of interference among subjects in each vector. With the proposed multi-vector based method, we ascertained that complex documents can be vectorized more accurately by eliminating the interference among subjects.

• Asia pacific journal of information systems
• /
• v.17 no.4
• /
• pp.187-206
• /
• 2007

The development of computer and information technology has been combined with the information superhighway internet infrastructure, so information widely spreads not only in special fields but also in the daily lives of people. Information ubiquity influences the traditional way of transaction, and leads a new E-commerce which distinguishes from the existing E-commerce. Not only goods as physical but also service as non-physical come into E-commerce. As the scale of E-Commerce is being enlarged as well. It keeps people from finding information they want. Recommender systems are now becoming the main tools for E-Commerce to mitigate the information overload. Recommender systems can be defined as systems for suggesting some Items(goods or service) considering customers' interests or tastes. They are being used by E-commerce web sites to suggest products to their customers who want to find something for them and to provide them with information to help them decide which to purchase. There are several approaches of recommending goods to customer in recommender system but in this study, the main subject is focused on collaborative filtering technique. This study presents a possibility of pre-evaluation for the prediction performance of customer's preference in collaborative filtering before the process of customer's preference prediction. Pre-evaluation for the prediction performance of each customer having low performance is classified by using the statistical features of ratings rated by each customer is conducted before the prediction process. In this study, MovieLens 100K dataset is used to analyze the accuracy of classification. The classification criteria are set by using the training sets divided 80% from the 100K dataset. In the process of classification, the customers are divided into two groups, classified group and non classified group. To compare the prediction performance of classified group and non classified group, the prediction process runs the 20% test set through the Neighborhood Based Collaborative Filtering Algorithm and Correspondence Mean Algorithm. The prediction errors from those prediction algorithm are allocated to each customer and compared with each user's error. Research hypothesis : Two research hypotheses are formulated in this study to test the accuracy of the classification criterion as follows. Hypothesis 1: The estimation accuracy of groups classified according to the standard deviation of each user's ratings has significant difference. To test the Hypothesis 1, the standard deviation is calculated for each user in training set which is divided 80% from MovieLens 100K dataset. Four groups are classified according to the quartile of the each user's standard deviations. It is compared to test the estimation errors of each group which results from test set are significantly different. Hypothesis 2: The estimation accuracy of groups that are classified according to the distribution of each user's ratings have significant differences. To test the Hypothesis 2, the distributions of each user's ratings are compared with the distribution of ratings of all customers in training set which is divided 80% from MovieLens 100K dataset. It assumes that the customers whose ratings' distribution are different from that of all customers would have low performance, so six types of different distributions are set to be compared. The test groups are classified into fit group or non-fit group according to the each type of different distribution assumed. The degrees in accordance with each type of distribution and each customer's distributions are tested by the test of ${\chi}^2$ goodness-of-fit and classified two groups for testing the difference of the mean of errors. Also, the degree of goodness-of-fit with the distribution of each user's ratings and the average distribution of the ratings in the training set are closely related to the prediction errors from those prediction algorithms. Through this study, the customers who have lower performance of prediction than the rest in the system are classified by those two criteria, which are set by statistical features of customers ratings in the training set, before the prediction process.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.4
• /
• pp.242-255
• /
• 2006

In the recent years, Popularization of mobile devices such as Smart Phones, PDAs and MP3 Players causes rapid increasing necessity of Power management technology because it is most essential factor of mobile devices. On the other hand, despite low price, hard disk has large capacity and high speed. Even it can be made small enough today, too. So it appropriates mobile devices. but it consumes too much power to embed In mobile devices. Due to these motivations, in this paper we had suggested methods of minimizing Power consumption while playing multimedia data in the disk media for real-time and we evaluated what we had suggested. Strict limitation of power consumption of mobile devices has a big impact on designing both hardware and software. One difference between real-time multimedia streaming data and legacy text based data is requirement about continuity of data supply. This fact is why disk drive must persist in active state for the entire playback duration, from power management point of view; it nay be a great burden. A legacy power management function of mobile disk drive affects quality of multimedia playback negatively because of excessive I/O requests when the disk is in standby state. Therefore, in this paper, we analyze power consumption profile of disk drive in detail, and we develop the algorithm which can play multimedia data effectively using less power. This algorithm calculates number of data block to be read and time duration of active/standby state. From this, the algorithm suggested in this paper does optimal scheduling that is ensuring continual playback of data blocks stored in mobile disk drive. And we implement our algorithms in publicly available MPEG player software. This MPEG player software saves up to 60% of power consumption as compared with full-time active stated disk drive, and 38% of power consumption by comparison with disk drive controlled by native power management method.

• Journal of Bio-Environment Control
• /
• v.22 no.2
• /
• pp.138-145
• /
• 2013

In order to develop the efficient control algorithm of the two-fluid fogging system, cooling experiments for the many different types of fogging cycles were conducted in tomato greenhouses. It showed that the cooling effect was 1.2 to $4.0^{\circ}C$ and the cooling efficiency was 8.2 to 32.9% on average. The cooling efficiency with fogging interval was highest in the case of the fogging cycle of 90 seconds. The cooling efficiency showed a tendency to increase as the fogging time increased and the stopping time decreased. As the spray rate of fog in the two-fluid fogging system increased, there was a tendency for the cooling efficiency to improve. However, as the inside air approaches its saturation level, even though the spray rate of fog increases, it does not lead to further evaporation. Thus, it can be inferred that increasing the spray rate of fog before the inside air reaches the saturation level could make higher the cooling efficiency. As cooling efficiency increases, the saturation deficit of inside air decreased and the difference between absolute humidity of inside and outside air increased. The more fog evaporated, the difference between absolute humidity of inside and outside air tended to increase and as the result, the discharge of vapor due to ventilation occurs more easily, which again lead to an increase in the evaporation rate and ultimately increase in the cooling efficiency. Regression analysis result on the saturation deficit of inside air showed that the fogging time needed to change of saturation deficit of $10g{\cdot}kg^{-1}$ was 120 seconds and stopping time was 60 seconds. But in order to decrease the amplitude of temperature and to increase the cooling efficiency, the fluctuation range of saturation deficit was set to $5g{\cdot}kg^{-1}$ and we decided that the fogging-stopping time of 60-30 seconds was more appropriate. Control types of two-fluid fogging systems were classified as computer control or simple control, and their control algorithms were derived. We recommend that if the two-fluid fogging system is controlled by manipulating only the set point of temperature, humidity, and on-off time, it would be best to set up the on-off time at 60-30 seconds in time control, the lower limit of air temperature at 30 to $32^{\circ}C$ and the upper limit of relative humidity at 85 to 90%.

• Investigative Magnetic Resonance Imaging
• /
• v.18 no.4
• /
• pp.303-313
• /
• 2014

Purpose : In-vivo conductivity reconstruction using transmit field ($B_1{^+}$) information of MRI was proposed. We assessed the accuracy of conductivity reconstruction in the presence of statistical noise in complex $B_1{^+}$ map and provided a parametric model of the conductivity-to-noise ratio value. Materials and Methods: The $B_1{^+}$ distribution was simulated for a cylindrical phantom model. By adding complex Gaussian noise to the simulated $B_1{^+}$ map, quantitative conductivity estimation error was evaluated. The quantitative evaluation process was repeated over several different parameters such as Larmor frequency, object radius and SNR of $B_1{^+}$ map. A parametric model for the conductivity-to-noise ratio was developed according to these various parameters. Results: According to the simulation results, conductivity estimation is more sensitive to statistical noise in $B_1{^+}$ phase than to noise in $B_1{^+}$ magnitude. The conductivity estimate of the object of interest does not depend on the external object surrounding it. The conductivity-to-noise ratio is proportional to the signal-to-noise ratio of the $B_1{^+}$ map, Larmor frequency, the conductivity value itself and the number of averaged pixels. To estimate accurate conductivity value of the targeted tissue, SNR of $B_1{^+}$ map and adequate filtering size have to be taken into account for conductivity reconstruction process. In addition, the simulation result was verified at 3T conventional MRI scanner. Conclusion: Through all these relationships, quantitative conductivity estimation error due to statistical noise in $B_1{^+}$ map is modeled. By using this model, further issues regarding filtering and reconstruction algorithms can be investigated for MREPT.

• Sleep Medicine and Psychophysiology
• /
• v.6 no.1
• /
• pp.52-60
• /
• 1999

Objectives: With the object of finding the appropriate conditions and algorithms for dimensional analysis of human EEG, we calculated correlation dimensions in the various condition of sampling rate and data aquisition time and improved the computation algorithm by taking advantage of bit operation instead of log operation. Methods: EEG signals from 13 scalp lead of a man were digitized with A-D converter under the condition of 12 bit resolution and 1000 Hertz of sampling rate during 32 seconds. From the original data, we made 15 time series data which have different sampling rate of 62.5, 125, 250, 500, 1000 hertz and data acqusition time of 10, 20, 30 second, respectively. New algorithm to shorten the calculation time using bit operation and the Least Trimmed Squares(LTS) estimator to get the optimal slope was applied to these data. Results: The values of the correlation dimension showed the increasing pattern as the data acquisition time becomes longer. The data with sampling rate of 62.5 Hz showed the highest value of correlation dimension regardless of sampling time but the correlation dimension at other sampling rates revealed similar values. The computation with bit operation instead of log operation had a statistically significant effect of shortening of calculation time and LTS method estimated more stably the slope of correlation dimension than the Least Squares estimator. Conclusion: The bit operation and LTS methods were successfully utilized to time-saving and efficient calculation of correlation dimension. In addition, time series of 20-sec length with sampling rate of 125 Hz was adequate to estimate the dimensional complexity of human EEG.

• Progress in Medical Physics
• /
• v.13 no.3
• /
• pp.149-155
• /
• 2002

Recent advances in radiation transport algorithms, computer hardware performance, and parallel computing make the clinical use of Monte Carlo based dose calculations possible. To compare the speed and accuracies of dose calculations between different developed codes, a benchmark tests were proposed at the XIIth ICCR (International Conference on the use of Computers in Radiation Therapy, Heidelberg, Germany 2000). A Monte Carlo treatment planning comprised of 28 various Intel Pentium CPUs was implemented for routine clinical use. The purpose of this study was to evaluate the performance of our system using the above benchmark tests. The benchmark procedures are comprised of three parts. a) speed of photon beams dose calculation inside a given phantom of 30.5 cm$\times$39.5 cm $\times$ 30 cm deep and filled with 5 ㎣ voxels within 2% statistical uncertainty. b) speed of electron beams dose calculation inside the same phantom as that of the photon beams. c) accuracy of photon and electron beam calculation inside heterogeneous slab phantom compared with the reference results of EGS4/PRESTA calculation. As results of the speed benchmark tests, it took 5.5 minutes to achieve less than 2% statistical uncertainty for 18 MV photon beams. Though the net calculation for electron beams was an order of faster than the photon beam, the overall calculation time was similar to that of photon beam case due to the overhead time to maintain parallel processing. Since our Monte Carlo code is EGSnrc, which is an improved version of EGS4, the accuracy tests of our system showed, as expected, very good agreement with the reference data. In conclusion, our Monte Carlo treatment planning system shows clinically meaningful results. Though other more efficient codes are developed such like MCDOSE and VMC＋＋, BEAMnrc based on EGSnrc code system may be used for routine clinical Monte Carlo treatment planning in conjunction with clustering technique.