• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.8
• /
• pp.321-328
• /
• 2014

In this paper, we present a MapReduce-based mobility pattern mining system which can predict efficiently the next place of mobile users. It learns the mobility pattern model of each user, represented by Hidden Markov Models(HMM), from a large-scale trajectory dataset, and then predicts the next place for the user to visit by applying the learned models to the current trajectory. Our system consists of two parts: the back-end part, in which the mobility pattern models are learned for individual users, and the front-end part, where the next place for a certain user to visit is predicted based on the mobility pattern models. While the back-end part comprises of three distinct MapReduce modules for POI extraction, trajectory transformation, and mobility pattern model learning, the front-end part has two different modules for candidate route generation and next place prediction. Map and reduce functions of each module in our system were designed to utilize the underlying Hadoop infrastructure enough to maximize the parallel processing. We performed experiments to evaluate the performance of the proposed system by using a large-scale open benchmark dataset, GeoLife, and then could make sure of high performance of our system as results of the experiments.

• Journal of Internet Computing and Services
• /
• v.10 no.2
• /
• pp.1-13
• /
• 2009

Intrusion detection is a process that identifies the attacks and responds to the malicious intrusion actions for the protection of the computer and the network resources. Due to the fast development of the Internet, the types of intrusions become more complex recently and need immediate and correct responses because the frequent occurrences of a new intrusion type rise rapidly. Therefore, to solve these problems of the intrusion detection systems, we propose a sequential pattern miner for analysis of the alert data in order to support intelligent and automatic detection of the intrusion. Sequential pattern mining is one of the methods to find the patterns among the extracted items that are frequent in the fixed sequences. We apply the prefixSpan algorithm to find out the alert sequences. This method can be used to predict the actions of the sequential patterns and to create the rules of the intrusions. In this paper, we propose an extended prefixSpan algorithm which is designed to consider the specific characteristics of the alert data. The extended sequential pattern miner will be used as a part of alert data analyzer of intrusion detection systems. By using the created rules from the sequential pattern miner, the HA(high-level alert analyzer) of PEP(policy enforcement point), usually called IDS, performs the prediction of the sequence behaviors and changing patterns that were not visibly checked.

• Journal of Digital Convergence
• /
• v.16 no.6
• /
• pp.85-93
• /
• 2018

The Latent Growth Modeling(LGM) is known as the arising analysis method of longitudinal data and it could be classified into unconditional model and conditional model. Unconditional model requires estimated value of intercept and slope to complete a model of fitness. However, the existing LGM is in absence of a structured methodology to estimate slope when longitudinal data is neither simple linear function nor the pre-defined function. This study used Sequential Pattern of Association Rule Mining to calculate slope of unconditional model. The applied dataset is 'the Youth Panel 2001-2006' from Korea Employment Information Service. The proposed methodology was able to identify increasing fitness of the model comparing to the existing simple linear function and visualizing process of slope estimation.

• The Journal of Society for e-Business Studies
• /
• v.23 no.2
• /
• pp.97-110
• /
• 2018

With the advent of the 4-th industrial revolution, manufacturing companies have increasing interests in the realization of smart manufacturing by utilizing their accumulated facilities data. However, most previous research dealt with the structured data such as sensor signals, and only a little focused on the unstructured data such as text, which actually comprises a large portion of the accumulated data. Therefore, we propose an association rule mining based facility error pattern extraction framework, where text data written by operators are analyzed. Specifically, phrases were extracted and utilized as a unit for text data analysis since a word, which normally used as a unit for text data analysis, is unable to deliver the technical meanings of facility errors. Performances of the proposed framework were evaluated by addressing a real-world case, and it is expected that the productivity of manufacturing companies will be enhanced by adopting the proposed framework.

• Journal of Korean Institute of Industrial Engineers
• /
• v.38 no.3
• /
• pp.191-197
• /
• 2012

One-class classification (OCC) is one of the recent growing areas in data mining and pattern recognition. In the present study we examine a k-nearest neighbors data description (kNNDD) algorithm, one of the OCC algorithms widely used. In particular, we propose to use nonparametric estimation methods to determine the threshold of the kNNDD algorithm. A simulation study has been conducted to explore the characteristics of the proposed approach and compare it with the existing approach that determines the threshold. The results demonstrate the usefulness and flexibility of the proposed approach.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.4
• /
• pp.985-995
• /
• 2015

Add-on selling can provide new chances to increase sellers' profits and meet customers' needs. Although prior studies have advocated add-on selling for its business value, there is an argument that add-on selling can cause customer repulsion. Therefore, we need to understand customer purchasing pattern related to add-on selling in order to promote it and to mitigate the customer repulsion. To that end, we applied data mining techniques to the 24,925 transactions of data from an online open market in Korea. We then conducted feature selection to investigate the most influential factors that can explain the characteristics of add-on selling transactions using a classification model. We also identified association rules among add-on selling and promotions. Finally, based on the findings in our experiments, we proposed add-on selling strategies for the target online market.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.713-716
• /
• 2005

In this paper, in order to identify and recognize attack patterns, we propose a Bayesian classification using frequent patterns. In theory, Bayesian classifiers guarantee the minimum error rate compared to all other classifiers. However, in practice this is not always the case owing to inaccuracies in the unrealistic assumption{ class conditional independence) made for its use. Our method addresses the problem of attribute dependence by discovering frequent patterns. It generates frequent patterns using an efficient FP-growth approach. Since the volume of patterns produced can be large, we propose a pruning technique for selection only interesting patterns. Also, this method estimates the probability of a new case using different product approximations, where each product approximation assumes different independence of the attributes. Our experiments show that the proposed classifier achieves higher accuracy and is more efficient than other classifiers.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.2
• /
• pp.209-215
• /
• 2003

A new GA-based methodology using information granules is suggested for the construction of fuzzy classifiers. The proposed scheme consists of three steps: selection of information granules, construction of the associated fuzzy sets, and tuning of the fuzzy rules. First, the genetic algorithm (GA) is applied to the development of the adequate information granules. The fuzzy sets are then constructed from the analysis of the developed information granules. An interpretable fuzzy classifier is designed by using the constructed fuzzy sets. Finally, the GA are utilized for tuning of the fuzzy rules, which can enhance the classification performance on the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example, the classification of the Iris data, is provided.

• Asia pacific journal of information systems
• /
• v.12 no.1
• /
• pp.1-16
• /
• 2002

Most of searching systems used by shopping-mall provide too much information for user requirements or fail to provide appropriate items reflecting customer's preference. This paper aims to design and implement the product searching systems based on customer preference which will enable efficient product selection in the internet shopping-mall. The proposed system consists of user/provider interface, searching and model agent, data management system, and model management system. Especially, we construct the searching pattern database to support fast search using association mining method. And this system includes the customer-oriented decision model which shows the highly preferred products. Input weight value per attribute and preference level should be needed to compute priority grade of preference.

• International Journal of Advanced Culture Technology
• /
• v.5 no.3
• /
• pp.40-45
• /
• 2017

This data mining technique was used to extract useful information from percutaneous coronary intervention data obtained from the US public data homepage. The experiment was performed by extracting data on the area, frequency of operation, and the number of deaths. It led us to finding of meaningful correlations, patterns, and trends using various algorithms, pattern techniques, and statistical techniques. In this paper, information is obtained through efficient decision tree and cluster analysis in predicting the incidence of percutaneous coronary intervention and mortality. In the cluster analysis, EM algorithm was used to evaluate the suitability of the algorithm for each situation based on performance tests and verification of results. In the cluster analysis, the experimental data were classified using the EM algorithm, and we evaluated which models are more effective in comparing functions. Using data mining technique, it was identified which areas had effective treatment techniques and which areas were vulnerable, and we can predict the frequency and mortality of percutaneous coronary intervention for heart disease.