• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.23-31
• /
• 2022

This study analyzed the actual frozen container operation data of Starcool provided by H Shipping. Through interviews with H's field experts, only Critical and Fatal Alarms among the four failure alarms were defined as failures, and it was confirmed that using all variables due to the nature of frozen containers resulted in cost inefficiency. Therefore, this study proposes a method for detecting failure of frozen containers through characteristic importance and PCA techniques. To improve the performance of the model, we select variables based on feature importance through tree series models such as XGBoost and LGBoost, and use PCA to reduce the dimension of the entire variables for each model. The boosting-based XGBoost and LGBoost techniques showed that the results of the model proposed in this study improved the reproduction rate by 0.36 and 0.39 respectively compared to the results of supervised learning using all 62 variables.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.05a
• /
• pp.443-446
• /
• 2007

A new association rule mining algorithm, which reflects the strategic importance of associative relationships between items, was developed and presented in this paper. This algorithm exploits the basic framework of Apriori procedures and TSAA(transitive support association Apriori) procedure developed by Hyun and Choi in evaluating non-frequent itemsets. The algorithm considers the strategic importance(weight) of feature variables in the association rule mining process. Sample feature variables of strategic importance include: profitability, marketing value, customer satisfaction, and frequency. A database with 730 transaction data set of a large scale discount store was used to compare and verify the performance of the presented algorithm against the existing Apriori and TSAA algorithms. The result clearly indicated that the new algorithm produced substantially different association itemsets according to the weights assigned to the strategic feature variables.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.4
• /
• pp.593-603
• /
• 2020

The advent of malicious code has increased exponentially due to the spread of malicious code generation tools in accordance with the development of the network, but there is a limit to the response through existing malicious code detection methods. According to this situation, a machine learning-based malicious code detection method is evolving, and in this paper, the feature of data is extracted from the PE header for machine-learning-based malicious code detection, and then it is used to automate the malware through autoencoder. Research on how to extract the indicated features and feature importance. In this paper, 549 features composed of information such as DLL/API that can be identified from PE files that are commonly used in malware analysis are extracted, and autoencoder is used through the extracted features to improve the performance of malware detection in machine learning. It was proved to be successful in providing excellent accuracy and reducing the processing time by 2 times by effectively extracting the features of the data by compressively storing the data. The test results have been shown to be useful for classifying malware groups, and in the future, a classifier such as SVM will be introduced to continue research for more accurate malware detection.

• The KIPS Transactions:PartB
• /
• v.15B no.4
• /
• pp.331-340
• /
• 2008

Feature Selection is one of methods to improve the classification accuracy of data in the field of machine learning. Many feature selection algorithms have been proposed and discussed for years. However, the problem of finding the optimal feature subset from full data still remains to be a difficult problem. Bio-inspired algorithms are well-known evolutionary algorithms based on the principles of behavior of organisms, and very useful methods to find the optimal solution in optimization problems. Bio-inspired algorithms are also used in the field of feature selection problems. So in this paper we proposed new improved bio-inspired algorithms for feature selection. We used well-known bio-inspired algorithms, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), to find the optimal subset of features that shows the best performance in classification accuracy. In addition, we modified the bio-inspired algorithms considering the prior importance (prior relevance) of each feature. We chose the mRMR method, which can measure the goodness of single feature, to set the prior importance of each feature. We modified the evolution operators of GA and PSO by using the prior importance of each feature. We verified the performance of the proposed methods by experiment with datasets. Feature selection methods using GA and PSO produced better performances in terms of the classification accuracy. The modified method with the prior importance demonstrated improved performances in terms of the evolution speed and the classification accuracy.

• Journal of Korean Society on Water Environment
• /
• v.38 no.6
• /
• pp.292-305
• /
• 2022

The Yeongsan River, one of the four largest rivers in South Korea, has been facing difficulties with water quality management with respect to algal bloom. The algal bloom menace has become bigger, especially after the construction of two weirs in the mainstream of the Yeongsan River. Therefore, the prediction and factor analysis of Chlorophyll-a (Chl-a) concentration is needed for effective water quality management. In this study, Chl-a prediction model was developed, and the performance evaluated using machine and deep learning methods, such as Deep Neural Network (DNN), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost). Moreover, the correlation analysis and the feature importance results were compared to identify the major factors affecting the concentration of Chl-a. All models showed high prediction performance with an R² value of 0.9 or higher. In particular, XGBoost showed the highest prediction accuracy of 0.95 in the test data.The results of feature importance suggested that Ammonia (NH₃-N) and Phosphate (PO₄-P) were common major factors for the three models to manage Chl-a concentration. From the results, it was confirmed that three machine learning methods, DNN, RF, and XGBoost are powerful methods for predicting water quality parameters. Also, the comparison between feature importance and correlation analysis would present a more accurate assessment of the important major factors.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.6
• /
• pp.1058-1062
• /
• 2008

In clinical decision support system(CDSS), unlike rule-based expert method, appropriate data-driven machine learning method can easily provide the information of individual feature(clinical test) for disease classification. However, currently developed methods focus on the improvement of the classification accuracy for diagnosis. With the analysis of feature importance in classification, one may infer the novel clinical test sets which highly differentiate the specific diseases or disease states. In this background, we introduce a novel CDSS that integrate a classifier and feature selection module together. Random forest algorithm is applied for the classifier and the feature importance measure. The system selects the significant clinical tests discriminating the diseases by examining the classification error during backward elimination of the features. The superior performance of random forest algorithm in clinical classification was assessed against artificial neural network and decision tree algorithm by using breast cancer, diabetes and heart disease data in UCI Machine Learning Repository. The test with the same data sets shows that the proposed system can successfully select the significant clinical test set for each disease.

• Electronics and Telecommunications Trends
• /
• v.36 no.2
• /
• pp.65-74
• /
• 2021

In this paper, we discussed the necessity and importance of introducing feature stores to establish a collaborative environment between data engineering work and data science work. We examined the technology trends of feature stores by analyzing the status of some major feature stores. Moreover, by introducing a feature store, we can reduce the cost of performing artificial intelligence (AI) projects and improve the performance and reliability of AI models and the convenience of model operation. The future task is to establish technical requirements for establishing a collaborative environment between data engineering work and data science work and develop a solution for providing a collaborative environment based on this.

• Advances in Energy Research
• /
• v.6 no.2
• /
• pp.91-101
• /
• 2019

This paper proposes an efficient data-driven approach to build models for predicting energy consumption in buildings. Data used in this research is collected by installing humidity and temperature sensors at different locations in a building. In addition to this, weather data from nearby weather station is also included in the dataset to study the impact of weather conditions on energy consumption. One of the main emphasize of this research is to make feature selection independent of domain knowledge. Therefore, to extract useful features from data, two different approaches are tested: one is feature selection through principal component analysis and second is relative importance-based feature selection in original domain. The regression model used in this research is gradient boosting regression and its optimal parameters are chosen through a two staged coarse-fine search approach. In order to evaluate the performance of model, different performance evaluation metrics like r2-score and root mean squared error are used. Results have shown that best performance is achieved, when relative importance-based feature selection is used with gradient boosting regressor. Results of proposed technique has also outperformed the results of support vector machines and neural network-based approaches tested on the same dataset.

• Journal of Information Technology Services
• /
• v.18 no.2
• /
• pp.143-159
• /
• 2019

Machine learning (ML) is a method of fitting given data to a mathematical model to derive insights or to predict. In the age of big data, where the amount of available data increases exponentially due to the development of information technology and smart devices, ML shows high prediction performance due to pattern detection without bias. The feature engineering that generates the features that can explain the problem to be solved in the ML process has a great influence on the performance and its importance is continuously emphasized. Despite this importance, however, it is still considered a difficult task as it requires a thorough understanding of the domain characteristics as well as an understanding of source data and the iterative procedure. Therefore, we propose methods to apply deep learning for solving the complexity and difficulty of feature extraction and improving the performance of ML model. Unlike other techniques, the most common reason for the superior performance of deep learning techniques in complex unstructured data processing is that it is possible to extract features from the source data itself. In order to apply these advantages to the business problems, we propose deep learning based methods that can automatically extract features from transaction data or directly predict and classify target variables. In particular, we applied techniques that show high performance in existing text processing based on the structural similarity between transaction data and text data. And we also verified the suitability of each method according to the characteristics of transaction data. Through our study, it is possible not only to search for the possibility of automated feature extraction but also to obtain a benchmark model that shows a certain level of performance before performing the feature extraction task by a human. In addition, it is expected that it will be able to provide guidelines for choosing a suitable deep learning model based on the business problem and the data characteristics.