• Journal of Information Processing Systems
• /
• v.4 no.1
• /
• pp.17-26
• /
• 2008

This research proposes a new strategy where documents are encoded into string vectors and modified version of KNN to be adaptable to string vectors for text categorization. Traditionally, when KNN are used for pattern classification, raw data should be encoded into numerical vectors. This encoding may be difficult, depending on a given application area of pattern classification. For example, in text categorization, encoding full texts given as raw data into numerical vectors leads to two main problems: huge dimensionality and sparse distribution. In this research, we encode full texts into string vectors, and modify the supervised learning algorithms adaptable to string vectors for text categorization.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.979-993
• /
• 2021

Acoustic scene classification (ASC) categorizes an audio file based on the environment in which it has been recorded. This has long been studied in the detection and classification of acoustic scenes and events (DCASE). In this study, we considered the problem that ASC faces in real-world applications that the model used should have low-complexity. We compared several models that apply light-weight techniques. First, a base CNN model was proposed using log mel-spectrogram, deltas, and delta-deltas features. Second, depthwise separable convolution, linear bottleneck inverted residual block was applied to the convolutional layer, and Quantization was applied to the models to develop a low-complexity model. The model considering low-complexity was similar or slightly inferior to the performance of the base model, but the model size was significantly reduced from 503 KB to 42.76 KB.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.163-169
• /
• 2020

Recently, research on automation and unmanned operation of machines in the industrial field has been conducted with the advent of AI, Big data, and the IoT, which are the core technologies of the Fourth Industrial Revolution. The machines for these automation processes are controlled based on the data collected from the sensors attached to them, and further, the processes are managed. Conventionally, the abnormalities of sensors are periodically checked and managed. However, due to various environmental factors and situations in the industrial field, there are cases where the inspection due to the failure is not missed or failures are not detected to prevent damage due to sensor failure. In addition, even if a failure occurs, it is not immediately detected, which worsens the process loss. Therefore, in order to prevent damage caused by such a sudden sensor failure, it is necessary to identify the failure of the sensor in an embedded system in real-time and to diagnose the failure and determine the type for a quick response. In this paper, a deep neural network-based fault diagnosis system is designed and implemented using Raspberry Pi to classify typical sensor fault types such as erratic fault, hard-over fault, spike fault, and stuck fault. In order to diagnose sensor failure, the network is constructed using Google's proposed Inverted residual block structure of MobilieNetV2. The proposed scheme reduces memory usage and improves the performance of the conventional CNN technique to classify sensor faults.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.46.3-47
• /
• 2021

Recently a multilayer spectral inversion (MLSI) model has been proposed to infer the physical parameters of plasmas in the solar chromosphere. The inversion solves a three-layer radiative transfer model using the strong absorption line profiles, H alpha and Ca II 8542 Å, taken by the Fast Imaging Solar Spectrograph (FISS). The model successfully provides the physical plasma parameters, such as source functions, Doppler velocities, and Doppler widths in the layers of the photosphere to the chromosphere. However, it is quite expensive to apply the MLSI to a huge number of line profiles. For example, the calculating time is an hour to several hours depending on the size of the scan raster. We apply deep neural network (DNN) to the inversion code to reduce the cost of calculating the physical parameters. We train the models using pairs of absorption line profiles from FISS and their 13 physical parameters (source functions, Doppler velocities, Doppler widths in the chromosphere, and the pre-determined parameters for the photosphere) calculated from the spectral inversion code for 49 scan rasters (~2,000,000 dataset) including quiet and active regions. We use fully connected dense layers for training the model. In addition, we utilize a skip connection to avoid a problem of vanishing gradients. We evaluate the model by comparing the pairs of absorption line profiles and their inverted physical parameters from other quiet and active regions. Our result shows that the deep learning model successfully reproduces physical parameter maps of a scan raster observation per second within 15% of mean absolute percentage error and the mean squared error of 0.3 to 0.003 depending on the parameters. Taking this advantage of high performance of the deep learning model, we plan to provide the physical parameter maps from the FISS observations to understand the chromospheric plasma conditions in various solar features.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1835-1845
• /
• 2021

To accurately identify black ice and warn the drivers of information in advance so they can control speed and take preventive measures. In this paper, we propose a lightweight black ice detection network based on infrared road images. A black ice recognition network model based on CNN transfer learning has been developed. Additionally, to further improve the accuracy of black ice recognition, an enhanced lightweight network based on MobileNetV2 has been developed. To reduce the amount of calculation, linear bottlenecks and inverse residuals was used, and four bottleneck groups were used. At the same time, to improve the recognition rate of the model, each bottleneck group was connected to a 3×3 convolutional layer to enhance regional feature extraction and increase the number of feature maps. Finally, a black ice recognition experiment was performed on the constructed infrared road black ice dataset. The network model proposed in this paper had an accurate recognition rate of 99.07% for black ice.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.2
• /
• pp.13-17
• /
• 2024

In this paper, union-based rule antecedent neuro-fuzzy controller, which can guarantee a parsimonious knowledge base with reduced number of rules, is proposed. The proposed neuro-fuzzy controller allows union operation of input fuzzy sets in the antecedents to cover bigger input domain compared with the complete structure rule which consists of AND combination of all input variables in its premise. To construct the proposed neuro-fuzzy controller, we consider the multiple-term unified logic processor (MULP) which consists of OR and AND fuzzy neurons. The fuzzy neurons exhibit learning abilities as they come with a collection of adjustable connection weights. In the development stage, the genetic algorithm (GA) constructs a Boolean skeleton of the proposed neuro-fuzzy controller, while the stochastic reinforcement learning refines the binary connections of the GA-optimized controller for further improvement of the performance index. An inverted pendulum system is considered to verify the effectiveness of the proposed method by simulation and experiment.

• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1153-1164
• /
• 2024

In recent years, unmanned ground vehicles (UGVs) have been used to search for lost or stolen radioactive sources to avoid radiation exposure for operators. To achieve autonomous localization of radioactive sources, the UGVs must have the ability to automatically determine the next radiation measurement location instead of following a predefined path. Also, the radiation field of radioactive sources has to be reconstructed or inverted utilizing discrete measurements to obtain the radiation intensity distribution in the area of interest. In this study, we propose an effective source localization framework and method, in which UGVs are able to autonomously explore in the radiation area to determine the location of radioactive sources through an iterative process: path planning, radiation field reconstruction and estimation of source location. In the search process, the next radiation measurement point of the UGVs is fully predicted by the design path planning algorithm. After obtaining the measurement points and their radiation measurements, the radiation field of radioactive sources is reconstructed by the Gaussian process regression (GPR) model based on machine learning method. Based on the reconstructed radiation field, the locations of radioactive sources can be determined by the peak analysis method. The proposed method is verified through extensive simulation experiments, and the real source localization experiment on a Cs-137 point source shows that the proposed method can accurately locate the radioactive source with an error of approximately 0.30 m. The experimental results reveal the important practicality of our proposed method for source autonomous localization tasks.

• Journal of Korea Trade
• /
• v.24 no.1
• /
• pp.82-98
• /
• 2020

Purpose - This study aims to overcome the limitations of existing studies, which linearly determine the precedence factors of competency in overseas subsidiaries. The research objectives are as follows. First, what kind of nonlinear effects does the level of control held by Korean headquarters over foreign subsidiaries have in terms of competency in the subsidiaries? Second, what kind of nonlinear effects do the local experiences of overseas subsidiaries have on their competency? Design/methodology - With data on Korean multinational corporations (MNCs), this paper analyzes the effects of control levels of headquarters (HQs) and host-country experiences of foreign subsidiaries regarding competency in overseas subsidiaries. In particular, this study focuses on nonlinear models, differentiating it from previous studies. In order to examine research hypotheses, this study conducted a survey of overseas subsidiaries of Korean corporations. Surveys were conducted through various methods including e-mail, online questionnaires, fax, and telephone calls. Copies of the questionnaire were distributed to a total of 2,246 overseas subsidiaries, and 409 completed responses were collected. Excluding 15 copies that were insufficiently answered, responses from a total of 394 copies were used for analysis. Findings - This study presents the following results. First, there is a U-shaped relationship between levels of HQ control and competency in foreign subsidiaries. This means that higher levels of HQ control negatively impact the competency levels of subsidiaries because strict control undermines autonomy in subsidiaries. However, if the level of HQ control exceeds a certain point, then the transfer of knowledge between HQs and subsidiaries is facilitated. Knowledge transferred from HQs can be used as prior knowledge by foreign subsidiaries to the benefit of all parties. Accordingly, knowledge transfer negates the negative effects of excessive HQ control and positively affects competency in subsidiaries. Second, there is an inverted U-shaped relationship between the local (host-country) experiences of subsidiaries and competency in foreign subsidiaries. This means that foreign subsidiaries can overcome the liabilities of foreignness and contribute to capability building by accumulating unique knowledge about their host countries. However, if local experiences accumulate excessively beyond a certain point, then the host country-specific experiences of foreign subsidiaries will offset the benefits discussed above. Excessive local experiences not only increase organizational inertia, but also create a problem of goal incongruence due to information asymmetry between HQs and subsidiaries. Therefore, excessive local experiences have negative effects on competency in foreign subsidiaries. Originality/value - This study suggests the following implications. First, unlike existing studies based mainly on linear models, this study presents important theoretical implications in its focus on nonlinear models and its analysis of the effects of HQ control and local experiences on competency in foreign subsidiaries from perspectives of organizational learning theory and agency theory. Second, in terms of practical implications, the results of this study suggest that optimally raising levels of HQ control and managing the local experiences of subsidiaries without increasing organizational inertia is important for enhancing competency in foreign subsidiaries.