• Journal of Digital Convergence
• /
• v.12 no.8
• /
• pp.321-327
• /
• 2014

Feature selection is the technique to improve the classification performance by using a minimal set by removing features that are not related with each other and characterized by redundancy. This study proposed new feature selection using the distance between the center of gravity of the bounded sum of weighted fuzzy membership functions (BSWFMs) provided by the neural network with weighted fuzzy membership functions (NEWFM) in order to improve the classification performance. The distance-based feature selection selects the minimum features by removing the worst features with the shortest distance between the center of gravity of BSWFMs from the 24 initial features one by one, and then 22 minimum features are selected with the highest performance result. The proposed methodology shows that sensitivity, specificity, and accuracy are 97.7%, 99.7%, and 98.7% with 22 minimum features, respectively.

• Journal of Internet Computing and Services
• /
• v.8 no.1
• /
• pp.125-132
• /
• 2007

This paper presents an approach to detect premature ventricular contractions(PVC) using the neural network with weighted fuzzy membership functions(NEWFM), NEWFM classifies normal and PVC beats by the trained weighted fuzzy membership functions using wavelet transformed coefficients extracted from the MIT-BIH PVC database. The eight most important coefficients of d3 and d4 are selected by the non-overlap area distribution measurement method. The selected 8 coefficients are used for 3 data sets showing reliable accuracy rates 99,80%, 99,21%, and 98.78%, respectively, which means the selected input features are less dependent to the data sets. The ECG signal segments and fuzzy membership functions of the 8 coefficients enable input features to interpret explicitly.

• ETRI Journal
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2019

In this paper, we propose a rank-weighted reconstruction feature to improve the robustness of a feed-forward deep neural network (FFDNN)-based acoustic model. In the FFDNN-based acoustic model, an input feature is constructed by vectorizing a submatrix that is created by slicing the feature vectors of frames within a context window. In this type of feature construction, the appropriate context window size is important because it determines the amount of trivial or discriminative information, such as redundancy, or temporal context of the input features. However, we ascertained whether a single parameter is sufficiently able to control the quantity of information. Therefore, we investigated the input feature construction from the perspectives of rank and nullity, and proposed a rank-weighted reconstruction feature herein, that allows for the retention of speech information components and the reduction in trivial components. The proposed method was evaluated in the TIMIT phone recognition and Wall Street Journal (WSJ) domains. The proposed method reduced the phone error rate of the TIMIT domain from 18.4% to 18.0%, and the word error rate of the WSJ domain from 4.70% to 4.43%.

• Genomics & Informatics
• /
• v.21 no.4
• /
• pp.45.1-45.11
• /
• 2023

Nonalcoholic fatty liver disease (NAFLD) is a common type of chronic liver disease, with severity levels ranging from nonalcoholic fatty liver to nonalcoholic steatohepatitis (NASH). The extent of liver fibrosis indicates the severity of NASH and the risk of liver cancer. However, the mechanism underlying NASH development, which is important for early screening and intervention, remains unclear. Weighted gene co-expression network analysis (WGCNA) is a useful method for identifying hub genes and screening specific targets for diseases. In this study, we utilized an mRNA dataset of the liver tissues of patients with NASH and conducted WGCNA for various stages of liver fibrosis. Subsequently, we employed two additional mRNA datasets for validation purposes. Gene set enrichment analysis (GSEA) was conducted to analyze gene function enrichment. Through WGCNA and subsequent analyses, complemented by validation using two additional datasets, we identified five genes (BICC1, C7, EFEMP1, LUM, and STMN2) as hub genes. GSEA analysis indicated that gene sets associated with liver metabolism and cholesterol homeostasis were uniformly downregulated. BICC1, C7, EFEMP1, LUM, and STMN2 were identified as hub genes of NASH, and were all related to liver metabolism, NAFLD, NASH, and related diseases. These hub genes might serve as potential targets for the early screening and treatment of NASH.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.123-128
• /
• 2019

This paper proposes the stock price prediction based on the artificial intelligence, where the model with recurrent convolution neural network (RCNN) layers is adopted. In the motivation of this prediction, long short-term memory model (LSTM)-based neural network can make the output of the time series prediction. On the other hand, the convolution neural network provides the data filtering, averaging, and augmentation. By combining the advantages mentioned above, the proposed technique predicts the estimated stock price of next day. In addition, in order to emphasize the recent time series, a custom weighted loss function is adopted. Moreover, stock data related to the stock price index are adopted to consider the market trends. In the experiments, the proposed stock price prediction reduces the test error by 3.19%, which is over other techniques by about 19%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.5
• /
• pp.1984-2002
• /
• 2020

With the appearance of wireless spectrum crisis in traditional cellular network, device-to-device (D2D) communication has been regarded as a promising solution to ease heavy traffic burden by enabling precise content delivery among mobile users. However, due to the channel sharing, the interference between D2D and cellular users can affect the transmission rate and narrow the throughput in the network. In this paper, we firstly present a weighted interference minimization cluster formation model involving both social attribute and physical closeness. The weighted-interference, which is evaluated under the susceptible-infected(SI) model, is utilized to gather user in social and physical proximity. Then, we address the cluster formation problem via spectrum clustering with iterative operation. Finally, we propose the stable matching theory algorithm in order to maximize rate oriented to accomplish the one-to-one resource allocation. Numerical results show that our proposed scheme acquires quite well clustering effect and increases the accumulative transmission rate compared with the other two advanced schemes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.5
• /
• pp.73-78
• /
• 2019

This paper proposes polynomial-time rule for maximum weighted independent set(MWIS) problem that is well known NP-hard. The well known distributed algorithm selects the maximum weighted node as a element of independent set in a local. But the merged independent nodes with less weighted nodes have more weights than maximum weighted node are frequently occur. In this case, existing algorithm fails to get the optimal solution. To deal with these problems, this paper constructs maximum weighted independent set in local area. Application result of proposed algorithm to various networks, this algorithm can be get the optimal solution that fail to existing algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.245-265
• /
• 2022

In response to problems such as insufficient extraction information, low detection accuracy, and frequent misdetection in the field of Thangka image defects, this paper proposes a YOLOv5 prediction algorithm fused with the attention mechanism. Firstly, the Backbone network is used for feature extraction, and the attention mechanism is fused to represent different features, so that the network can fully extract the texture and semantic features of the defect area. The extracted features are then weighted and fused, so as to reduce the loss of information. Next, the weighted fused features are transferred to the Neck network, the semantic features and texture features of different layers are fused by FPN, and the defect target is located more accurately by PAN. In the detection network, the CIOU loss function is used to replace the GIOU loss function to locate the image defect area quickly and accurately, generate the bounding box, and predict the defect category. The results show that compared with the original network, YOLOv5-SE and YOLOv5-CBAM achieve an improvement of 8.95% and 12.87% in detection accuracy respectively. The improved networks can identify the location and category of defects more accurately, and greatly improve the accuracy of defect detection of Thangka images.

• Journal of Information Technology Applications and Management
• /
• v.20 no.3_spc
• /
• pp.219-230
• /
• 2013

Real time accessiblity and agility in Ubiquitous-commerce is required under ubiquitous computing environment. The Research has been actively processed in e-commerce so as to improve the accuracy of recommendation. Existing Collaborative filtering (CF) can not reflect contents of the items and has the problem of the process of selection in the neighborhood user group and the problems of sparsity and scalability as well. Although a system has been practically used to improve these defects, it still does not reflect attributes of the item. In this paper, to solve this problem, We can use a implicit method which is used by customer's data and purchase history data. We propose a new clustering method of weighted preference for customer using k-means clustering and Bayesian network in order to improve the accuracy of recommendation. To verify improved performance of the proposed system, we make experiments with dataset collected in a cosmetic internet shopping mall.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.2
• /
• pp.1-11
• /
• 2009

This study presents HCWS algorithm for researcher grouping on a weighted researcher network. The weights represent intensity of connections among researchers based on the number of co-authors and the number of co-authored research papers. To confirm the validity of the proposed technique, this study conducted an experimentation on about 80 research papers. As a consequence, it is proved that HCWS algorithm is able to bring about more realistic clustering compared with HCS algorithm which presents semantic relations among researchers in simple connections. In addition, it is found that HCWS algorithm can address the problems of existing HCS algorithm; researchers are disconnected since their connections are classified as weak even though they are strong, and vise versa. The technique described in this research paper can be applied to efficiently establish social networks of researchers considering relations such as collaboration histories among researchers or to create communities of researchers.