• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.67-80
• /
• 2021

The node embedding technique for learning graph representation plays an important role in obtaining good quality results in graph mining. Until now, representative node embedding techniques have been studied for homogeneous graphs, and thus it is difficult to learn knowledge graphs with unique meanings for each edge. To resolve this problem, the conventional Triple2Vec technique builds an embedding model by learning a triple graph having a node pair and an edge of the knowledge graph as one node. However, the Triple2 Vec embedding model has limitations in improving performance because it calculates the relationship between triple nodes as a simple measure. Therefore, this paper proposes a feature extraction technique based on a graph convolutional neural network to improve the Triple2Vec embedding model. The proposed method extracts the neighborliness vector of the triple graph and learns the relationship between neighboring nodes for each node in the triple graph. We proves that the embedding model applying the proposed method is superior to the existing Triple2Vec model through category classification experiments using DBLP, DBpedia, and IMDB datasets.

• International Journal of Computer Science & Network Security
• /
• v.24 no.10
• /
• pp.1-16
• /
• 2024

Due to its complexity and high diagnosis and treatment costs, heart attack (HA) is the top cause of death globally. Heart failure's widespread effect and high morbidity and death rates make accurate and fast prognosis and diagnosis crucial. Due to the complexity of medical data, early and accurate prediction of HA is difficult. Healthcare providers must evaluate data quickly and accurately to intervene. This novel hybrid approach predicts HA using Long Short-Term Memory (LSTM) networks, Deep belief networks (DBNs) with attention mechanism, and robust data mining to fill this essential gap. HA is predicted using Kaggle, PhysioNet, and UCI datasets. Wearable sensor data, ECG signals, and demographic and clinical data provide a solid analytical base. To maintain consistency, ECG signals are normalized and segmented after thorough cleaning to remove missing values and noise. Feature extraction employs complex approaches like Principal Component Analysis (PCA) and Autoencoders to pick time-domain (MNN, SDNN, RMSSD, PNN50) and frequency-domain (PSD at VLF, LF, HF bands) characteristics. The hybrid model architecture uses LSTM networks for sequence learning and DBNs for feature representation and selection to create a robust and comprehensive prediction model. Accuracy, precision, recall, F1-score, and ROC-AUC are measured after cross-entropy loss and SGD optimization. The LSTM-DBN model outperforms predictive methods in accuracy, sensitivity, and specificity. The findings show that several data sources and powerful algorithms can improve heart attack predictions. The proposed architecture performed well on many datasets, with an accuracy rate of 96.00%, sensitivity of 98%, AUC of 0.98, and F1-score of 0.97. High performance proves this system's dependability. Moreover, the proposed approach is outperformed compared to state-of-the-art systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3762-3781
• /
• 2020

Infrared pedestrian target recognition is a vital problem of significant interest in computer vision. In this work, a novel infrared pedestrian target recognition method that uses heterogeneous features with multi-kernel transfer learning is proposed. Firstly, to exploit the characteristics of infrared pedestrian targets fully, a novel multi-scale monogenic filtering-based completed local binary pattern descriptor, referred to as MSMF-CLBP, is designed to extract the texture information, and then an improved histogram of oriented gradient-fisher vector descriptor, referred to as HOG-FV, is proposed to extract the shape information. Second, to enrich the semantic content of feature expression, these two heterogeneous features are integrated to get more complete representation for infrared pedestrian targets. Third, to overcome the defects, such as poor generalization, scarcity of tagged infrared samples, distributional and semantic deviations between the training and testing samples, of the state-of-the-art classifiers, an effective multi-kernel transfer learning classifier called MK-TrAdaBoost is designed. Experimental results show that the proposed method outperforms many state-of-the-art recognition approaches for infrared pedestrian targets.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.5
• /
• pp.277-281
• /
• 2013

Modern speaker verification systems based on support vector machines (SVMs) use Gaussian mixture model (GMM) supervectors as their input feature vectors, and the maximum a posteriori (MAP) adaptation is a conventional method for generating speaker-dependent GMMs by adapting a universal background model (UBM). MAP adaptation requires the appropriate amount of input utterance due to the number of model parameters to be estimated. On the other hand, with limited utterances, unreliable MAP adaptation can be performed, which causes adaptation noise even though the Bayesian priors used in the MAP adaptation smooth the movements between the UBM and speaker dependent GMMs. This paper proposes a sparse MAP adaptation method, which is known to perform well in the automatic speech recognition area. By introducing sparse MAP adaptation to the GMM-SVM-based speaker verification system, the adaptation noise can be mitigated effectively. The proposed method utilizes the L0 norm as a regularizer to induce sparsity. The experimental results on the TIMIT database showed that the sparse MAP-based GMM-SVM speaker verification system yields a 42.6% relative reduction in the equal error rate with few additional computations.

• Korean Institute of Interior Design Journal
• /
• v.27 no.2
• /
• pp.175-185
• /
• 2018

The purpose of this study is to analyze the concept of 'Gradual Differentiation' in parametric design in terms of pure model logic and thus describe the distinctive feature from the previous design method. To meet the purpose, it explores external cases like gradual factor identified in natural phenomenon and artworks and define the inherent model principles into "Self-similarity', "Correlation', and 'Temporality' by examining these features in terms of algorithm. Meanwhile, it identified the principle of gradual model representation in parametric design within a single system called 'Attractor System' by applying these three concepts into specific methods of parametric design, and by interpreting the logical structure through the association among 'Attractor', 'Field', and 'Differentiation'. The creative utilization of parameter shows that gradual model process in parametric design does not mean a passive "conversion process" merely replacing natural parameter with algorithm; rather, it refers to an active "generating process" creating new meanings and value. By continuing this process of conceptual understanding and insight, creative perspective and practical ability to interpret parameter can be improved.

• International Journal of Human Ecology
• /
• v.14 no.1
• /
• pp.87-102
• /
• 2013

This study investigated residents' preferences for spatial features of sitting areas in assisted living facilities, and provides recommendations for planning sitting areas to support residents' spatial preferences and social interaction. The study participants were 69 residents of eight assisted living facilities (30+ resident capacity), located in south central Texas. A photographic comparison method was used, in which residents were shown 20 matched pairs of photos, with a single feature digitally modified in each pair, and asked to select which environmental representation they preferred. The hypothesized spatial characteristics were identified in practice based literature as those that may encourage usage of sitting areas: viewability, variety, homelikeness, and privacy. Most of the hypothesized features were preferred by participants, with the highest preference found for non-institutional furniture arrangements and naturalness, followed by increasing enclosure and variety of seating. Preference was less significant for domestic cues such as carpeted floors, divided light windows, and boundaries defined by different colored material or columns, possibly due to their physical impairments or preference for visual openness. Participants' level of mobility assistance was significantly related to their preference for some features, such as seating with people-watching capability, and carpeted floors. The findings have implications for facility architects and administrators engaged in resident-oriented spatial planning.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.2
• /
• pp.77-85
• /
• 2009

In this paper, we propose data format for smell, taste, touch with speech and vision which can be transmitted and implement a floral scent detection and recognition system. We provide representation method of data of smell, taste, and touch. Also, proposed floral scent recognition system consists of three module such as floral scent acquisition module using Metal Oxide Semiconductor (MOS) sensor array, entropy-based floral scent detection module, and floral scent recognition module using correlation coefficients. The proposed system calculates correlation coefficients of the individual sensor between feature vector(16 sensors) from floral scent input point until the stable region and 12 types of reference models. Then, this system selects the floral scent with the maximum similarity to the calculated average of individual correlation coefficients. To evaluate the floral scent recognition system using correlation coefficients, we implemented an individual floral scent recognition system using K-NN with PCA and LDA that are generally used in conventional electronic noses. In the experimental results, the proposed system performs approximately 95.7% average recognition rate.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.7
• /
• pp.1517-1524
• /
• 2011

In this paper, a new document summarization method, which uses the semantic features and the pseudo relevance feedback (PRF) by using WordNet, is introduced to extract meaningful sentences relevant to a user query. The proposed method can improve the quality of document summaries because the inherent semantic of the documents are well reflected by the semantic feature from NMF. In addition, it uses the PRF by the semantic features and WordNet to reduce the semantic gap between the high level user's requirement and the low level vector representation. The experimental results demonstrate that the proposed method achieves better performance that the other methods.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1877-1885
• /
• 2015

This paper presents a new Nearest-Neighbors based weighted representation for images and weighted K-Nearest-Neighbors (WKNN) classifier to improve the precision of image classification using the Bag of Visual Words (BOVW) based models. Scale-invariant feature transform (SIFT) features are firstly extracted from images. Then, the K-means++ algorithm is adopted in place of the conventional K-means algorithm to generate a more effective visual dictionary. Furthermore, the histogram of visual words becomes more expressive by utilizing the proposed weighted vector quantization (WVQ). Finally, WKNN classifier is applied to enhance the properties of the classification task between images in which similar levels of background noise are present. Average precision and absolute change degree are calculated to assess the classification performance and the stability of K-means++ algorithm, respectively. Experimental results on three diverse datasets: Caltech-101, Caltech-256 and PASCAL VOC 2011 show that the proposed WVQ method and WKNN method further improve the performance of classification.

• Journal of the Korean Society for information Management
• /
• v.27 no.3
• /
• pp.227-239
• /
• 2010

Opinion retrieval is to retrieve items which are relevant to the user information need topically and include opinion about the topic. This paper aims to find a method to represent user information need for effective opinion retrieval and to analyze the combination methods for opinion features through various experiments. The experiments are carried out in the inference network framework using the Blogs06 collection and 100 TREC test topics. The results show that our suggested representation method based on hidden 'opinion' concept is effective, and the compact model with very small opinion lexicon shows the comparable performance to the previous model on the same test data set.